"rsync for cloud storage" - Google Drive, S3, Dropbox, Backblaze B2, One Drive, Swift, Hubic, Wasabi, Google Cloud Storage, Azure Blob, Azure Files, Yandex Files
Go azure-blobazure-blob-storageazure-filesbackblaze-b2cloud-storagedropboxencryptionftpfuse-filesystemgogolanggoogle-cloud-storagegoogle-driveonedriveopenstack-swiftrclones3sftpsyncwebdav
Version v1.76.0
Nick Craig-Woode982ff36668f
diff --git a/MANUAL.html b/MANUAL.htmlindex 4a397f5b7..32e331e47 100644--- a/MANUAL.html+++ b/MANUAL.html@@ -233,7 +233,7 @@<header id="title-block-header"><h1 class="title">rclone(1) User Manual</h1><p class="author">Nick Craig-Wood</p>-<p class="date">Jul 31, 2026</p>+<p class="date">Nov 06, 2026</p></header><h1 id="name">NAME</h1><p>rclone - manage files on cloud storage</p>@@ -268,6 +268,7 @@ Available commands:gui Open the web based GUI.hashsum Produces a hashsum file for all the objects in the path.help Show help for rclone commands, flags and backends.+ index Write static directory listings into a remote.link Generate public link to file/folder.listremotes List all the remotes in the config file and defined in environment variables.ls List the objects in the path with size and path.@@ -422,6 +423,7 @@ S3, that work out of the box.)</p><li>Akamai Netstorage</li><li>Alibaba Cloud (Aliyun) Object Storage System (OSS)</li><li>Amazon S3</li>+<li>Autonomys Auto Drive</li><li>Bizfly Cloud Simple Storage</li><li>Backblaze B2</li><li>Box</li>@@ -429,11 +431,13 @@ S3, that work out of the box.)</p><li>China Mobile Ecloud Elastic Object Storage (EOS)</li><li>Arvan Cloud Object Storage (AOS)</li><li>Citrix ShareFile</li>+<li>Clever Cloud Cellar</li><li>Cloudflare R2</li><li>Cloudinary</li><li>Cubbit DS3</li><li>DigitalOcean Spaces</li><li>Digi Storage</li>+<li>dosya.dev</li><li>Dreamhost</li><li>Drime</li><li>Dropbox</li>@@ -467,8 +471,11 @@ S3, that work out of the box.)</p><li>IBM COS S3</li><li>IDrive e2</li><li>Impossible Cloud</li>+<li>Infomaniak Swiss Backup (S3)</li>+<li>Infomaniak Swiss Backup (Swift)</li><li>Intercolo Object Storage</li><li>IONOS Cloud</li>+<li>kDrive</li><li>Koofr</li><li>Leviia Object Storage</li><li>Liara Object Storage</li>@@ -484,6 +491,7 @@ S3, that work out of the box.)</p><li>Microsoft Azure Files Storage</li><li>Microsoft OneDrive</li><li>Minio</li>+<li>Neon Object Storage</li><li>Nextcloud</li><li>Blomp Cloud Storage</li><li>OpenDrive</li>@@ -508,6 +516,7 @@ S3, that work out of the box.)</p><li>RackCorp Object Storage</li><li>Rackspace Cloud Files</li><li>Rclone Serve S3</li>+<li>RelAix.Storage</li><li>rsync.net</li><li>Scaleway</li><li>Scality (RING / ARTESCA)</li>@@ -752,12 +761,6 @@ alt="Packaging status" /></a></p><p>The rclone developers maintain a <ahref="https://hub.docker.com/r/rclone/rclone">docker image forrclone</a>.</p>-<p><strong>Note:</strong> We also now offer a paid version of rclone-with enterprise-grade security and zero CVEs through our partner <a-href="https://securebuild.com/blog/introducing-securebuild">SecureBuild</a>.-If you are interested, check out their website and the <a-href="https://securebuild.com/images/rclone">Rclone SecureBuild-Image</a>.</p><p>These images are built as part of the release process based on aminimal Alpine Linux.</p><p>The <code>:latest</code> tag will always point to the latest stable@@ -868,7 +871,7 @@ current version is as below.</p>src="https://snapcraft.io/rclone/badge.svg" alt="rclone" /></a></p><h2 id="source">Source installation</h2><p>Make sure you have git and <a href="https://golang.org/">Go</a>-installed. Go version 1.25 or newer is required, the latest release is+installed. Go version 1.26 or newer is required, the latest release isrecommended. You can get it from your package manager, or download itfrom <a href="https://golang.org/dl/">golang.org/dl</a>. Then you canrun the following:</p>@@ -1182,6 +1185,7 @@ remotes</li>Spaces</a></li><li><a href="https://rclone.org/koofr/#digi-storage">DigiStorage</a></li>+<li><a href="https://rclone.org/dosya/">dosya.dev</a></li><li><a href="https://rclone.org/drime/">Drime</a></li><li><a href="https://rclone.org/dropbox/">Dropbox</a></li><li><a href="https://rclone.org/filefabric/">Enterprise File@@ -1208,6 +1212,7 @@ Storage Box</a></li>Archive</a></li><li><a href="https://rclone.org/internxt/">Internxt</a></li><li><a href="https://rclone.org/jottacloud/">Jottacloud</a></li>+<li><a href="https://rclone.org/kdrive/">kDrive</a></li><li><a href="https://rclone.org/koofr/">Koofr</a></li><li><a href="https://rclone.org/linkbox/">Linkbox</a></li><li><a href="https://rclone.org/mailru/">Mail.ru Cloud</a></li>@@ -2259,6 +2264,10 @@ use <code>--max-depth 1</code> to stop the recursion.</p><p>The other list commands<code>lsd</code>,<code>lsf</code>,<code>lsjson</code> do not recurse bydefault - use <code>-R</code> to make them recurse.</p>+<p>On the bucket-based remotes (e.g. s3, gcs, b2) a directory is a key+prefix rather than an object, so a listing without <code>-R</code> shows+each prefix as one directory entry and not the objects stored under+it.</p><p>List commands prefer a recursive method that uses more memory butfewer transactions by default. Use <code>--disable ListR</code> tosuppress the behavior. See <a@@ -2317,9 +2326,9 @@ for rclone commands, flags and backends.</li><p>Lists the directories in the source path to standard output. Does notrecurse by default. Use the <code>-R</code> flag to recurse.</p><p>This command lists the total size of the directory (if known, -1 if-not), the modification time (if known, the current time if not), the-number of objects in the directory (if known, -1 if not) and the name of-the directory, E.g.</p>+not), the modification time (if known, the <code>--default-time</code>+if not), the number of objects in the directory (if known, -1 if not)+and the name of the directory, E.g.</p><pre class="console"><code>$ rclone lsd swift:494000 2018-04-26 08:43:20 10000 10000files65 2018-04-26 08:43:20 1 1File</code></pre>@@ -2351,6 +2360,10 @@ use <code>--max-depth 1</code> to stop the recursion.</p><p>The other list commands<code>lsd</code>,<code>lsf</code>,<code>lsjson</code> do not recurse bydefault - use <code>-R</code> to make them recurse.</p>+<p>On the bucket-based remotes (e.g. s3, gcs, b2) a directory is a key+prefix rather than an object, so a listing without <code>-R</code> shows+each prefix as one directory entry and not the objects stored under+it.</p><p>List commands prefer a recursive method that uses more memory butfewer transactions by default. Use <code>--disable ListR</code> tosuppress the behavior. See <a@@ -2437,6 +2450,10 @@ use <code>--max-depth 1</code> to stop the recursion.</p><p>The other list commands<code>lsd</code>,<code>lsf</code>,<code>lsjson</code> do not recurse bydefault - use <code>-R</code> to make them recurse.</p>+<p>On the bucket-based remotes (e.g. s3, gcs, b2) a directory is a key+prefix rather than an object, so a listing without <code>-R</code> shows+each prefix as one directory entry and not the objects stored under+it.</p><p>List commands prefer a recursive method that uses more memory butfewer transactions by default. Use <code>--disable ListR</code> tosuppress the behavior. See <a@@ -3292,6 +3309,7 @@ for details.</p>--force Bypass --max-delete safety check and run the sync. Consider using with --verbose-h, --help help for bisync--ignore-listing-checksum Do not use checksums for listings (add --ignore-checksum to additionally skip post-copy checksum checks)+ --max-delete-renames-aware Exclude tracked renames from the --max-delete safety check (requires --track-renames)--max-lock Duration Consider lock files older than this to be expired (default: 0 (never expire)) (minimum: 2m) (default 0s)--no-cleanup Retain working files (useful for troubleshooting and testing).--no-slow-hash Ignore listing checksums only on backends where they are slow@@ -3576,7 +3594,9 @@ for rclone commands, flags and backends.</li><p>Output completion script for a given shell.</p><h2 id="synopsis-28">Synopsis</h2><p>Generates a shell completion script for rclone. Run with-<code>--help</code> to list the supported shells.</p>+<code>--help</code> to list the supported shells. Enum-valued flags such+as <code>--log-level</code> and <code>--vfs-cache-mode</code> also+complete their supported values.</p><h2 id="options-28">Options</h2><pre><code> -h, --help help for completion</code></pre><p>See the <a href="https://rclone.org/flags/">global flags page</a> for@@ -4623,9 +4643,9 @@ SquareBracket</code></pre><pre class="console"><code>rclone convmv "stories/The Quick Brown Fox!.txt" --name-transform "all,command=echo"// Output: stories/The Quick Brown Fox!.txt</code></pre><pre class="console"><code>rclone convmv "stories/The Quick Brown Fox!" --name-transform "date=-{YYYYMMDD}"-// Output: stories/The Quick Brown Fox!-20260731</code></pre>+// Output: stories/The Quick Brown Fox!-20261106</code></pre><pre class="console"><code>rclone convmv "stories/The Quick Brown Fox!" --name-transform "date=-{macfriendlytime}"-// Output: stories/The Quick Brown Fox!-2026-07-31 0340PM</code></pre>+// Output: stories/The Quick Brown Fox!-2026-11-06 0220PM</code></pre><pre class="console"><code>rclone convmv "stories/The Quick Brown Fox!.txt" --name-transform "all,regex=[\\.\\w]/ab"// Output: ababababababab/ababab ababababab ababababab ababab!abababab</code></pre><p>The regex command generally accepts Perl-style regular expressions,@@ -5211,13 +5231,31 @@ for rclone commands, flags and backends.</li><h2 id="synopsis-52">Synopsis</h2><p>Remove a single file from remote. Unlike <code>delete</code> itcannot be used to remove a directory and it doesn't obey include/exclude-filters - if the specified file exists, it will always be removed.</p>+filters - if the specified file exists, it will always be removed from+its original location.</p>+<p>The <code>--backup-dir</code> and <code>--suffix</code> flags are+honored when set.</p><pre><code>rclone deletefile remote:path [flags]</code></pre><h2 id="options-59">Options</h2><pre><code> -h, --help help for deletefile</code></pre><p>Options shared with other commands are described next. See the <ahref="https://rclone.org/flags/">global flags page</a> for globaloptions not listed here.</p>+<h3 id="sync-options-2">Sync Options</h3>+<p>Flags used for sync commands</p>+<pre class="text"><code> --backup-dir string Make backups into hierarchy based in DIR+ --delete-after When synchronizing, delete files on destination after transferring (default)+ --delete-before When synchronizing, delete files on destination before transferring+ --delete-during When synchronizing, delete files during transfer+ --fix-case Force rename of case insensitive dest to match source+ --ignore-errors Delete even if there are I/O errors+ --list-cutoff int To save memory, sort directory listings on disk above this threshold (default 1000000)+ --max-delete int When synchronizing, limit the number of deletes (default -1)+ --max-delete-size SizeSuffix When synchronizing, limit the total size of deletes (default off)+ --suffix string Suffix to add to changed files+ --suffix-keep-extension Preserve the extension when using --suffix+ --track-renames When synchronizing, track file renames and do a server-side move if possible+ --track-renames-strategy string Strategies to use when synchronizing using track-renames hash|modtime|leaf (default "hash")</code></pre><h3 id="important-options-14">Important Options</h3><p>Important flags useful for most commands</p><pre class="text"><code> -n, --dry-run Do a trial run with no permanent changes@@ -5385,12 +5423,7 @@ options not listed here.</p>--rc-server-write-timeout Duration Timeout for server writing data (default 1h0m0s)--rc-template string User-specified template--rc-user string User name for authentication- --rc-user-from-header string User name from a defined HTTP header- --rc-web-fetch-url string URL to fetch the releases for webgui (default "https://api.github.com/repos/rclone/rclone-webui-react/releases/latest")- --rc-web-gui Launch WebGUI on localhost- --rc-web-gui-force-update Force update to latest version of web gui- --rc-web-gui-no-open-browser Don't open the browser automatically- --rc-web-gui-update Check and update to latest version of web gui</code></pre>+ --rc-user-from-header string User name from a defined HTTP header</code></pre><h2 id="see-also-62">See Also</h2><!-- markdownlint-capture --><!-- markdownlint-disable ul-style line-length -->@@ -5427,7 +5460,8 @@ Supported hashes are:- sha512- blake3- xxh3-- xxh128</code></pre>+- xxh128+- crc32c</code></pre><p>Then</p><pre class="console"><code>rclone hashsum MD5 remote:path</code></pre><p>Note that hash names are case insensitive and values are output in@@ -5480,9 +5514,309 @@ options not listed here.</p>for rclone commands, flags and backends.</li></ul><!-- markdownlint-restore -->+<h1 id="rclone-index">rclone index</h1>+<p>Write static directory listings into a remote.</p>+<h2 id="synopsis-57">Synopsis</h2>+<p>Write a directory listing, <code>index.html</code> by default, into+every directory of <code>remote:path</code> so that it can be browsed+when it is served as a static website, for example from an S3 website+endpoint, a bucket behind a CDN, or a plain web server.</p>+<p>This works like <code>rclone sync</code>: each run makes the minimum+changes needed to bring the listings into line with the files. Listings+which haven't changed aren't rewritten, so a run with no changes makes+no uploads. The listings never show the listing files themselves.</p>+<p><strong>Note</strong> that <code>rclone index</code> treats every+file with an output name (<code>index.html</code> by default) in the+directories it indexes as its own, and will overwrite it or delete it+without checking where it came from. To protect a hand-written index,+exclude its directory with <code>--index-exclude</code>.</p>+<h2 id="outputs">Outputs</h2>+<p><code>--output NAME=FORMAT</code> says what to write in each+directory and can be repeated. NAME is the file name, and FORMAT is one+of the built-in formats below or the path of a Go template file.</p>+<table>+<colgroup>+<col style="width: 27%" />+<col style="width: 37%" />+<col style="width: 34%" />+</colgroup>+<thead>+<tr class="header">+<th style="text-align: left;">Format</th>+<th style="text-align: left;">MIME type</th>+<th style="text-align: left;">Contents</th>+</tr>+</thead>+<tbody>+<tr class="odd">+<td style="text-align: left;"><code>html</code></td>+<td style="text-align: left;"><code>text/html; charset=utf-8</code></td>+<td style="text-align: left;">The HTML listing from the+<code>rclone serve http</code> template</td>+</tr>+<tr class="even">+<td style="text-align: left;"><code>json</code></td>+<td style="text-align: left;"><code>application/json</code></td>+<td style="text-align: left;">JSON in the format of+<code>rclone lsjson</code> for the directory</td>+</tr>+<tr class="odd">+<td style="text-align: left;"><code>caddy</code></td>+<td style="text-align: left;"><code>application/json</code></td>+<td style="text-align: left;">JSON in the format of Caddy's+<code>file_server browse</code></td>+</tr>+</tbody>+</table>+<p>The default is <code>--output index.html=html</code>. The built-in+formats are templates, so they can be printed with+<code>--print-template FORMAT</code>, copied and changed. Outputs whose+name ends in <code>.html</code> or <code>.htm</code> are rendered with+Go's <code>html/template</code> and everything else with+<code>text/template</code>, which has a <code>json</code> function for+safe encoding. The template data is that of+<code>rclone serve http</code> with <code>.Static</code> set (see <a+href="https://rclone.org/commands/rclone_serve_http/#template">rclone+serve http</a>). Outputs are uploaded with the MIME type of the built-in+format, or that of the output name's extension for template outputs,+since static hosts serve the stored type.</p>+<h2 id="filters">Filters</h2>+<p>The normal filters (<code>--include</code>, <code>--exclude</code>,+<code>--filter-from</code>, <code>--max-age</code>, ...) control what+appears in the listings. An excluded file isn't listed, and an excluded+directory isn't listed or indexed.</p>+<p>The <code>--index-include</code>, <code>--index-exclude</code>,+<code>--index-filter</code> flags and their <code>-from</code> variants+control which directories get listings. They take the same patterns as+the normal filters, applied to directory paths, and+<code>--index-include</code> implies excluding all other directories. A+directory excluded here still appears in its parent's listing, but+nothing is written or deleted inside it, so it can have its own+hand-made index. <code>--index-max-depth</code> limits how deep listings+are written.</p>+<h2 id="modification-times">Modification times</h2>+<p>Listings show the modification time of each file. When using the s3,+oracleobjectstorage and swift backends rclone stores the modification+time as metadata which isn't returned by the bucket listing, so+<strong>every object needs a HEAD request</strong>, which can mean+thousands of transactions per run. Avoid this with one of:</p>+<ul>+<li><code>--use-server-modtime</code> uses the time the object was+uploaded instead, which is what the web server sends as+<code>Last-Modified</code>. Recommended for s3 unless the original+modification times matter.</li>+<li><code>--no-modtime</code> shows no times at all, which is the+cheapest option.</li>+</ul>+<p><code>--dir-time</code> sets the time shown for directories:</p>+<ul>+<li><code>newest</code> (default): the newest modification time of+anything below the directory. Note that this means one new file changes+the listing of every directory above it.</li>+<li><code>dir</code>: the directory's own modification time. On most+file systems writing a listing into a directory changes its time, so+this can take more than one run to settle.</li>+<li><code>none</code>: no time.</li>+</ul>+<p>The listing files are given the newest modification time below their+directory, so the <code>Last-Modified</code> header of a listing is+meaningful.</p>+<h2 id="links-1">Links</h2>+<p>Directory links are <code>dir/</code> by default, which works on+hosts that serve index documents: S3 website endpoints, GCS and Azure+static websites, nginx and Apache, and R2 custom domains with a URL+rewrite rule. For hosts that don't (plain S3 REST URLs, B2 friendly+URLs) use <code>--link-index</code> to make directory links+<code>dir/index.html</code> instead.</p>+<h2 id="partial-runs">Partial runs</h2>+<p>After a sync which changed a known set of files there is no need to+walk the whole remote. Tell <code>rclone index</code> what changed and+it re-indexes only the directories that could be affected:</p>+<pre><code>rclone index r2:bucket --changed v1.76.0/ --changed version.txt+rclone index r2:bucket --changed-from changes.txt+rclone sync ./public r2:bucket --exclude index.html --combined - | rclone index r2:bucket --changed-combined -</code></pre>+<ul>+<li><code>--changed PATH</code> names a changed file or directory+relative to <code>remote:path</code> and can be repeated. A directory,+given with a trailing slash or found to be one, is re-indexed+completely.</li>+<li><code>--changed-from FILE</code> reads one path per line exactly as+written.</li>+<li><code>--changed-combined FILE</code> reads the+<code>--combined</code> report of <code>rclone sync</code>: lines+starting with <code>+</code>, <code>-</code>, <code>*</code> and+<code>!</code> are changes and <code>=</code> lines are ignored.</li>+<li>Both files accept <code>-</code> for standard input and all three+flags can be combined.</li>+</ul>+<p>A changed file re-indexes its directory and every directory above it,+since their times and their listings can change too, and on bucket based+storage a directory can cease to exist. Each of those is one directory+listing. With <code>--dir-time newest</code> the directories above also+need the times of their unchanged subdirectories, which are read from+the modification times of those subdirectories' listings at the cost of+one small request each. With <code>--use-server-modtime</code> that is+the time the listing was written rather than the time of the newest+file, so partial runs can drift by the indexing delay until the next+full run. <code>--dir-time dir</code> and <code>none</code> need no such+lookups.</p>+<p>Directories which are neither named nor above a named path are left+alone, as are directories deeper than <code>--max-depth</code>. An empty+change list does nothing, and naming the root does a full run. There is+no automatic cutover to a full run since that depends on the size of the+whole remote: a partial run costs one listing per affected directory, a+full run one request per thousand objects on S3-like storage or one per+directory elsewhere. The number of directories a partial run will list+is logged with <code>-v</code>, and <code>--changed-max-dirs</code>+makes it fall back to a full run above that number. A good pattern is a+partial run after each upload and a scheduled full run to catch anything+missed.</p>+<h2 id="site-icon">Site icon</h2>+<p>The listings carry no icon or branding of their own. To give a site+an icon, upload a <code>favicon.ico</code> to the root of the remote,+which browsers fetch from there for every page. Exclude it from the+listings with <code>--exclude /favicon.ico</code> so it stays on the+site but isn't shown.</p>+<h2 id="deleting-listings">Deleting listings</h2>+<p>On backends which can't have empty directories (S3, B2 and other+bucket-based storage) a directory only exists while it has files in it.+Once the last file in a directory has been removed its listing is+deleted, since the directory only existed because of it. Backends which+can have empty directories keep a listing for each empty directory.</p>+<h2 id="using-with-sync">Using with sync</h2>+<p>A sync from a local build deletes files on the destination which+aren't in the source, which includes the listings. Either index the+local build before syncing:</p>+<pre><code>rclone index ./public+rclone sync ./public r2:bucket</code></pre>+<p>or exclude the listings from the sync so it neither uploads nor+deletes them, then index the remote:</p>+<pre><code>rclone sync ./public r2:bucket --exclude index.html+rclone index r2:bucket</code></pre>+<h2 id="examples">Examples</h2>+<p>An S3 website bucket, using the upload times so no object needs a+HEAD request:</p>+<pre><code>rclone index s3:my-site --use-server-modtime</code></pre>+<p>HTML and rclone JSON listings, setting <code>Cache-Control</code> on+them:</p>+<pre><code>rclone index s3:bucket --output index.html=html --output index.json=json --header-upload "Cache-Control: max-age=300"</code></pre>+<p>Only list release files, and don't write listings under+<code>/private</code>:</p>+<pre><code>rclone index r2:bucket --include "*.{zip,deb,rpm}" --index-exclude "/private/**"</code></pre>+<p>A B2 bucket served from its friendly URLs:</p>+<pre><code>rclone index b2:bucket --link-index</code></pre>+<p>See what would change:</p>+<pre><code>rclone index r2:bucket --dry-run -v</code></pre>+<p>Listings which haven't changed aren't normally rewritten, so changing+the headers or metadata set on them needs <code>--index-rewrite</code>,+which writes every listing a run would render whether it changed or+not:</p>+<pre><code>rclone index r2:bucket --index-rewrite --header-upload "Cache-Control: max-age=3600"</code></pre>+<pre><code>rclone index remote:path [flags]</code></pre>+<h2 id="options-64">Options</h2>+<pre><code> --changed stringArray Only re-index directories affected by this changed path (directories end in /)+ --changed-combined stringArray Read changed paths from a sync --combined report (use - to read from stdin)+ --changed-from stringArray Read changed paths from file, one per line (use - to read from stdin)+ --changed-max-dirs int Do a full index if a partial one would list more than this many directories (0 for no limit)+ --dir-time newest|dir|none How to work out the time shown for a directory (default newest)+ -h, --help help for index+ --index-exclude stringArray Don't write listings in directories matching pattern+ --index-exclude-from stringArray Read directory exclude patterns from file (use - to read from stdin)+ --index-filter stringArray Add a rule for which directories get listings+ --index-filter-from stringArray Read directory rules from a file (use - to read from stdin)+ --index-include stringArray Only write listings in directories matching pattern+ --index-include-from stringArray Read directory include patterns from file (use - to read from stdin)+ --index-max-depth int Only write listings this many directories deep (-1 for no limit) (default -1)+ --index-rewrite Write every listing even if it is unchanged+ --link-index Link to dir/NAME rather than dir/ for hosts without index documents+ --no-modtime Don't show modification times in listings+ --output stringArray Listing to write in each directory as NAME=FORMAT (default [index.html=html])+ --print-template string Print the built-in template for FORMAT and exit</code></pre>+<p>Options shared with other commands are described next. See the <a+href="https://rclone.org/flags/">global flags page</a> for global+options not listed here.</p>+<h3 id="copy-options-6">Copy Options</h3>+<p>Flags for anything which can copy a file</p>+<pre class="text"><code> --check-first Do all the checks before starting transfers+ -c, --checksum Check for changes with size & checksum (if available, or fallback to size only)+ --compare-dest stringArray Include additional server-side paths during comparison+ --copy-dest stringArray Implies --compare-dest but also copies files from paths into destination+ --cutoff-mode HARD|SOFT|CAUTIOUS Mode to stop transfers when reaching the max transfer limit HARD|SOFT|CAUTIOUS (default HARD)+ --ignore-case-sync Ignore case when synchronizing+ --ignore-checksum Skip post copy check of checksums+ --ignore-existing Skip all files that exist on destination+ --ignore-size Ignore size when skipping use modtime or checksum+ -I, --ignore-times Don't skip items that match size and time - transfer all unconditionally+ --immutable Do not modify files, fail if existing files have been modified+ --inplace Download directly to destination file instead of atomic download to temp/rename+ -l, --links Translate symlinks to/from regular files with a '.rclonelink' extension+ --max-backlog int Maximum number of objects in sync or check backlog (default 10000)+ --max-duration Duration Maximum duration rclone will transfer data for (default 0s)+ --max-transfer SizeSuffix Maximum size of data to transfer (default off)+ -M, --metadata If set, preserve metadata when copying objects+ --modify-window Duration Max time diff to be considered the same (default 1ns)+ --multi-thread-chunk-size SizeSuffix Chunk size for multi-thread downloads / uploads, if not set by filesystem (default 64Mi)+ --multi-thread-cutoff SizeSuffix Use multi-thread downloads for files above this size (default 256Mi)+ --multi-thread-streams int Number of streams to use for multi-thread downloads (default 4)+ --multi-thread-write-buffer-size SizeSuffix In memory buffer size for writing when in multi-thread mode (default 128Ki)+ --name-transform stringArray Transform paths during the copy process+ --no-check-dest Don't check the destination, copy regardless+ --no-traverse Don't traverse destination file system on copy+ --no-update-dir-modtime Don't update directory modification times+ --no-update-modtime Don't update destination modtime if files identical+ --order-by string Instructions on how to order the transfers, e.g. 'size,descending'+ --partial-suffix string Add partial-suffix to temporary file name when --inplace is not used (default ".partial")+ --refresh-times Refresh the modtime of remote files+ --server-side-across-configs Allow server-side operations (e.g. copy) to work across different configs+ --size-only Skip based on size only, not modtime or checksum+ --streaming-upload-cutoff SizeSuffix Cutoff for switching to chunked upload if file size is unknown, upload starts after reaching cutoff or when file ends (default 100Ki)+ -u, --update Skip files that are newer on the destination</code></pre>+<h3 id="important-options-15">Important Options</h3>+<p>Important flags useful for most commands</p>+<pre class="text"><code> -n, --dry-run Do a trial run with no permanent changes+ -i, --interactive Enable interactive mode+ -v, --verbose count Print lots more stuff (repeat for more)</code></pre>+<h3 id="filter-options-18">Filter Options</h3>+<p>Flags for filtering directory listings</p>+<pre class="text"><code> --delete-excluded Delete files on dest excluded from sync+ --exclude stringArray Exclude files matching pattern+ --exclude-from stringArray Read file exclude patterns from file (use - to read from stdin)+ --exclude-if-present stringArray Exclude directories if filename is present+ --files-from stringArray Read list of source-file names from file (use - to read from stdin)+ --files-from-raw stringArray Read list of source-file names from file without any processing of lines (use - to read from stdin)+ --files-from0 stringArray Read list of source-file names from file using NUL as separator (use - to read from stdin)+ -f, --filter stringArray Add a file filtering rule+ --filter-from stringArray Read file filtering patterns from a file (use - to read from stdin)+ --hash-filter string Partition filenames by hash k/n or randomly @/n+ --ignore-case Ignore case in filters (case insensitive)+ --include stringArray Include files matching pattern+ --include-from stringArray Read file include patterns from file (use - to read from stdin)+ --max-age Duration Only transfer files younger than this in s or suffix ms|s|m|h|d|w|M|y (default off)+ --max-depth int If set limits the recursion depth to this (default -1)+ --max-size SizeSuffix Only transfer files smaller than this in KiB or suffix B|K|M|G|T|P (default off)+ --metadata-exclude stringArray Exclude metadatas matching pattern+ --metadata-exclude-from stringArray Read metadata exclude patterns from file (use - to read from stdin)+ --metadata-filter stringArray Add a metadata filtering rule+ --metadata-filter-from stringArray Read metadata filtering patterns from a file (use - to read from stdin)+ --metadata-include stringArray Include metadatas matching pattern+ --metadata-include-from stringArray Read metadata include patterns from file (use - to read from stdin)+ --min-age Duration Only transfer files older than this in s or suffix ms|s|m|h|d|w|M|y (default off)+ --min-size SizeSuffix Only transfer files bigger than this in KiB or suffix B|K|M|G|T|P (default off)</code></pre>+<h3 id="listing-options-17">Listing Options</h3>+<p>Flags for listing directories</p>+<pre class="text"><code> --default-time Time Time to show if modtime is unknown for files and directories (default 2000-01-01T00:00:00Z)+ --fast-list Use recursive list if available; uses more memory but fewer transactions</code></pre>+<h2 id="see-also-64">See Also</h2>+<!-- markdownlint-capture -->+<!-- markdownlint-disable ul-style line-length -->+<ul>+<li><a href="https://rclone.org/commands/rclone/">rclone</a> - Show help+for rclone commands, flags and backends.</li>+</ul>+<!-- markdownlint-restore --><h1 id="rclone-link">rclone link</h1><p>Generate public link to file/folder.</p>-<h2 id="synopsis-57">Synopsis</h2>+<h2 id="synopsis-58">Synopsis</h2><p>Create, retrieve or remove a public link to the given file orfolder.</p><pre class="console"><code>rclone link remote:path/to/file@@ -5501,13 +5835,13 @@ Exact capabilities depend on the remote, but the link will always bydefault be created with the least constraints - e.g. no expiry, nopassword protection, accessible without account.</p><pre><code>rclone link remote:path [flags]</code></pre>-<h2 id="options-64">Options</h2>+<h2 id="options-65">Options</h2><pre><code> --expire Duration The amount of time that the link will be valid (default off)-h, --help help for link--unlink Remove existing public link to file/folder</code></pre><p>See the <a href="https://rclone.org/flags/">global flags page</a> forglobal options not listed here.</p>-<h2 id="see-also-64">See Also</h2>+<h2 id="see-also-65">See Also</h2><!-- markdownlint-capture --><!-- markdownlint-disable ul-style line-length --><ul>@@ -5518,7 +5852,7 @@ for rclone commands, flags and backends.</li><h1 id="rclone-listremotes">rclone listremotes</h1><p>List all the remotes in the config file and defined in environmentvariables.</p>-<h2 id="synopsis-58">Synopsis</h2>+<h2 id="synopsis-59">Synopsis</h2><p>Lists all the available remotes from the config file, or the remotesmatching an optional filter.</p><p>Prints the result in human-readable format by default, and as a@@ -5535,7 +5869,7 @@ syntax.</p><code>--type box</code> also matches <code>dropbox</code>. Use<code>--exact</code> to match complete values only.</p><pre><code>rclone listremotes [<filter>] [flags]</code></pre>-<h2 id="options-65">Options</h2>+<h2 id="options-66">Options</h2><pre><code> --description string Filter remotes by description--exact Match filter strings exactly instead of using non-anchored glob matching-h, --help help for listremotes@@ -5547,7 +5881,7 @@ syntax.</p>--type string Filter remotes by type</code></pre><p>See the <a href="https://rclone.org/flags/">global flags page</a> forglobal options not listed here.</p>-<h2 id="see-also-65">See Also</h2>+<h2 id="see-also-66">See Also</h2><!-- markdownlint-capture --><!-- markdownlint-disable ul-style line-length --><ul>@@ -5558,7 +5892,7 @@ for rclone commands, flags and backends.</li><h1 id="rclone-lsf">rclone lsf</h1><p>List directories and objects in remote:path formatted forparsing.</p>-<h2 id="synopsis-59">Synopsis</h2>+<h2 id="synopsis-60">Synopsis</h2><p>List the contents of the source path (directories and objects) tostandard output in a form which is easy to parse by scripts. By defaultthis will just be the names of the objects and directories, one per@@ -5674,6 +6008,10 @@ use <code>--max-depth 1</code> to stop the recursion.</p><p>The other list commands<code>lsd</code>,<code>lsf</code>,<code>lsjson</code> do not recurse bydefault - use <code>-R</code> to make them recurse.</p>+<p>On the bucket-based remotes (e.g. s3, gcs, b2) a directory is a key+prefix rather than an object, so a listing without <code>-R</code> shows+each prefix as one directory entry and not the objects stored under+it.</p><p>List commands prefer a recursive method that uses more memory butfewer transactions by default. Use <code>--disable ListR</code> tosuppress the behavior. See <a@@ -5683,7 +6021,7 @@ for more details.</p>remotes which can't have empty directories (e.g. s3, swift, or gcs - thebucket-based remotes).</p><pre><code>rclone lsf remote:path [flags]</code></pre>-<h2 id="options-66">Options</h2>+<h2 id="options-67">Options</h2><pre><code> --absolute Put a leading / in front of path names--csv Output in CSV format-d, --dir-slash Append a slash to directory names (default true)@@ -5698,7 +6036,7 @@ bucket-based remotes).</p><p>Options shared with other commands are described next. See the <ahref="https://rclone.org/flags/">global flags page</a> for globaloptions not listed here.</p>-<h3 id="filter-options-18">Filter Options</h3>+<h3 id="filter-options-19">Filter Options</h3><p>Flags for filtering directory listings</p><pre class="text"><code> --delete-excluded Delete files on dest excluded from sync--exclude stringArray Exclude files matching pattern@@ -5724,11 +6062,11 @@ options not listed here.</p>--metadata-include-from stringArray Read metadata include patterns from file (use - to read from stdin)--min-age Duration Only transfer files older than this in s or suffix ms|s|m|h|d|w|M|y (default off)--min-size SizeSuffix Only transfer files bigger than this in KiB or suffix B|K|M|G|T|P (default off)</code></pre>-<h3 id="listing-options-17">Listing Options</h3>+<h3 id="listing-options-18">Listing Options</h3><p>Flags for listing directories</p><pre class="text"><code> --default-time Time Time to show if modtime is unknown for files and directories (default 2000-01-01T00:00:00Z)--fast-list Use recursive list if available; uses more memory but fewer transactions</code></pre>-<h2 id="see-also-66">See Also</h2>+<h2 id="see-also-67">See Also</h2><!-- markdownlint-capture --><!-- markdownlint-disable ul-style line-length --><ul>@@ -5738,29 +6076,29 @@ for rclone commands, flags and backends.</li><!-- markdownlint-restore --><h1 id="rclone-lsjson">rclone lsjson</h1><p>List directories and objects in the path in JSON format.</p>-<h2 id="synopsis-60">Synopsis</h2>+<h2 id="synopsis-61">Synopsis</h2><p>List directories and objects in the path in JSON format.</p><p>The output is an array of Items, where each Item looks like this:</p>-<div class="sourceCode" id="cb351"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb351-1"><a href="#cb351-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>-<span id="cb351-2"><a href="#cb351-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Hashes"</span> <span class="fu">:</span> <span class="fu">{</span></span>-<span id="cb351-3"><a href="#cb351-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"SHA-1"</span> <span class="fu">:</span> <span class="st">"f572d396fae9206628714fb2ce00f72e94f2258f"</span><span class="fu">,</span></span>-<span id="cb351-4"><a href="#cb351-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"MD5"</span> <span class="fu">:</span> <span class="st">"b1946ac92492d2347c6235b4d2611184"</span><span class="fu">,</span></span>-<span id="cb351-5"><a href="#cb351-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"DropboxHash"</span> <span class="fu">:</span> <span class="st">"ecb65bb98f9d905b70458986c39fcbad7715e5f2fcc3b1f07767d7c83e2438cc"</span></span>-<span id="cb351-6"><a href="#cb351-6" aria-hidden="true" tabindex="-1"></a> <span class="fu">},</span></span>-<span id="cb351-7"><a href="#cb351-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"ID"</span><span class="fu">:</span> <span class="st">"y2djkhiujf83u33"</span><span class="fu">,</span></span>-<span id="cb351-8"><a href="#cb351-8" aria-hidden="true" tabindex="-1"></a> <span class="dt">"OrigID"</span><span class="fu">:</span> <span class="st">"UYOJVTUW00Q1RzTDA"</span><span class="fu">,</span></span>-<span id="cb351-9"><a href="#cb351-9" aria-hidden="true" tabindex="-1"></a> <span class="dt">"IsBucket"</span> <span class="fu">:</span> <span class="kw">false</span><span class="fu">,</span></span>-<span id="cb351-10"><a href="#cb351-10" aria-hidden="true" tabindex="-1"></a> <span class="dt">"IsDir"</span> <span class="fu">:</span> <span class="kw">false</span><span class="fu">,</span></span>-<span id="cb351-11"><a href="#cb351-11" aria-hidden="true" tabindex="-1"></a> <span class="dt">"MimeType"</span> <span class="fu">:</span> <span class="st">"application/octet-stream"</span><span class="fu">,</span></span>-<span id="cb351-12"><a href="#cb351-12" aria-hidden="true" tabindex="-1"></a> <span class="dt">"ModTime"</span> <span class="fu">:</span> <span class="st">"2017-05-31T16:15:57.034468261+01:00"</span><span class="fu">,</span></span>-<span id="cb351-13"><a href="#cb351-13" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Name"</span> <span class="fu">:</span> <span class="st">"file.txt"</span><span class="fu">,</span></span>-<span id="cb351-14"><a href="#cb351-14" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Encrypted"</span> <span class="fu">:</span> <span class="st">"v0qpsdq8anpci8n929v3uu9338"</span><span class="fu">,</span></span>-<span id="cb351-15"><a href="#cb351-15" aria-hidden="true" tabindex="-1"></a> <span class="dt">"EncryptedPath"</span> <span class="fu">:</span> <span class="st">"kja9098349023498/v0qpsdq8anpci8n929v3uu9338"</span><span class="fu">,</span></span>-<span id="cb351-16"><a href="#cb351-16" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Path"</span> <span class="fu">:</span> <span class="st">"full/path/goes/here/file.txt"</span><span class="fu">,</span></span>-<span id="cb351-17"><a href="#cb351-17" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Size"</span> <span class="fu">:</span> <span class="dv">6</span><span class="fu">,</span></span>-<span id="cb351-18"><a href="#cb351-18" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Tier"</span> <span class="fu">:</span> <span class="st">"hot"</span><span class="fu">,</span></span>-<span id="cb351-19"><a href="#cb351-19" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<div class="sourceCode" id="cb367"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb367-1"><a href="#cb367-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb367-2"><a href="#cb367-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Hashes"</span> <span class="fu">:</span> <span class="fu">{</span></span>+<span id="cb367-3"><a href="#cb367-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"SHA-1"</span> <span class="fu">:</span> <span class="st">"f572d396fae9206628714fb2ce00f72e94f2258f"</span><span class="fu">,</span></span>+<span id="cb367-4"><a href="#cb367-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"MD5"</span> <span class="fu">:</span> <span class="st">"b1946ac92492d2347c6235b4d2611184"</span><span class="fu">,</span></span>+<span id="cb367-5"><a href="#cb367-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"DropboxHash"</span> <span class="fu">:</span> <span class="st">"ecb65bb98f9d905b70458986c39fcbad7715e5f2fcc3b1f07767d7c83e2438cc"</span></span>+<span id="cb367-6"><a href="#cb367-6" aria-hidden="true" tabindex="-1"></a> <span class="fu">},</span></span>+<span id="cb367-7"><a href="#cb367-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"ID"</span><span class="fu">:</span> <span class="st">"y2djkhiujf83u33"</span><span class="fu">,</span></span>+<span id="cb367-8"><a href="#cb367-8" aria-hidden="true" tabindex="-1"></a> <span class="dt">"OrigID"</span><span class="fu">:</span> <span class="st">"UYOJVTUW00Q1RzTDA"</span><span class="fu">,</span></span>+<span id="cb367-9"><a href="#cb367-9" aria-hidden="true" tabindex="-1"></a> <span class="dt">"IsBucket"</span> <span class="fu">:</span> <span class="kw">false</span><span class="fu">,</span></span>+<span id="cb367-10"><a href="#cb367-10" aria-hidden="true" tabindex="-1"></a> <span class="dt">"IsDir"</span> <span class="fu">:</span> <span class="kw">false</span><span class="fu">,</span></span>+<span id="cb367-11"><a href="#cb367-11" aria-hidden="true" tabindex="-1"></a> <span class="dt">"MimeType"</span> <span class="fu">:</span> <span class="st">"application/octet-stream"</span><span class="fu">,</span></span>+<span id="cb367-12"><a href="#cb367-12" aria-hidden="true" tabindex="-1"></a> <span class="dt">"ModTime"</span> <span class="fu">:</span> <span class="st">"2017-05-31T16:15:57.034468261+01:00"</span><span class="fu">,</span></span>+<span id="cb367-13"><a href="#cb367-13" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Name"</span> <span class="fu">:</span> <span class="st">"file.txt"</span><span class="fu">,</span></span>+<span id="cb367-14"><a href="#cb367-14" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Encrypted"</span> <span class="fu">:</span> <span class="st">"v0qpsdq8anpci8n929v3uu9338"</span><span class="fu">,</span></span>+<span id="cb367-15"><a href="#cb367-15" aria-hidden="true" tabindex="-1"></a> <span class="dt">"EncryptedPath"</span> <span class="fu">:</span> <span class="st">"kja9098349023498/v0qpsdq8anpci8n929v3uu9338"</span><span class="fu">,</span></span>+<span id="cb367-16"><a href="#cb367-16" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Path"</span> <span class="fu">:</span> <span class="st">"full/path/goes/here/file.txt"</span><span class="fu">,</span></span>+<span id="cb367-17"><a href="#cb367-17" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Size"</span> <span class="fu">:</span> <span class="dv">6</span><span class="fu">,</span></span>+<span id="cb367-18"><a href="#cb367-18" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Tier"</span> <span class="fu">:</span> <span class="st">"hot"</span><span class="fu">,</span></span>+<span id="cb367-19"><a href="#cb367-19" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div><p>The exact set of properties included depends on the backend:</p><ul><li>The property IsBucket will only be included for bucket-based@@ -5841,6 +6179,10 @@ use <code>--max-depth 1</code> to stop the recursion.</p><p>The other list commands<code>lsd</code>,<code>lsf</code>,<code>lsjson</code> do not recurse bydefault - use <code>-R</code> to make them recurse.</p>+<p>On the bucket-based remotes (e.g. s3, gcs, b2) a directory is a key+prefix rather than an object, so a listing without <code>-R</code> shows+each prefix as one directory entry and not the objects stored under+it.</p><p>List commands prefer a recursive method that uses more memory butfewer transactions by default. Use <code>--disable ListR</code> tosuppress the behavior. See <a@@ -5850,7 +6192,7 @@ for more details.</p>remotes which can't have empty directories (e.g. s3, swift, or gcs - thebucket-based remotes).</p><pre><code>rclone lsjson remote:path [flags]</code></pre>-<h2 id="options-67">Options</h2>+<h2 id="options-68">Options</h2><pre><code> --dirs-only Show only directories in the listing--encrypted Show the encrypted names--files-only Show only files in the listing@@ -5866,7 +6208,7 @@ bucket-based remotes).</p><p>Options shared with other commands are described next. See the <ahref="https://rclone.org/flags/">global flags page</a> for globaloptions not listed here.</p>-<h3 id="filter-options-19">Filter Options</h3>+<h3 id="filter-options-20">Filter Options</h3><p>Flags for filtering directory listings</p><pre class="text"><code> --delete-excluded Delete files on dest excluded from sync--exclude stringArray Exclude files matching pattern@@ -5892,11 +6234,11 @@ options not listed here.</p>--metadata-include-from stringArray Read metadata include patterns from file (use - to read from stdin)--min-age Duration Only transfer files older than this in s or suffix ms|s|m|h|d|w|M|y (default off)--min-size SizeSuffix Only transfer files bigger than this in KiB or suffix B|K|M|G|T|P (default off)</code></pre>-<h3 id="listing-options-18">Listing Options</h3>+<h3 id="listing-options-19">Listing Options</h3><p>Flags for listing directories</p><pre class="text"><code> --default-time Time Time to show if modtime is unknown for files and directories (default 2000-01-01T00:00:00Z)--fast-list Use recursive list if available; uses more memory but fewer transactions</code></pre>-<h2 id="see-also-67">See Also</h2>+<h2 id="see-also-68">See Also</h2><!-- markdownlint-capture --><!-- markdownlint-disable ul-style line-length --><ul>@@ -5906,7 +6248,7 @@ for rclone commands, flags and backends.</li><!-- markdownlint-restore --><h1 id="rclone-mount">rclone mount</h1><p>Mount the remote as file system on a mountpoint.</p>-<h2 id="synopsis-61">Synopsis</h2>+<h2 id="synopsis-62">Synopsis</h2><p>Rclone mount allows Linux, FreeBSD, macOS and Windows to mount any ofRclone's cloud storage systems as a file system with FUSE.</p><p>First set up your remote using <code>rclone config</code>. Check it@@ -6164,6 +6506,12 @@ server.</p>and <code>serve</code> commands on macOS. For details, see <ahref="https://rclone.org/commands/rclone_mount/#vfs-case-sensitivity">vfs-case-sensitivity</a>.</p><h3 id="nfs-mount">NFS mount</h3>+<p>For macOS (and other platforms where this path is supported), prefer+the dedicated <a+href="https://rclone.org/commands/rclone_nfsmount/">rclone nfsmount</a>+command. It starts the NFS server and performs the mount for you.+<code>rclone mount</code> itself still uses FUSE (macFUSE/FUSE-T) and+does not switch to NFS via a flag.</p><p>This method spins up an NFS server using <ahref="https://rclone.org/commands/rclone_serve_nfs/">serve nfs</a>command and mounts it to the specified mountpoint. If you run this in@@ -6331,7 +6679,7 @@ however there is more chance of the corruption issue above.</p>then there is no chance of corruption.</p><p>This is the same as setting the attr_timeout option inmount.fuse.</p>-<h2 id="filters">Filters</h2>+<h2 id="filters-1">Filters</h2><p>Note that all the rclone filters can be used to select a subset ofthe files to be visible in the mount.</p><h2 id="systemd">systemd</h2>@@ -6364,25 +6712,25 @@ detect it and translate command-line arguments appropriately.</p><p>Now you can run classic mounts like this:</p><pre class="console"><code>mount sftp1:subdir /mnt/data -t rclone -o vfs_cache_mode=writes,sftp_key_file=/path/to/pem</code></pre><p>or create systemd mount units:</p>-<div class="sourceCode" id="cb367"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb367-1"><a href="#cb367-1" aria-hidden="true" tabindex="-1"></a><span class="co"># /etc/systemd/system/mnt-data.mount</span></span>-<span id="cb367-2"><a href="#cb367-2" aria-hidden="true" tabindex="-1"></a><span class="kw">[Unit]</span></span>-<span id="cb367-3"><a href="#cb367-3" aria-hidden="true" tabindex="-1"></a><span class="dt">Description</span><span class="ot">=</span><span class="st">Mount for /mnt/data</span></span>-<span id="cb367-4"><a href="#cb367-4" aria-hidden="true" tabindex="-1"></a><span class="kw">[Mount]</span></span>-<span id="cb367-5"><a href="#cb367-5" aria-hidden="true" tabindex="-1"></a><span class="dt">Type</span><span class="ot">=</span><span class="st">rclone</span></span>-<span id="cb367-6"><a href="#cb367-6" aria-hidden="true" tabindex="-1"></a><span class="dt">What</span><span class="ot">=</span><span class="st">sftp1:subdir</span></span>-<span id="cb367-7"><a href="#cb367-7" aria-hidden="true" tabindex="-1"></a><span class="dt">Where</span><span class="ot">=</span><span class="st">/mnt/data</span></span>-<span id="cb367-8"><a href="#cb367-8" aria-hidden="true" tabindex="-1"></a><span class="dt">Options</span><span class="ot">=</span><span class="st">rw,_netdev,allow_other,args2env,vfs-cache-mode=writes,config=/etc/rclone.conf,cache-dir=/var/rclone</span></span></code></pre></div>+<div class="sourceCode" id="cb383"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb383-1"><a href="#cb383-1" aria-hidden="true" tabindex="-1"></a><span class="co"># /etc/systemd/system/mnt-data.mount</span></span>+<span id="cb383-2"><a href="#cb383-2" aria-hidden="true" tabindex="-1"></a><span class="kw">[Unit]</span></span>+<span id="cb383-3"><a href="#cb383-3" aria-hidden="true" tabindex="-1"></a><span class="dt">Description</span><span class="ot">=</span><span class="st">Mount for /mnt/data</span></span>+<span id="cb383-4"><a href="#cb383-4" aria-hidden="true" tabindex="-1"></a><span class="kw">[Mount]</span></span>+<span id="cb383-5"><a href="#cb383-5" aria-hidden="true" tabindex="-1"></a><span class="dt">Type</span><span class="ot">=</span><span class="st">rclone</span></span>+<span id="cb383-6"><a href="#cb383-6" aria-hidden="true" tabindex="-1"></a><span class="dt">What</span><span class="ot">=</span><span class="st">sftp1:subdir</span></span>+<span id="cb383-7"><a href="#cb383-7" aria-hidden="true" tabindex="-1"></a><span class="dt">Where</span><span class="ot">=</span><span class="st">/mnt/data</span></span>+<span id="cb383-8"><a href="#cb383-8" aria-hidden="true" tabindex="-1"></a><span class="dt">Options</span><span class="ot">=</span><span class="st">rw,_netdev,allow_other,args2env,vfs-cache-mode=writes,config=/etc/rclone.conf,cache-dir=/var/rclone</span></span></code></pre></div><p>optionally accompanied by systemd automount unit</p>-<div class="sourceCode" id="cb368"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb368-1"><a href="#cb368-1" aria-hidden="true" tabindex="-1"></a><span class="co"># /etc/systemd/system/mnt-data.automount</span></span>-<span id="cb368-2"><a href="#cb368-2" aria-hidden="true" tabindex="-1"></a><span class="kw">[Unit]</span></span>-<span id="cb368-3"><a href="#cb368-3" aria-hidden="true" tabindex="-1"></a><span class="dt">Description</span><span class="ot">=</span><span class="st">AutoMount for /mnt/data</span></span>-<span id="cb368-4"><a href="#cb368-4" aria-hidden="true" tabindex="-1"></a><span class="kw">[Automount]</span></span>-<span id="cb368-5"><a href="#cb368-5" aria-hidden="true" tabindex="-1"></a><span class="dt">Where</span><span class="ot">=</span><span class="st">/mnt/data</span></span>-<span id="cb368-6"><a href="#cb368-6" aria-hidden="true" tabindex="-1"></a><span class="dt">TimeoutIdleSec</span><span class="ot">=</span><span class="dv">600</span></span>-<span id="cb368-7"><a href="#cb368-7" aria-hidden="true" tabindex="-1"></a><span class="kw">[Install]</span></span>-<span id="cb368-8"><a href="#cb368-8" aria-hidden="true" tabindex="-1"></a><span class="dt">WantedBy</span><span class="ot">=</span><span class="st">multi-user.target</span></span></code></pre></div>+<div class="sourceCode" id="cb384"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb384-1"><a href="#cb384-1" aria-hidden="true" tabindex="-1"></a><span class="co"># /etc/systemd/system/mnt-data.automount</span></span>+<span id="cb384-2"><a href="#cb384-2" aria-hidden="true" tabindex="-1"></a><span class="kw">[Unit]</span></span>+<span id="cb384-3"><a href="#cb384-3" aria-hidden="true" tabindex="-1"></a><span class="dt">Description</span><span class="ot">=</span><span class="st">AutoMount for /mnt/data</span></span>+<span id="cb384-4"><a href="#cb384-4" aria-hidden="true" tabindex="-1"></a><span class="kw">[Automount]</span></span>+<span id="cb384-5"><a href="#cb384-5" aria-hidden="true" tabindex="-1"></a><span class="dt">Where</span><span class="ot">=</span><span class="st">/mnt/data</span></span>+<span id="cb384-6"><a href="#cb384-6" aria-hidden="true" tabindex="-1"></a><span class="dt">TimeoutIdleSec</span><span class="ot">=</span><span class="dv">600</span></span>+<span id="cb384-7"><a href="#cb384-7" aria-hidden="true" tabindex="-1"></a><span class="kw">[Install]</span></span>+<span id="cb384-8"><a href="#cb384-8" aria-hidden="true" tabindex="-1"></a><span class="dt">WantedBy</span><span class="ot">=</span><span class="st">multi-user.target</span></span></code></pre></div><p>or add in <code>/etc/fstab</code> a line like</p><pre class="console"><code>sftp1:subdir /mnt/data rclone rw,noauto,nofail,_netdev,x-systemd.automount,args2env,vfs_cache_mode=writes,config=/etc/rclone.conf,cache_dir=/var/cache/rclone 0 0</code></pre><p>or use classic Automountd. Remember to provide explicit@@ -6453,7 +6801,7 @@ that will be used to buffer data in advance.</p>memory at all times. The buffered data is bound to one open file andwon't be shared.</p><p>This flag is a upper limit for the used memory per open file. The-buffer will only use memory for data that is downloaded but not not yet+buffer will only use memory for data that is downloaded but not yetread. If the buffer is empty, only a small amount of memory will beused.</p><p>The maximum memory used by rclone for buffering can be up to@@ -6496,11 +6844,12 @@ will start with files that haven't been accessed for the longest. Thiscache flushing strategy is efficient and more relevant files are likelyto remain cached.</p><p>The <code>--vfs-cache-max-age</code> will evict files from the cache-after the set time since last access has passed. The default value of 1-hour will start evicting files from cache that haven't been accessed for-1 hour. When a cached file is accessed the 1 hour timer is reset to 0-and will wait for 1 more hour before evicting. Specify the time with-standard notation, s, m, h, d, w .</p>+after the set time since last access has passed; it is based on access+time, not on when the file was first added to the cache. The default+value of 1 hour will start evicting files from cache that haven't been+accessed for 1 hour. When a cached file is accessed the 1 hour timer is+reset to 0 and will wait for 1 more hour before evicting. Specify the+time with standard notation, s, m, h, d, w .</p><p>You <strong>should not</strong> run two copies of rclone using thesame VFS cache with the same or overlapping remotes if using<code>--vfs-cache-mode > off</code>. This can potentially cause data@@ -6625,8 +6974,8 @@ id="vfs-read-chunk-streams-0-1"><code>--vfs-read-chunk-streams</code><p>Rclone reads <code>--vfs-read-chunk-streams</code> chunks of size<code>--vfs-read-chunk-size</code> concurrently. The size for each readwill stay constant.</p>-<p>This improves performance performance massively on high latency links-or very high bandwidth links to high performance object stores.</p>+<p>This improves performance massively on high latency links or very+high bandwidth links to high performance object stores.</p><p>Some experimentation will be needed to find the optimum values of<code>--vfs-read-chunk-size</code> and<code>--vfs-read-chunk-streams</code> as these will depend on the@@ -6806,7 +7155,7 @@ total 1048578and if there is an error reading the metadata the error will be returnedas <code>{"error":"error string"}</code>.</p><pre><code>rclone mount remote:path /path/to/mountpoint [flags]</code></pre>-<h2 id="options-68">Options</h2>+<h2 id="options-69">Options</h2><pre><code> --allow-idmap Allow id-mapped mounts (Linux 6.12+, mount2 only)--allow-non-empty Allow mounting over a non-empty directory (not supported on Windows)--allow-other Allow access to other users (not supported on Windows)@@ -6866,7 +7215,7 @@ as <code>{"error":"error string"}</code>.</p><p>Options shared with other commands are described next. See the <ahref="https://rclone.org/flags/">global flags page</a> for globaloptions not listed here.</p>-<h3 id="filter-options-20">Filter Options</h3>+<h3 id="filter-options-21">Filter Options</h3><p>Flags for filtering directory listings</p><pre class="text"><code> --delete-excluded Delete files on dest excluded from sync--exclude stringArray Exclude files matching pattern@@ -6892,7 +7241,7 @@ options not listed here.</p>--metadata-include-from stringArray Read metadata include patterns from file (use - to read from stdin)--min-age Duration Only transfer files older than this in s or suffix ms|s|m|h|d|w|M|y (default off)--min-size SizeSuffix Only transfer files bigger than this in KiB or suffix B|K|M|G|T|P (default off)</code></pre>-<h2 id="see-also-68">See Also</h2>+<h2 id="see-also-69">See Also</h2><!-- markdownlint-capture --><!-- markdownlint-disable ul-style line-length --><ul>@@ -6902,7 +7251,7 @@ for rclone commands, flags and backends.</li><!-- markdownlint-restore --><h1 id="rclone-moveto">rclone moveto</h1><p>Move file or directory from source to dest.</p>-<h2 id="synopsis-62">Synopsis</h2>+<h2 id="synopsis-63">Synopsis</h2><p>If source:path is a file or directory then it moves it to a file ordirectory named dest:path.</p><p>This can be used to rename files or upload single files to other than@@ -6978,7 +7327,7 @@ scenarios are not currently supported:</p>after, so it is most useful as a predictor of what SHOULD happen to eachfile (which may or may not match what actually DID).</p><pre><code>rclone moveto source:path dest:path [flags]</code></pre>-<h2 id="options-69">Options</h2>+<h2 id="options-70">Options</h2><pre><code> --absolute Put a leading / in front of path names--combined string Make a combined report of changes to this file--csv Output in CSV format@@ -6999,7 +7348,7 @@ file (which may or may not match what actually DID).</p><p>Options shared with other commands are described next. See the <ahref="https://rclone.org/flags/">global flags page</a> for globaloptions not listed here.</p>-<h3 id="copy-options-6">Copy Options</h3>+<h3 id="copy-options-7">Copy Options</h3><p>Flags for anything which can copy a file</p><pre class="text"><code> --check-first Do all the checks before starting transfers-c, --checksum Check for changes with size & checksum (if available, or fallback to size only)@@ -7035,12 +7384,12 @@ options not listed here.</p>--size-only Skip based on size only, not modtime or checksum--streaming-upload-cutoff SizeSuffix Cutoff for switching to chunked upload if file size is unknown, upload starts after reaching cutoff or when file ends (default 100Ki)-u, --update Skip files that are newer on the destination</code></pre>-<h3 id="important-options-15">Important Options</h3>+<h3 id="important-options-16">Important Options</h3><p>Important flags useful for most commands</p><pre class="text"><code> -n, --dry-run Do a trial run with no permanent changes-i, --interactive Enable interactive mode-v, --verbose count Print lots more stuff (repeat for more)</code></pre>-<h3 id="filter-options-21">Filter Options</h3>+<h3 id="filter-options-22">Filter Options</h3><p>Flags for filtering directory listings</p><pre class="text"><code> --delete-excluded Delete files on dest excluded from sync--exclude stringArray Exclude files matching pattern@@ -7066,11 +7415,11 @@ options not listed here.</p>--metadata-include-from stringArray Read metadata include patterns from file (use - to read from stdin)--min-age Duration Only transfer files older than this in s or suffix ms|s|m|h|d|w|M|y (default off)--min-size SizeSuffix Only transfer files bigger than this in KiB or suffix B|K|M|G|T|P (default off)</code></pre>-<h3 id="listing-options-19">Listing Options</h3>+<h3 id="listing-options-20">Listing Options</h3><p>Flags for listing directories</p><pre class="text"><code> --default-time Time Time to show if modtime is unknown for files and directories (default 2000-01-01T00:00:00Z)--fast-list Use recursive list if available; uses more memory but fewer transactions</code></pre>-<h2 id="see-also-69">See Also</h2>+<h2 id="see-also-70">See Also</h2><!-- markdownlint-capture --><!-- markdownlint-disable ul-style line-length --><ul>@@ -7080,7 +7429,7 @@ for rclone commands, flags and backends.</li><!-- markdownlint-restore --><h1 id="rclone-ncdu">rclone ncdu</h1><p>Explore a remote with a text based user interface.</p>-<h2 id="synopsis-63">Synopsis</h2>+<h2 id="synopsis-64">Synopsis</h2><p>This displays a text based user interface allowing the navigation ofa remote. It is most useful for answering the question - "What is usingall my disk space?".</p>@@ -7135,12 +7484,12 @@ href="https://rclone.org/commands/rclone_tree/">tree</a> command. Tojust get the total size of the remote you can also use the <ahref="https://rclone.org/commands/rclone_size/">size</a> command.</p><pre><code>rclone ncdu remote:path [flags]</code></pre>-<h2 id="options-70">Options</h2>+<h2 id="options-71">Options</h2><pre><code> -h, --help help for ncdu</code></pre><p>Options shared with other commands are described next. See the <ahref="https://rclone.org/flags/">global flags page</a> for globaloptions not listed here.</p>-<h3 id="filter-options-22">Filter Options</h3>+<h3 id="filter-options-23">Filter Options</h3><p>Flags for filtering directory listings</p><pre class="text"><code> --delete-excluded Delete files on dest excluded from sync--exclude stringArray Exclude files matching pattern@@ -7166,11 +7515,11 @@ options not listed here.</p>--metadata-include-from stringArray Read metadata include patterns from file (use - to read from stdin)--min-age Duration Only transfer files older than this in s or suffix ms|s|m|h|d|w|M|y (default off)--min-size SizeSuffix Only transfer files bigger than this in KiB or suffix B|K|M|G|T|P (default off)</code></pre>-<h3 id="listing-options-20">Listing Options</h3>+<h3 id="listing-options-21">Listing Options</h3><p>Flags for listing directories</p><pre class="text"><code> --default-time Time Time to show if modtime is unknown for files and directories (default 2000-01-01T00:00:00Z)--fast-list Use recursive list if available; uses more memory but fewer transactions</code></pre>-<h2 id="see-also-70">See Also</h2>+<h2 id="see-also-71">See Also</h2><!-- markdownlint-capture --><!-- markdownlint-disable ul-style line-length --><ul>@@ -7180,7 +7529,7 @@ for rclone commands, flags and backends.</li><!-- markdownlint-restore --><h1 id="rclone-nfsmount">rclone nfsmount</h1><p>Mount the remote as file system on a mountpoint.</p>-<h2 id="synopsis-64">Synopsis</h2>+<h2 id="synopsis-65">Synopsis</h2><p>Rclone nfsmount allows Linux, FreeBSD, macOS and Windows to mount anyof Rclone's cloud storage systems as a file system with FUSE.</p><p>First set up your remote using <code>rclone config</code>. Check it@@ -7438,6 +7787,12 @@ server.</p>and <code>serve</code> commands on macOS. For details, see <ahref="https://rclone.org/commands/rclone_mount/#vfs-case-sensitivity">vfs-case-sensitivity</a>.</p><h3 id="nfs-mount-1">NFS mount</h3>+<p>For macOS (and other platforms where this path is supported), prefer+the dedicated <a+href="https://rclone.org/commands/rclone_nfsmount/">rclone nfsmount</a>+command. It starts the NFS server and performs the mount for you.+<code>rclone mount</code> itself still uses FUSE (macFUSE/FUSE-T) and+does not switch to NFS via a flag.</p><p>This method spins up an NFS server using <ahref="https://rclone.org/commands/rclone_serve_nfs/">serve nfs</a>command and mounts it to the specified mountpoint. If you run this in@@ -7605,7 +7960,7 @@ however there is more chance of the corruption issue above.</p>then there is no chance of corruption.</p><p>This is the same as setting the attr_timeout option inmount.fuse.</p>-<h2 id="filters-1">Filters</h2>+<h2 id="filters-2">Filters</h2><p>Note that all the rclone filters can be used to select a subset ofthe files to be visible in the mount.</p><h2 id="systemd-1">systemd</h2>@@ -7638,25 +7993,25 @@ detect it and translate command-line arguments appropriately.</p><p>Now you can run classic mounts like this:</p><pre class="console"><code>mount sftp1:subdir /mnt/data -t rclone -o vfs_cache_mode=writes,sftp_key_file=/path/to/pem</code></pre><p>or create systemd mount units:</p>-<div class="sourceCode" id="cb411"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb411-1"><a href="#cb411-1" aria-hidden="true" tabindex="-1"></a><span class="co"># /etc/systemd/system/mnt-data.mount</span></span>-<span id="cb411-2"><a href="#cb411-2" aria-hidden="true" tabindex="-1"></a><span class="kw">[Unit]</span></span>-<span id="cb411-3"><a href="#cb411-3" aria-hidden="true" tabindex="-1"></a><span class="dt">Description</span><span class="ot">=</span><span class="st">Mount for /mnt/data</span></span>-<span id="cb411-4"><a href="#cb411-4" aria-hidden="true" tabindex="-1"></a><span class="kw">[Mount]</span></span>-<span id="cb411-5"><a href="#cb411-5" aria-hidden="true" tabindex="-1"></a><span class="dt">Type</span><span class="ot">=</span><span class="st">rclone</span></span>-<span id="cb411-6"><a href="#cb411-6" aria-hidden="true" tabindex="-1"></a><span class="dt">What</span><span class="ot">=</span><span class="st">sftp1:subdir</span></span>-<span id="cb411-7"><a href="#cb411-7" aria-hidden="true" tabindex="-1"></a><span class="dt">Where</span><span class="ot">=</span><span class="st">/mnt/data</span></span>-<span id="cb411-8"><a href="#cb411-8" aria-hidden="true" tabindex="-1"></a><span class="dt">Options</span><span class="ot">=</span><span class="st">rw,_netdev,allow_other,args2env,vfs-cache-mode=writes,config=/etc/rclone.conf,cache-dir=/var/rclone</span></span></code></pre></div>+<div class="sourceCode" id="cb427"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb427-1"><a href="#cb427-1" aria-hidden="true" tabindex="-1"></a><span class="co"># /etc/systemd/system/mnt-data.mount</span></span>+<span id="cb427-2"><a href="#cb427-2" aria-hidden="true" tabindex="-1"></a><span class="kw">[Unit]</span></span>+<span id="cb427-3"><a href="#cb427-3" aria-hidden="true" tabindex="-1"></a><span class="dt">Description</span><span class="ot">=</span><span class="st">Mount for /mnt/data</span></span>+<span id="cb427-4"><a href="#cb427-4" aria-hidden="true" tabindex="-1"></a><span class="kw">[Mount]</span></span>+<span id="cb427-5"><a href="#cb427-5" aria-hidden="true" tabindex="-1"></a><span class="dt">Type</span><span class="ot">=</span><span class="st">rclone</span></span>+<span id="cb427-6"><a href="#cb427-6" aria-hidden="true" tabindex="-1"></a><span class="dt">What</span><span class="ot">=</span><span class="st">sftp1:subdir</span></span>+<span id="cb427-7"><a href="#cb427-7" aria-hidden="true" tabindex="-1"></a><span class="dt">Where</span><span class="ot">=</span><span class="st">/mnt/data</span></span>+<span id="cb427-8"><a href="#cb427-8" aria-hidden="true" tabindex="-1"></a><span class="dt">Options</span><span class="ot">=</span><span class="st">rw,_netdev,allow_other,args2env,vfs-cache-mode=writes,config=/etc/rclone.conf,cache-dir=/var/rclone</span></span></code></pre></div><p>optionally accompanied by systemd automount unit</p>-<div class="sourceCode" id="cb412"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb412-1"><a href="#cb412-1" aria-hidden="true" tabindex="-1"></a><span class="co"># /etc/systemd/system/mnt-data.automount</span></span>-<span id="cb412-2"><a href="#cb412-2" aria-hidden="true" tabindex="-1"></a><span class="kw">[Unit]</span></span>-<span id="cb412-3"><a href="#cb412-3" aria-hidden="true" tabindex="-1"></a><span class="dt">Description</span><span class="ot">=</span><span class="st">AutoMount for /mnt/data</span></span>-<span id="cb412-4"><a href="#cb412-4" aria-hidden="true" tabindex="-1"></a><span class="kw">[Automount]</span></span>-<span id="cb412-5"><a href="#cb412-5" aria-hidden="true" tabindex="-1"></a><span class="dt">Where</span><span class="ot">=</span><span class="st">/mnt/data</span></span>-<span id="cb412-6"><a href="#cb412-6" aria-hidden="true" tabindex="-1"></a><span class="dt">TimeoutIdleSec</span><span class="ot">=</span><span class="dv">600</span></span>-<span id="cb412-7"><a href="#cb412-7" aria-hidden="true" tabindex="-1"></a><span class="kw">[Install]</span></span>-<span id="cb412-8"><a href="#cb412-8" aria-hidden="true" tabindex="-1"></a><span class="dt">WantedBy</span><span class="ot">=</span><span class="st">multi-user.target</span></span></code></pre></div>+<div class="sourceCode" id="cb428"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb428-1"><a href="#cb428-1" aria-hidden="true" tabindex="-1"></a><span class="co"># /etc/systemd/system/mnt-data.automount</span></span>+<span id="cb428-2"><a href="#cb428-2" aria-hidden="true" tabindex="-1"></a><span class="kw">[Unit]</span></span>+<span id="cb428-3"><a href="#cb428-3" aria-hidden="true" tabindex="-1"></a><span class="dt">Description</span><span class="ot">=</span><span class="st">AutoMount for /mnt/data</span></span>+<span id="cb428-4"><a href="#cb428-4" aria-hidden="true" tabindex="-1"></a><span class="kw">[Automount]</span></span>+<span id="cb428-5"><a href="#cb428-5" aria-hidden="true" tabindex="-1"></a><span class="dt">Where</span><span class="ot">=</span><span class="st">/mnt/data</span></span>+<span id="cb428-6"><a href="#cb428-6" aria-hidden="true" tabindex="-1"></a><span class="dt">TimeoutIdleSec</span><span class="ot">=</span><span class="dv">600</span></span>+<span id="cb428-7"><a href="#cb428-7" aria-hidden="true" tabindex="-1"></a><span class="kw">[Install]</span></span>+<span id="cb428-8"><a href="#cb428-8" aria-hidden="true" tabindex="-1"></a><span class="dt">WantedBy</span><span class="ot">=</span><span class="st">multi-user.target</span></span></code></pre></div><p>or add in <code>/etc/fstab</code> a line like</p><pre class="console"><code>sftp1:subdir /mnt/data rclone rw,noauto,nofail,_netdev,x-systemd.automount,args2env,vfs_cache_mode=writes,config=/etc/rclone.conf,cache_dir=/var/cache/rclone 0 0</code></pre><p>or use classic Automountd. Remember to provide explicit@@ -7727,7 +8082,7 @@ that will be used to buffer data in advance.</p>memory at all times. The buffered data is bound to one open file andwon't be shared.</p><p>This flag is a upper limit for the used memory per open file. The-buffer will only use memory for data that is downloaded but not not yet+buffer will only use memory for data that is downloaded but not yetread. If the buffer is empty, only a small amount of memory will beused.</p><p>The maximum memory used by rclone for buffering can be up to@@ -7770,11 +8125,12 @@ will start with files that haven't been accessed for the longest. Thiscache flushing strategy is efficient and more relevant files are likelyto remain cached.</p><p>The <code>--vfs-cache-max-age</code> will evict files from the cache-after the set time since last access has passed. The default value of 1-hour will start evicting files from cache that haven't been accessed for-1 hour. When a cached file is accessed the 1 hour timer is reset to 0-and will wait for 1 more hour before evicting. Specify the time with-standard notation, s, m, h, d, w .</p>+after the set time since last access has passed; it is based on access+time, not on when the file was first added to the cache. The default+value of 1 hour will start evicting files from cache that haven't been+accessed for 1 hour. When a cached file is accessed the 1 hour timer is+reset to 0 and will wait for 1 more hour before evicting. Specify the+time with standard notation, s, m, h, d, w .</p><p>You <strong>should not</strong> run two copies of rclone using thesame VFS cache with the same or overlapping remotes if using<code>--vfs-cache-mode > off</code>. This can potentially cause data@@ -7900,8 +8256,8 @@ id="vfs-read-chunk-streams-0-3"><code>--vfs-read-chunk-streams</code><p>Rclone reads <code>--vfs-read-chunk-streams</code> chunks of size<code>--vfs-read-chunk-size</code> concurrently. The size for each readwill stay constant.</p>-<p>This improves performance performance massively on high latency links-or very high bandwidth links to high performance object stores.</p>+<p>This improves performance massively on high latency links or very+high bandwidth links to high performance object stores.</p><p>Some experimentation will be needed to find the optimum values of<code>--vfs-read-chunk-size</code> and<code>--vfs-read-chunk-streams</code> as these will depend on the@@ -8081,7 +8437,7 @@ total 1048578and if there is an error reading the metadata the error will be returnedas <code>{"error":"error string"}</code>.</p><pre><code>rclone nfsmount remote:path /path/to/mountpoint [flags]</code></pre>-<h2 id="options-71">Options</h2>+<h2 id="options-72">Options</h2><pre><code> --addr string IPaddress:Port or :Port to bind server to--allow-idmap Allow id-mapped mounts (Linux 6.12+, mount2 only)--allow-non-empty Allow mounting over a non-empty directory (not supported on Windows)@@ -8147,7 +8503,7 @@ as <code>{"error":"error string"}</code>.</p><p>Options shared with other commands are described next. See the <ahref="https://rclone.org/flags/">global flags page</a> for globaloptions not listed here.</p>-<h3 id="filter-options-23">Filter Options</h3>+<h3 id="filter-options-24">Filter Options</h3><p>Flags for filtering directory listings</p><pre class="text"><code> --delete-excluded Delete files on dest excluded from sync--exclude stringArray Exclude files matching pattern@@ -8173,7 +8529,7 @@ options not listed here.</p>--metadata-include-from stringArray Read metadata include patterns from file (use - to read from stdin)--min-age Duration Only transfer files older than this in s or suffix ms|s|m|h|d|w|M|y (default off)--min-size SizeSuffix Only transfer files bigger than this in KiB or suffix B|K|M|G|T|P (default off)</code></pre>-<h2 id="see-also-71">See Also</h2>+<h2 id="see-also-72">See Also</h2><!-- markdownlint-capture --><!-- markdownlint-disable ul-style line-length --><ul>@@ -8183,7 +8539,7 @@ for rclone commands, flags and backends.</li><!-- markdownlint-restore --><h1 id="rclone-obscure">rclone obscure</h1><p>Obscure password for use in the rclone config file.</p>-<h2 id="synopsis-65">Synopsis</h2>+<h2 id="synopsis-66">Synopsis</h2><p>In the rclone config file, human-readable passwords are obscured.Obscuring them is done by encrypting them and writing them out inbase64. This is <strong>not</strong> a secure way of encrypting these@@ -8204,11 +8560,11 @@ encryption - see <ahref="https://rclone.org/commands/rclone_config/">rclone config</a> formore info.</p><pre><code>rclone obscure password [flags]</code></pre>-<h2 id="options-72">Options</h2>+<h2 id="options-73">Options</h2><pre><code> -h, --help help for obscure</code></pre><p>See the <a href="https://rclone.org/flags/">global flags page</a> forglobal options not listed here.</p>-<h2 id="see-also-72">See Also</h2>+<h2 id="see-also-73">See Also</h2><!-- markdownlint-capture --><!-- markdownlint-disable ul-style line-length --><ul>@@ -8218,7 +8574,7 @@ for rclone commands, flags and backends.</li><!-- markdownlint-restore --><h1 id="rclone-rc">rclone rc</h1><p>Run a command against a running rclone.</p>-<h2 id="synopsis-66">Synopsis</h2>+<h2 id="synopsis-67">Synopsis</h2><p>This runs a command against a running rclone. Use the<code>--url</code> flag to specify an non default URL to connect on.This can be either a ":port" which is taken to mean <a@@ -8232,11 +8588,11 @@ class="uri">http://host:port</a>.</p><code>--user</code>, <code>--pass</code>.</p><p>The <code>--unix-socket</code> flag can be used to connect over aunix socket like this</p>-<div class="sourceCode" id="cb433"><pre-class="sourceCode sh"><code class="sourceCode bash"><span id="cb433-1"><a href="#cb433-1" aria-hidden="true" tabindex="-1"></a><span class="co"># start server on /tmp/my.socket</span></span>-<span id="cb433-2"><a href="#cb433-2" aria-hidden="true" tabindex="-1"></a><span class="ex">rclone</span> rcd <span class="at">--rc-addr</span> unix:///tmp/my.socket</span>-<span id="cb433-3"><a href="#cb433-3" aria-hidden="true" tabindex="-1"></a><span class="co"># Connect to it</span></span>-<span id="cb433-4"><a href="#cb433-4" aria-hidden="true" tabindex="-1"></a><span class="ex">rclone</span> rc <span class="at">--unix-socket</span> /tmp/my.socket core/stats</span></code></pre></div>+<div class="sourceCode" id="cb449"><pre+class="sourceCode sh"><code class="sourceCode bash"><span id="cb449-1"><a href="#cb449-1" aria-hidden="true" tabindex="-1"></a><span class="co"># start server on /tmp/my.socket</span></span>+<span id="cb449-2"><a href="#cb449-2" aria-hidden="true" tabindex="-1"></a><span class="ex">rclone</span> rcd <span class="at">--rc-addr</span> unix:///tmp/my.socket</span>+<span id="cb449-3"><a href="#cb449-3" aria-hidden="true" tabindex="-1"></a><span class="co"># Connect to it</span></span>+<span id="cb449-4"><a href="#cb449-4" aria-hidden="true" tabindex="-1"></a><span class="ex">rclone</span> rc <span class="at">--unix-socket</span> /tmp/my.socket core/stats</span></code></pre></div><p>Arguments should be passed in as parameter=value.</p><p>The result will be returned as a JSON object by default.</p><p>The <code>--json</code> parameter can be used to pass in a JSON blob@@ -8249,25 +8605,25 @@ This is useful for rc commands which take the "opt" parameter which byconvention is a dictionary of strings.</p><pre class="text"><code>-o key=value -o key2</code></pre><p>Will place this in the "opt" value</p>-<div class="sourceCode" id="cb435"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb435-1"><a href="#cb435-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span><span class="dt">"key"</span><span class="fu">:</span><span class="st">"value"</span><span class="fu">,</span> <span class="dt">"key2"</span><span class="fu">,</span><span class="dt">""</span><span class="er">)</span></span></code></pre></div>+<div class="sourceCode" id="cb451"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb451-1"><a href="#cb451-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span><span class="dt">"key"</span><span class="fu">:</span><span class="st">"value"</span><span class="fu">,</span> <span class="dt">"key2"</span><span class="fu">,</span><span class="dt">""</span><span class="er">)</span></span></code></pre></div><p>The <code>-a</code>/<code>--arg</code> option can be used to setstrings in the "arg" value. It can be repeated as many times asrequired. This is useful for rc commands which take the "arg" parameterwhich by convention is a list of strings.</p><pre class="text"><code>-a value -a value2</code></pre><p>Will place this in the "arg" value</p>-<div class="sourceCode" id="cb437"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb437-1"><a href="#cb437-1" aria-hidden="true" tabindex="-1"></a><span class="ot">[</span><span class="st">"value"</span><span class="ot">,</span> <span class="st">"value2"</span><span class="ot">]</span></span></code></pre></div>+<div class="sourceCode" id="cb453"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb453-1"><a href="#cb453-1" aria-hidden="true" tabindex="-1"></a><span class="ot">[</span><span class="st">"value"</span><span class="ot">,</span> <span class="st">"value2"</span><span class="ot">]</span></span></code></pre></div><p>Use <code>--loopback</code> to connect to the rclone instance running<code>rclone rc</code>. This is very useful for testing commands withouthaving to run an rclone rc server, e.g.:</p>-<div class="sourceCode" id="cb438"><pre-class="sourceCode sh"><code class="sourceCode bash"><span id="cb438-1"><a href="#cb438-1" aria-hidden="true" tabindex="-1"></a><span class="ex">rclone</span> rc <span class="at">--loopback</span> operations/about fs=/</span></code></pre></div>+<div class="sourceCode" id="cb454"><pre+class="sourceCode sh"><code class="sourceCode bash"><span id="cb454-1"><a href="#cb454-1" aria-hidden="true" tabindex="-1"></a><span class="ex">rclone</span> rc <span class="at">--loopback</span> operations/about fs=/</span></code></pre></div><p>Use <code>rclone rc</code> to see a list of all possiblecommands.</p><pre><code>rclone rc commands parameter [flags]</code></pre>-<h2 id="options-73">Options</h2>+<h2 id="options-74">Options</h2><pre><code> -a, --arg stringArray Argument placed in the "arg" array-h, --help help for rc--json string Input JSON - use instead of key=value args@@ -8280,7 +8636,7 @@ commands.</p>--user string Username to use to rclone remote control</code></pre><p>See the <a href="https://rclone.org/flags/">global flags page</a> forglobal options not listed here.</p>-<h2 id="see-also-73">See Also</h2>+<h2 id="see-also-74">See Also</h2><!-- markdownlint-capture --><!-- markdownlint-disable ul-style line-length --><ul>@@ -8290,7 +8646,7 @@ for rclone commands, flags and backends.</li><!-- markdownlint-restore --><h1 id="rclone-rcat">rclone rcat</h1><p>Copies standard input to file on remote.</p>-<h2 id="synopsis-67">Synopsis</h2>+<h2 id="synopsis-68">Synopsis</h2><p>Reads from standard input (stdin) and copies it to a single remotefile.</p><pre class="console"><code>echo "hello world" | rclone rcat remote:path/to/file@@ -8317,18 +8673,18 @@ chunks can be retried. If you need to transfer a lot of data, you may bebetter off caching it locally and then <code>rclone move</code> it tothe destination which can use retries.</p><pre><code>rclone rcat remote:path [flags]</code></pre>-<h2 id="options-74">Options</h2>+<h2 id="options-75">Options</h2><pre><code> -h, --help help for rcat--size int File size hint to preallocate (default -1)</code></pre><p>Options shared with other commands are described next. See the <ahref="https://rclone.org/flags/">global flags page</a> for globaloptions not listed here.</p>-<h3 id="important-options-16">Important Options</h3>+<h3 id="important-options-17">Important Options</h3><p>Important flags useful for most commands</p><pre class="text"><code> -n, --dry-run Do a trial run with no permanent changes-i, --interactive Enable interactive mode-v, --verbose count Print lots more stuff (repeat for more)</code></pre>-<h2 id="see-also-74">See Also</h2>+<h2 id="see-also-75">See Also</h2><!-- markdownlint-capture --><!-- markdownlint-disable ul-style line-length --><ul>@@ -8338,7 +8694,7 @@ for rclone commands, flags and backends.</li><!-- markdownlint-restore --><h1 id="rclone-rcd">rclone rcd</h1><p>Run rclone listening to remote control commands only.</p>-<h2 id="synopsis-68">Synopsis</h2>+<h2 id="synopsis-69">Synopsis</h2><p>This runs rclone so that it only listens to remote controlcommands.</p><p>This is useful if you are controlling rclone via the rc API.</p>@@ -8372,6 +8728,14 @@ bytes the server will accept in the HTTP header.</p>for all responses, will overriding existing values. The flag may berepeated to add multiple headers. Use the format<code>Header-Name: value</code>.</p>+<p><code>--rc-allow-origin</code> enables cross-origin (CORS) requests+from browsers by setting the <code>Access-Control-Allow-Origin</code>+header on every response to the origin given, e.g.+<code>--rc-allow-origin https://app.example.com</code>. Browser CORS+preflight OPTIONS requests are answered by rclone without needing+authentication as browsers never send credentials with them. Setting+this to <code>*</code> allows any origin which is a security risk on an+authenticated server.</p><p><code>--rc-baseurl</code> controls the URL prefix that rclone servesfrom. By default rclone will serve from the root. If you used<code>--rc-baseurl "/rclone"</code> then rclone would serve from a URL@@ -8468,55 +8832,101 @@ asc).</td><td style="text-align: left;">Currently unused.</td></tr><tr class="even">+<td style="text-align: left;">.Path</td>+<td style="text-align: left;"></td>+<td style="text-align: left;">The path of the directory from the root+with leading and trailing slashes, e.g. '/dir/'.</td>+</tr>+<tr class="odd">+<td style="text-align: left;">.IsRoot</td>+<td style="text-align: left;"></td>+<td style="text-align: left;">Boolean for if this is the root+directory.</td>+</tr>+<tr class="even">+<td style="text-align: left;">.UpLink</td>+<td style="text-align: left;"></td>+<td style="text-align: left;">The link to the parent directory.</td>+</tr>+<tr class="odd">+<td style="text-align: left;">.Static</td>+<td style="text-align: left;"></td>+<td style="text-align: left;">Boolean for if the listing is being+written as a static page by rclone index.</td>+</tr>+<tr class="even">+<td style="text-align: left;">.NumDirs</td>+<td style="text-align: left;"></td>+<td style="text-align: left;">The number of directories in the+listing.</td>+</tr>+<tr class="odd">+<td style="text-align: left;">.NumFiles</td>+<td style="text-align: left;"></td>+<td style="text-align: left;">The number of files in the listing.</td>+</tr>+<tr class="even">+<td style="text-align: left;">.TotalSize</td>+<td style="text-align: left;"></td>+<td style="text-align: left;">The total size in bytes of the files in+the listing.</td>+</tr>+<tr class="odd"><td style="text-align: left;">.Breadcrumb</td><td style="text-align: left;"></td><td style="text-align: left;">Allows for creating a relativenavigation.</td></tr>-<tr class="odd">+<tr class="even"><td style="text-align: left;"></td><td style="text-align: left;">.Link</td><td style="text-align: left;">The link of the Text relative to theroot.</td></tr>-<tr class="even">+<tr class="odd"><td style="text-align: left;"></td><td style="text-align: left;">.Text</td><td style="text-align: left;">The Name of the directory.</td></tr>-<tr class="odd">+<tr class="even"><td style="text-align: left;">.Entries</td><td style="text-align: left;"></td><td style="text-align: left;">Information about a specificfile/directory.</td></tr>-<tr class="even">+<tr class="odd"><td style="text-align: left;"></td><td style="text-align: left;">.URL</td><td style="text-align: left;">The url of an entry.</td></tr>-<tr class="odd">+<tr class="even"><td style="text-align: left;"></td><td style="text-align: left;">.Leaf</td><td style="text-align: left;">Currently same as '.URL' but intended tobe just the name.</td></tr>-<tr class="even">+<tr class="odd"><td style="text-align: left;"></td><td style="text-align: left;">.IsDir</td><td style="text-align: left;">Boolean for if an entry is a directory ornot.</td></tr>-<tr class="odd">+<tr class="even"><td style="text-align: left;"></td><td style="text-align: left;">.Size</td><td style="text-align: left;">Size in bytes of the entry.</td></tr>-<tr class="even">+<tr class="odd"><td style="text-align: left;"></td><td style="text-align: left;">.ModTime</td><td style="text-align: left;">The UTC timestamp of an entry.</td></tr>+<tr class="even">+<td style="text-align: left;"></td>+<td style="text-align: left;">.MimeType</td>+<td style="text-align: left;">The MIME type of an entry guessed from its+name, or 'inode/directory'.</td>+</tr></tbody></table><p>The server also makes the following functions available so that they@@ -8579,12 +8989,22 @@ and BCrypt for basic authentication. Bcrypt is recommended.</p><pre class="console"><code>touch htpasswdhtpasswd -B htpasswd userhtpasswd -B htpasswd anotherUser</code></pre>+<p>Use the same hash format and cost for every entry in the file. If the+entries differ, the time taken to reject a wrong password can reveal+which usernames exist.</p><p>The password file can be updated while rclone is running.</p>+<p>When authentication is enabled every request must be authenticated,+including HTTP OPTIONS requests, as the answer to an OPTIONS request can+reveal whether a path exists. Health checks or monitoring which use+unauthenticated OPTIONS requests will need to send credentials. The+exception is browser CORS preflight requests when+<code>--rc-allow-origin</code> is set, which are answered without+authentication as browsers never send credentials with them.</p><p>Use <code>--rc-realm</code> to set the authentication realm.</p><p>Use <code>--rc-salt</code> to change the password hashing salt fromthe default.</p><pre><code>rclone rcd <path to files to serve>* [flags]</code></pre>-<h2 id="options-75">Options</h2>+<h2 id="options-76">Options</h2><pre><code> -h, --help help for rcd</code></pre><p>Options shared with other commands are described next. See the <ahref="https://rclone.org/flags/">global flags page</a> for global@@ -8616,13 +9036,8 @@ options not listed here.</p>--rc-server-write-timeout Duration Timeout for server writing data (default 1h0m0s)--rc-template string User-specified template--rc-user string User name for authentication- --rc-user-from-header string User name from a defined HTTP header- --rc-web-fetch-url string URL to fetch the releases for webgui (default "https://api.github.com/repos/rclone/rclone-webui-react/releases/latest")- --rc-web-gui Launch WebGUI on localhost- --rc-web-gui-force-update Force update to latest version of web gui- --rc-web-gui-no-open-browser Don't open the browser automatically- --rc-web-gui-update Check and update to latest version of web gui</code></pre>-<h2 id="see-also-75">See Also</h2>+ --rc-user-from-header string User name from a defined HTTP header</code></pre>+<h2 id="see-also-76">See Also</h2><!-- markdownlint-capture --><!-- markdownlint-disable ul-style line-length --><ul>@@ -8632,7 +9047,7 @@ for rclone commands, flags and backends.</li><!-- markdownlint-restore --><h1 id="rclone-rmdirs">rclone rmdirs</h1><p>Remove empty directories under the path.</p>-<h2 id="synopsis-69">Synopsis</h2>+<h2 id="synopsis-70">Synopsis</h2><p>This recursively removes any empty directories (including directoriesthat only contain empty directories), that it finds under the path. Theroot path itself will also be removed if it is empty, unless you supply@@ -8651,18 +9066,18 @@ number.</p><p>To delete a path and any objects in it, use the <ahref="https://rclone.org/commands/rclone_purge/">purge</a> command.</p><pre><code>rclone rmdirs remote:path [flags]</code></pre>-<h2 id="options-76">Options</h2>+<h2 id="options-77">Options</h2><pre><code> -h, --help help for rmdirs--leave-root Do not remove root directory if empty</code></pre><p>Options shared with other commands are described next. See the <ahref="https://rclone.org/flags/">global flags page</a> for globaloptions not listed here.</p>-<h3 id="important-options-17">Important Options</h3>+<h3 id="important-options-18">Important Options</h3><p>Important flags useful for most commands</p><pre class="text"><code> -n, --dry-run Do a trial run with no permanent changes-i, --interactive Enable interactive mode-v, --verbose count Print lots more stuff (repeat for more)</code></pre>-<h2 id="see-also-76">See Also</h2>+<h2 id="see-also-77">See Also</h2><!-- markdownlint-capture --><!-- markdownlint-disable ul-style line-length --><ul>@@ -8672,7 +9087,7 @@ for rclone commands, flags and backends.</li><!-- markdownlint-restore --><h1 id="rclone-selfupdate">rclone selfupdate</h1><p>Update the rclone binary.</p>-<h2 id="synopsis-70">Synopsis</h2>+<h2 id="synopsis-71">Synopsis</h2><p>This command downloads the latest release of rclone and replaces thecurrently running binary. The download is verified with a hashsum andcryptographically signed signature; see <a@@ -8724,7 +9139,7 @@ success.</p>manually following the <a href="https://rclone.org/install/">installdocumentation</a>.</p><pre><code>rclone selfupdate [flags]</code></pre>-<h2 id="options-77">Options</h2>+<h2 id="options-78">Options</h2><pre><code> --beta Install beta release--check Check for latest release, do not download-h, --help help for selfupdate@@ -8734,7 +9149,7 @@ documentation</a>.</p>--version string Install the given rclone version (default: latest)</code></pre><p>See the <a href="https://rclone.org/flags/">global flags page</a> forglobal options not listed here.</p>-<h2 id="see-also-77">See Also</h2>+<h2 id="see-also-78">See Also</h2><!-- markdownlint-capture --><!-- markdownlint-disable ul-style line-length --><ul>@@ -8744,7 +9159,7 @@ for rclone commands, flags and backends.</li><!-- markdownlint-restore --><h1 id="rclone-serve">rclone serve</h1><p>Serve a remote over a protocol.</p>-<h2 id="synopsis-71">Synopsis</h2>+<h2 id="synopsis-72">Synopsis</h2><p>Serve a remote over a given protocol. Requires the use of asubcommand to specify the protocol, e.g.</p><pre class="console"><code>rclone serve http remote:</code></pre>@@ -8755,11 +9170,11 @@ fields depends on whether the remote supports metadata.</p><p>Each subcommand has its own options which you can see in theirhelp.</p><pre><code>rclone serve <protocol> [opts] <remote> [flags]</code></pre>-<h2 id="options-78">Options</h2>+<h2 id="options-79">Options</h2><pre><code> -h, --help help for serve</code></pre><p>See the <a href="https://rclone.org/flags/">global flags page</a> forglobal options not listed here.</p>-<h2 id="see-also-78">See Also</h2>+<h2 id="see-also-79">See Also</h2><!-- markdownlint-capture --><!-- markdownlint-disable ul-style line-length --><ul>@@ -8787,7 +9202,7 @@ serve webdav</a> - Serve remote:path over WebDAV.</li><!-- markdownlint-restore --><h1 id="rclone-serve-dlna">rclone serve dlna</h1><p>Serve remote:path over DLNA</p>-<h2 id="synopsis-72">Synopsis</h2>+<h2 id="synopsis-73">Synopsis</h2><p>Run a DLNA media server for media stored in an rclone remote. Manydevices, such as the Xbox and PlayStation, can automatically discoverthis server in the LAN and play audio/video from it. VLC is also@@ -8847,7 +9262,7 @@ that will be used to buffer data in advance.</p>memory at all times. The buffered data is bound to one open file andwon't be shared.</p><p>This flag is a upper limit for the used memory per open file. The-buffer will only use memory for data that is downloaded but not not yet+buffer will only use memory for data that is downloaded but not yetread. If the buffer is empty, only a small amount of memory will beused.</p><p>The maximum memory used by rclone for buffering can be up to@@ -8890,11 +9305,12 @@ will start with files that haven't been accessed for the longest. Thiscache flushing strategy is efficient and more relevant files are likelyto remain cached.</p><p>The <code>--vfs-cache-max-age</code> will evict files from the cache-after the set time since last access has passed. The default value of 1-hour will start evicting files from cache that haven't been accessed for-1 hour. When a cached file is accessed the 1 hour timer is reset to 0-and will wait for 1 more hour before evicting. Specify the time with-standard notation, s, m, h, d, w .</p>+after the set time since last access has passed; it is based on access+time, not on when the file was first added to the cache. The default+value of 1 hour will start evicting files from cache that haven't been+accessed for 1 hour. When a cached file is accessed the 1 hour timer is+reset to 0 and will wait for 1 more hour before evicting. Specify the+time with standard notation, s, m, h, d, w .</p><p>You <strong>should not</strong> run two copies of rclone using thesame VFS cache with the same or overlapping remotes if using<code>--vfs-cache-mode > off</code>. This can potentially cause data@@ -9020,8 +9436,8 @@ id="vfs-read-chunk-streams-0-5"><code>--vfs-read-chunk-streams</code><p>Rclone reads <code>--vfs-read-chunk-streams</code> chunks of size<code>--vfs-read-chunk-size</code> concurrently. The size for each readwill stay constant.</p>-<p>This improves performance performance massively on high latency links-or very high bandwidth links to high performance object stores.</p>+<p>This improves performance massively on high latency links or very+high bandwidth links to high performance object stores.</p><p>Some experimentation will be needed to find the optimum values of<code>--vfs-read-chunk-size</code> and<code>--vfs-read-chunk-streams</code> as these will depend on the@@ -9201,7 +9617,7 @@ total 1048578and if there is an error reading the metadata the error will be returnedas <code>{"error":"error string"}</code>.</p><pre><code>rclone serve dlna remote:path [flags]</code></pre>-<h2 id="options-79">Options</h2>+<h2 id="options-80">Options</h2><pre><code> --addr string The ip:port or :port to bind the DLNA http server to (default ":7879")--announce-interval Duration The interval between SSDP announcements (default 12m0s)--dir-cache-time Duration Time to cache directory entries for (default 5m0s)@@ -9244,7 +9660,7 @@ as <code>{"error":"error string"}</code>.</p><p>Options shared with other commands are described next. See the <ahref="https://rclone.org/flags/">global flags page</a> for globaloptions not listed here.</p>-<h3 id="filter-options-24">Filter Options</h3>+<h3 id="filter-options-25">Filter Options</h3><p>Flags for filtering directory listings</p><pre class="text"><code> --delete-excluded Delete files on dest excluded from sync--exclude stringArray Exclude files matching pattern@@ -9270,7 +9686,7 @@ options not listed here.</p>--metadata-include-from stringArray Read metadata include patterns from file (use - to read from stdin)--min-age Duration Only transfer files older than this in s or suffix ms|s|m|h|d|w|M|y (default off)--min-size SizeSuffix Only transfer files bigger than this in KiB or suffix B|K|M|G|T|P (default off)</code></pre>-<h2 id="see-also-79">See Also</h2>+<h2 id="see-also-80">See Also</h2><!-- markdownlint-capture --><!-- markdownlint-disable ul-style line-length --><ul>@@ -9280,7 +9696,7 @@ options not listed here.</p><!-- markdownlint-restore --><h1 id="rclone-serve-docker">rclone serve docker</h1><p>Serve any remote on docker's volume plugin API.</p>-<h2 id="synopsis-73">Synopsis</h2>+<h2 id="synopsis-74">Synopsis</h2><p>This command implements the Docker volume plugin API allowing dockerto use rclone as a data storage mechanism for various cloud providers.rclone provides <a href="/docker">docker volume plugin</a> based on@@ -9411,7 +9827,7 @@ that will be used to buffer data in advance.</p>memory at all times. The buffered data is bound to one open file andwon't be shared.</p><p>This flag is a upper limit for the used memory per open file. The-buffer will only use memory for data that is downloaded but not not yet+buffer will only use memory for data that is downloaded but not yetread. If the buffer is empty, only a small amount of memory will beused.</p><p>The maximum memory used by rclone for buffering can be up to@@ -9454,11 +9870,12 @@ will start with files that haven't been accessed for the longest. Thiscache flushing strategy is efficient and more relevant files are likelyto remain cached.</p><p>The <code>--vfs-cache-max-age</code> will evict files from the cache-after the set time since last access has passed. The default value of 1-hour will start evicting files from cache that haven't been accessed for-1 hour. When a cached file is accessed the 1 hour timer is reset to 0-and will wait for 1 more hour before evicting. Specify the time with-standard notation, s, m, h, d, w .</p>+after the set time since last access has passed; it is based on access+time, not on when the file was first added to the cache. The default+value of 1 hour will start evicting files from cache that haven't been+accessed for 1 hour. When a cached file is accessed the 1 hour timer is+reset to 0 and will wait for 1 more hour before evicting. Specify the+time with standard notation, s, m, h, d, w .</p><p>You <strong>should not</strong> run two copies of rclone using thesame VFS cache with the same or overlapping remotes if using<code>--vfs-cache-mode > off</code>. This can potentially cause data@@ -9584,8 +10001,8 @@ id="vfs-read-chunk-streams-0-7"><code>--vfs-read-chunk-streams</code><p>Rclone reads <code>--vfs-read-chunk-streams</code> chunks of size<code>--vfs-read-chunk-size</code> concurrently. The size for each readwill stay constant.</p>-<p>This improves performance performance massively on high latency links-or very high bandwidth links to high performance object stores.</p>+<p>This improves performance massively on high latency links or very+high bandwidth links to high performance object stores.</p><p>Some experimentation will be needed to find the optimum values of<code>--vfs-read-chunk-size</code> and<code>--vfs-read-chunk-streams</code> as these will depend on the@@ -9765,7 +10182,7 @@ total 1048578and if there is an error reading the metadata the error will be returnedas <code>{"error":"error string"}</code>.</p><pre><code>rclone serve docker [flags]</code></pre>-<h2 id="options-80">Options</h2>+<h2 id="options-81">Options</h2><pre><code> --allow-idmap Allow id-mapped mounts (Linux 6.12+, mount2 only)--allow-non-empty Allow mounting over a non-empty directory (not supported on Windows)--allow-other Allow access to other users (not supported on Windows)@@ -9830,7 +10247,7 @@ as <code>{"error":"error string"}</code>.</p><p>Options shared with other commands are described next. See the <ahref="https://rclone.org/flags/">global flags page</a> for globaloptions not listed here.</p>-<h3 id="filter-options-25">Filter Options</h3>+<h3 id="filter-options-26">Filter Options</h3><p>Flags for filtering directory listings</p><pre class="text"><code> --delete-excluded Delete files on dest excluded from sync--exclude stringArray Exclude files matching pattern@@ -9856,7 +10273,7 @@ options not listed here.</p>--metadata-include-from stringArray Read metadata include patterns from file (use - to read from stdin)--min-age Duration Only transfer files older than this in s or suffix ms|s|m|h|d|w|M|y (default off)--min-size SizeSuffix Only transfer files bigger than this in KiB or suffix B|K|M|G|T|P (default off)</code></pre>-<h2 id="see-also-80">See Also</h2>+<h2 id="see-also-81">See Also</h2><!-- markdownlint-capture --><!-- markdownlint-disable ul-style line-length --><ul>@@ -9866,7 +10283,7 @@ options not listed here.</p><!-- markdownlint-restore --><h1 id="rclone-serve-ftp">rclone serve ftp</h1><p>Serve remote:path over FTP.</p>-<h2 id="synopsis-74">Synopsis</h2>+<h2 id="synopsis-75">Synopsis</h2><p>Run a basic FTP server to serve a remote over FTP protocol. This canbe viewed with a FTP client or you can make a remote of type FTP to readand write it.</p>@@ -9920,7 +10337,7 @@ that will be used to buffer data in advance.</p>memory at all times. The buffered data is bound to one open file andwon't be shared.</p><p>This flag is a upper limit for the used memory per open file. The-buffer will only use memory for data that is downloaded but not not yet+buffer will only use memory for data that is downloaded but not yetread. If the buffer is empty, only a small amount of memory will beused.</p><p>The maximum memory used by rclone for buffering can be up to@@ -9963,11 +10380,12 @@ will start with files that haven't been accessed for the longest. Thiscache flushing strategy is efficient and more relevant files are likelyto remain cached.</p><p>The <code>--vfs-cache-max-age</code> will evict files from the cache-after the set time since last access has passed. The default value of 1-hour will start evicting files from cache that haven't been accessed for-1 hour. When a cached file is accessed the 1 hour timer is reset to 0-and will wait for 1 more hour before evicting. Specify the time with-standard notation, s, m, h, d, w .</p>+after the set time since last access has passed; it is based on access+time, not on when the file was first added to the cache. The default+value of 1 hour will start evicting files from cache that haven't been+accessed for 1 hour. When a cached file is accessed the 1 hour timer is+reset to 0 and will wait for 1 more hour before evicting. Specify the+time with standard notation, s, m, h, d, w .</p><p>You <strong>should not</strong> run two copies of rclone using thesame VFS cache with the same or overlapping remotes if using<code>--vfs-cache-mode > off</code>. This can potentially cause data@@ -10093,8 +10511,8 @@ id="vfs-read-chunk-streams-0-9"><code>--vfs-read-chunk-streams</code><p>Rclone reads <code>--vfs-read-chunk-streams</code> chunks of size<code>--vfs-read-chunk-size</code> concurrently. The size for each readwill stay constant.</p>-<p>This improves performance performance massively on high latency links-or very high bandwidth links to high performance object stores.</p>+<p>This improves performance massively on high latency links or very+high bandwidth links to high performance object stores.</p><p>Some experimentation will be needed to find the optimum values of<code>--vfs-read-chunk-size</code> and<code>--vfs-read-chunk-streams</code> as these will depend on the@@ -10296,35 +10714,69 @@ proxy program to make a complete config.</p><ul><li><code>_root</code> - root to use for the backend</li></ul>-<p>And it may have this parameter</p>+<p>And it may have these parameters</p><ul><li><code>_obscure</code> - comma separated strings for parameters toobscure</li>+<li><code>_secret_access_key</code> - the secret for S3 access key auth+(see below)</li></ul><p>If password authentication was used by the client, input to the proxyprocess (on STDIN) would look similar to this:</p>-<div class="sourceCode" id="cb504"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb504-1"><a href="#cb504-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>-<span id="cb504-2"><a href="#cb504-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"user"</span><span class="fu">:</span> <span class="st">"me"</span><span class="fu">,</span></span>-<span id="cb504-3"><a href="#cb504-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"pass"</span><span class="fu">:</span> <span class="st">"mypassword"</span></span>-<span id="cb504-4"><a href="#cb504-4" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<div class="sourceCode" id="cb520"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb520-1"><a href="#cb520-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb520-2"><a href="#cb520-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"user"</span><span class="fu">:</span> <span class="st">"me"</span><span class="fu">,</span></span>+<span id="cb520-3"><a href="#cb520-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"pass"</span><span class="fu">:</span> <span class="st">"mypassword"</span><span class="fu">,</span></span>+<span id="cb520-4"><a href="#cb520-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"client_ip"</span><span class="fu">:</span> <span class="st">"192.168.1.1"</span></span>+<span id="cb520-5"><a href="#cb520-5" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div><p>If public-key authentication was used by the client, input to theproxy process (on STDIN) would look similar to this:</p>-<div class="sourceCode" id="cb505"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb505-1"><a href="#cb505-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>-<span id="cb505-2"><a href="#cb505-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"user"</span><span class="fu">:</span> <span class="st">"me"</span><span class="fu">,</span></span>-<span id="cb505-3"><a href="#cb505-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"public_key"</span><span class="fu">:</span> <span class="st">"AAAAB3NzaC1yc2EAAAADAQABAAABAQDuwESFdAe14hVS6omeyX7edc...JQdf"</span></span>-<span id="cb505-4"><a href="#cb505-4" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<div class="sourceCode" id="cb521"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb521-1"><a href="#cb521-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb521-2"><a href="#cb521-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"user"</span><span class="fu">:</span> <span class="st">"me"</span><span class="fu">,</span></span>+<span id="cb521-3"><a href="#cb521-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"public_key"</span><span class="fu">:</span> <span class="st">"AAAAB3NzaC1yc2EAAAADAQABAAABAQDuwESFdAe14hVS6omeyX7edc...JQdf"</span><span class="fu">,</span></span>+<span id="cb521-4"><a href="#cb521-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"client_ip"</span><span class="fu">:</span> <span class="st">"192.168.1.1"</span></span>+<span id="cb521-5"><a href="#cb521-5" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<p>If the client authenticated with an S3 access key+(<code>rclone serve s3</code>), the client never sends its secret, only+a signature made with it, so the input contains just the access key ID+as the <code>user</code> with no <code>pass</code> or+<code>public_key</code>:</p>+<div class="sourceCode" id="cb522"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb522-1"><a href="#cb522-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb522-2"><a href="#cb522-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"user"</span><span class="fu">:</span> <span class="st">"AKIAIOSFODNN7EXAMPLE"</span><span class="fu">,</span></span>+<span id="cb522-3"><a href="#cb522-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"client_ip"</span><span class="fu">:</span> <span class="st">"192.168.1.1"</span></span>+<span id="cb522-4"><a href="#cb522-4" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<p>In this case the program must look up the secret access key for that+access key ID and return it in the <code>_secret_access_key</code> field+of the output. Rclone then uses that secret to verify the signature on+the request, refusing the request if it does not match. This means the+proxy program is the source of truth for both the credentials and the+backend they map to. If the program does not return+<code>_secret_access_key</code> or returns it empty the request is+refused.</p>+<p>The program's answer for an access key ID is cached (see below) but+is checked with the program again after 5 minutes even if the access key+ID is in constant use, so revoking an access key ID in the program takes+effect within 5 minutes. A rotated secret takes effect on the first+request signed with it.</p>+<p>The <code>client_ip</code> key holds the IP address the client+connected from, without a port number. It can be used to restrict logins+to certain networks, or to log authentication attempts centrally. It is+omitted if the client has no IP address, for example when connecting+over a unix socket. Note that if rclone is behind a reverse proxy this+will be the address of the reverse proxy and not the original+client.</p><p>And as an example return this on STDOUT</p>-<div class="sourceCode" id="cb506"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb506-1"><a href="#cb506-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>-<span id="cb506-2"><a href="#cb506-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"type"</span><span class="fu">:</span> <span class="st">"sftp"</span><span class="fu">,</span></span>-<span id="cb506-3"><a href="#cb506-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"_root"</span><span class="fu">:</span> <span class="st">""</span><span class="fu">,</span></span>-<span id="cb506-4"><a href="#cb506-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"_obscure"</span><span class="fu">:</span> <span class="st">"pass"</span><span class="fu">,</span></span>-<span id="cb506-5"><a href="#cb506-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"user"</span><span class="fu">:</span> <span class="st">"me"</span><span class="fu">,</span></span>-<span id="cb506-6"><a href="#cb506-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">"pass"</span><span class="fu">:</span> <span class="st">"mypassword"</span><span class="fu">,</span></span>-<span id="cb506-7"><a href="#cb506-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"host"</span><span class="fu">:</span> <span class="st">"sftp.example.com"</span></span>-<span id="cb506-8"><a href="#cb506-8" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<div class="sourceCode" id="cb523"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb523-1"><a href="#cb523-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb523-2"><a href="#cb523-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"type"</span><span class="fu">:</span> <span class="st">"sftp"</span><span class="fu">,</span></span>+<span id="cb523-3"><a href="#cb523-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"_root"</span><span class="fu">:</span> <span class="st">""</span><span class="fu">,</span></span>+<span id="cb523-4"><a href="#cb523-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"_obscure"</span><span class="fu">:</span> <span class="st">"pass"</span><span class="fu">,</span></span>+<span id="cb523-5"><a href="#cb523-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"user"</span><span class="fu">:</span> <span class="st">"me"</span><span class="fu">,</span></span>+<span id="cb523-6"><a href="#cb523-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">"pass"</span><span class="fu">:</span> <span class="st">"mypassword"</span><span class="fu">,</span></span>+<span id="cb523-7"><a href="#cb523-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"host"</span><span class="fu">:</span> <span class="st">"sftp.example.com"</span></span>+<span id="cb523-8"><a href="#cb523-8" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div><p>This would mean that an SFTP backend would be created on the fly forthe <code>user</code> and <code>pass</code>/<code>public_key</code>returned in the output to the host given. Note that since@@ -10337,16 +10789,23 @@ make the <code>user</code> be <code>[email protected]</code> and then setthe <code>host</code> to <code>example.com</code> in the output and theuser to <code>user</code>. For security you'd probably want to restrictthe <code>host</code> to a limited list.</p>-<p>An internal cache of backends is keyed on the <code>user</code> and a-hash of the <code>pass</code> or <code>public_key</code>. This means-that if a user's password or public-key changes, or the proxy returns-different config parameters (eg a rotated <code>api_key</code>), a fresh-backend will be created on the next request rather than the cached one-being reused.</p>+<p>An internal cache of backends is keyed on the <code>user</code>, the+kind of credential presented, a hash of the <code>pass</code> or+<code>public_key</code>, and the <code>client_ip</code>. This means that+if a user's password or public-key changes, the client connects from a+new IP address, or the proxy returns different config parameters (eg a+rotated <code>api_key</code>), a fresh backend will be created on the+next request rather than the cached one being reused. A backend created+for a public key login is never reused for a password login, so a public+key cannot be sent as a password to skip the program.</p>+<p>A cached answer is checked with the program again after 5 minutes+even if the user is in constant use, so revoking a user in the program+takes effect within 5 minutes. For <code>serve sftp</code> this applies+to new logins - an existing connection is not closed.</p><p>This can be used to build general purpose proxies to any kind ofbackend that rclone supports.</p><pre><code>rclone serve ftp remote:path [flags]</code></pre>-<h2 id="options-81">Options</h2>+<h2 id="options-82">Options</h2><pre><code> --addr string IPaddress:Port or :Port to bind server to (default "localhost:2121")--auth-proxy string A program to use to create the backend from the auth--cert string TLS PEM key (concatenation of certificate and CA certificate)@@ -10392,7 +10851,7 @@ backend that rclone supports.</p><p>Options shared with other commands are described next. See the <ahref="https://rclone.org/flags/">global flags page</a> for globaloptions not listed here.</p>-<h3 id="filter-options-26">Filter Options</h3>+<h3 id="filter-options-27">Filter Options</h3><p>Flags for filtering directory listings</p><pre class="text"><code> --delete-excluded Delete files on dest excluded from sync--exclude stringArray Exclude files matching pattern@@ -10418,7 +10877,7 @@ options not listed here.</p>--metadata-include-from stringArray Read metadata include patterns from file (use - to read from stdin)--min-age Duration Only transfer files older than this in s or suffix ms|s|m|h|d|w|M|y (default off)--min-size SizeSuffix Only transfer files bigger than this in KiB or suffix B|K|M|G|T|P (default off)</code></pre>-<h2 id="see-also-81">See Also</h2>+<h2 id="see-also-82">See Also</h2><!-- markdownlint-capture --><!-- markdownlint-disable ul-style line-length --><ul>@@ -10428,7 +10887,7 @@ options not listed here.</p><!-- markdownlint-restore --><h1 id="rclone-serve-http">rclone serve http</h1><p>Serve the remote over HTTP.</p>-<h2 id="synopsis-75">Synopsis</h2>+<h2 id="synopsis-76">Synopsis</h2><p>Run a basic web server to serve a remote over HTTP. This can beviewed in a web browser or you can make a remote of type http read fromit.</p>@@ -10462,6 +10921,14 @@ the server will accept in the HTTP header.</p>all responses, will overriding existing values. The flag may be repeatedto add multiple headers. Use the format<code>Header-Name: value</code>.</p>+<p><code>--allow-origin</code> enables cross-origin (CORS) requests from+browsers by setting the <code>Access-Control-Allow-Origin</code> header+on every response to the origin given, e.g.+<code>--allow-origin https://app.example.com</code>. Browser CORS+preflight OPTIONS requests are answered by rclone without needing+authentication as browsers never send credentials with them. Setting+this to <code>*</code> allows any origin which is a security risk on an+authenticated server.</p><p><code>--baseurl</code> controls the URL prefix that rclone servesfrom. By default rclone will serve from the root. If you used<code>--baseurl "/rclone"</code> then rclone would serve from a URL@@ -10558,55 +11025,101 @@ asc).</td><td style="text-align: left;">Currently unused.</td></tr><tr class="even">+<td style="text-align: left;">.Path</td>+<td style="text-align: left;"></td>+<td style="text-align: left;">The path of the directory from the root+with leading and trailing slashes, e.g. '/dir/'.</td>+</tr>+<tr class="odd">+<td style="text-align: left;">.IsRoot</td>+<td style="text-align: left;"></td>+<td style="text-align: left;">Boolean for if this is the root+directory.</td>+</tr>+<tr class="even">+<td style="text-align: left;">.UpLink</td>+<td style="text-align: left;"></td>+<td style="text-align: left;">The link to the parent directory.</td>+</tr>+<tr class="odd">+<td style="text-align: left;">.Static</td>+<td style="text-align: left;"></td>+<td style="text-align: left;">Boolean for if the listing is being+written as a static page by rclone index.</td>+</tr>+<tr class="even">+<td style="text-align: left;">.NumDirs</td>+<td style="text-align: left;"></td>+<td style="text-align: left;">The number of directories in the+listing.</td>+</tr>+<tr class="odd">+<td style="text-align: left;">.NumFiles</td>+<td style="text-align: left;"></td>+<td style="text-align: left;">The number of files in the listing.</td>+</tr>+<tr class="even">+<td style="text-align: left;">.TotalSize</td>+<td style="text-align: left;"></td>+<td style="text-align: left;">The total size in bytes of the files in+the listing.</td>+</tr>+<tr class="odd"><td style="text-align: left;">.Breadcrumb</td><td style="text-align: left;"></td><td style="text-align: left;">Allows for creating a relativenavigation.</td></tr>-<tr class="odd">+<tr class="even"><td style="text-align: left;"></td><td style="text-align: left;">.Link</td><td style="text-align: left;">The link of the Text relative to theroot.</td></tr>-<tr class="even">+<tr class="odd"><td style="text-align: left;"></td><td style="text-align: left;">.Text</td><td style="text-align: left;">The Name of the directory.</td></tr>-<tr class="odd">+<tr class="even"><td style="text-align: left;">.Entries</td><td style="text-align: left;"></td><td style="text-align: left;">Information about a specificfile/directory.</td></tr>-<tr class="even">+<tr class="odd"><td style="text-align: left;"></td><td style="text-align: left;">.URL</td><td style="text-align: left;">The url of an entry.</td></tr>-<tr class="odd">+<tr class="even"><td style="text-align: left;"></td><td style="text-align: left;">.Leaf</td><td style="text-align: left;">Currently same as '.URL' but intended tobe just the name.</td></tr>-<tr class="even">+<tr class="odd"><td style="text-align: left;"></td><td style="text-align: left;">.IsDir</td><td style="text-align: left;">Boolean for if an entry is a directory ornot.</td></tr>-<tr class="odd">+<tr class="even"><td style="text-align: left;"></td><td style="text-align: left;">.Size</td><td style="text-align: left;">Size in bytes of the entry.</td></tr>-<tr class="even">+<tr class="odd"><td style="text-align: left;"></td><td style="text-align: left;">.ModTime</td><td style="text-align: left;">The UTC timestamp of an entry.</td></tr>+<tr class="even">+<td style="text-align: left;"></td>+<td style="text-align: left;">.MimeType</td>+<td style="text-align: left;">The MIME type of an entry guessed from its+name, or 'inode/directory'.</td>+</tr></tbody></table><p>The server also makes the following functions available so that they@@ -10669,7 +11182,17 @@ BCrypt for basic authentication. Bcrypt is recommended.</p><pre class="console"><code>touch htpasswdhtpasswd -B htpasswd userhtpasswd -B htpasswd anotherUser</code></pre>+<p>Use the same hash format and cost for every entry in the file. If the+entries differ, the time taken to reject a wrong password can reveal+which usernames exist.</p><p>The password file can be updated while rclone is running.</p>+<p>When authentication is enabled every request must be authenticated,+including HTTP OPTIONS requests, as the answer to an OPTIONS request can+reveal whether a path exists. Health checks or monitoring which use+unauthenticated OPTIONS requests will need to send credentials. The+exception is browser CORS preflight requests when+<code>--allow-origin</code> is set, which are answered without+authentication as browsers never send credentials with them.</p><p>Use <code>--realm</code> to set the authentication realm.</p><p>Use <code>--salt</code> to change the password hashing salt from thedefault.</p>@@ -10711,7 +11234,7 @@ that will be used to buffer data in advance.</p>memory at all times. The buffered data is bound to one open file andwon't be shared.</p><p>This flag is a upper limit for the used memory per open file. The-buffer will only use memory for data that is downloaded but not not yet+buffer will only use memory for data that is downloaded but not yetread. If the buffer is empty, only a small amount of memory will beused.</p><p>The maximum memory used by rclone for buffering can be up to@@ -10754,11 +11277,12 @@ will start with files that haven't been accessed for the longest. Thiscache flushing strategy is efficient and more relevant files are likelyto remain cached.</p><p>The <code>--vfs-cache-max-age</code> will evict files from the cache-after the set time since last access has passed. The default value of 1-hour will start evicting files from cache that haven't been accessed for-1 hour. When a cached file is accessed the 1 hour timer is reset to 0-and will wait for 1 more hour before evicting. Specify the time with-standard notation, s, m, h, d, w .</p>+after the set time since last access has passed; it is based on access+time, not on when the file was first added to the cache. The default+value of 1 hour will start evicting files from cache that haven't been+accessed for 1 hour. When a cached file is accessed the 1 hour timer is+reset to 0 and will wait for 1 more hour before evicting. Specify the+time with standard notation, s, m, h, d, w .</p><p>You <strong>should not</strong> run two copies of rclone using thesame VFS cache with the same or overlapping remotes if using<code>--vfs-cache-mode > off</code>. This can potentially cause data@@ -10884,8 +11408,8 @@ id="vfs-read-chunk-streams-0-11"><code>--vfs-read-chunk-streams</code><p>Rclone reads <code>--vfs-read-chunk-streams</code> chunks of size<code>--vfs-read-chunk-size</code> concurrently. The size for each readwill stay constant.</p>-<p>This improves performance performance massively on high latency links-or very high bandwidth links to high performance object stores.</p>+<p>This improves performance massively on high latency links or very+high bandwidth links to high performance object stores.</p><p>Some experimentation will be needed to find the optimum values of<code>--vfs-read-chunk-size</code> and<code>--vfs-read-chunk-streams</code> as these will depend on the@@ -11087,35 +11611,69 @@ proxy program to make a complete config.</p><ul><li><code>_root</code> - root to use for the backend</li></ul>-<p>And it may have this parameter</p>+<p>And it may have these parameters</p><ul><li><code>_obscure</code> - comma separated strings for parameters toobscure</li>+<li><code>_secret_access_key</code> - the secret for S3 access key auth+(see below)</li></ul><p>If password authentication was used by the client, input to the proxyprocess (on STDIN) would look similar to this:</p>-<div class="sourceCode" id="cb525"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb525-1"><a href="#cb525-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>-<span id="cb525-2"><a href="#cb525-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"user"</span><span class="fu">:</span> <span class="st">"me"</span><span class="fu">,</span></span>-<span id="cb525-3"><a href="#cb525-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"pass"</span><span class="fu">:</span> <span class="st">"mypassword"</span></span>-<span id="cb525-4"><a href="#cb525-4" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<div class="sourceCode" id="cb542"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb542-1"><a href="#cb542-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb542-2"><a href="#cb542-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"user"</span><span class="fu">:</span> <span class="st">"me"</span><span class="fu">,</span></span>+<span id="cb542-3"><a href="#cb542-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"pass"</span><span class="fu">:</span> <span class="st">"mypassword"</span><span class="fu">,</span></span>+<span id="cb542-4"><a href="#cb542-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"client_ip"</span><span class="fu">:</span> <span class="st">"192.168.1.1"</span></span>+<span id="cb542-5"><a href="#cb542-5" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div><p>If public-key authentication was used by the client, input to theproxy process (on STDIN) would look similar to this:</p>-<div class="sourceCode" id="cb526"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb526-1"><a href="#cb526-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>-<span id="cb526-2"><a href="#cb526-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"user"</span><span class="fu">:</span> <span class="st">"me"</span><span class="fu">,</span></span>-<span id="cb526-3"><a href="#cb526-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"public_key"</span><span class="fu">:</span> <span class="st">"AAAAB3NzaC1yc2EAAAADAQABAAABAQDuwESFdAe14hVS6omeyX7edc...JQdf"</span></span>-<span id="cb526-4"><a href="#cb526-4" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<div class="sourceCode" id="cb543"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb543-1"><a href="#cb543-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb543-2"><a href="#cb543-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"user"</span><span class="fu">:</span> <span class="st">"me"</span><span class="fu">,</span></span>+<span id="cb543-3"><a href="#cb543-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"public_key"</span><span class="fu">:</span> <span class="st">"AAAAB3NzaC1yc2EAAAADAQABAAABAQDuwESFdAe14hVS6omeyX7edc...JQdf"</span><span class="fu">,</span></span>+<span id="cb543-4"><a href="#cb543-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"client_ip"</span><span class="fu">:</span> <span class="st">"192.168.1.1"</span></span>+<span id="cb543-5"><a href="#cb543-5" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<p>If the client authenticated with an S3 access key+(<code>rclone serve s3</code>), the client never sends its secret, only+a signature made with it, so the input contains just the access key ID+as the <code>user</code> with no <code>pass</code> or+<code>public_key</code>:</p>+<div class="sourceCode" id="cb544"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb544-1"><a href="#cb544-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb544-2"><a href="#cb544-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"user"</span><span class="fu">:</span> <span class="st">"AKIAIOSFODNN7EXAMPLE"</span><span class="fu">,</span></span>+<span id="cb544-3"><a href="#cb544-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"client_ip"</span><span class="fu">:</span> <span class="st">"192.168.1.1"</span></span>+<span id="cb544-4"><a href="#cb544-4" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<p>In this case the program must look up the secret access key for that+access key ID and return it in the <code>_secret_access_key</code> field+of the output. Rclone then uses that secret to verify the signature on+the request, refusing the request if it does not match. This means the+proxy program is the source of truth for both the credentials and the+backend they map to. If the program does not return+<code>_secret_access_key</code> or returns it empty the request is+refused.</p>+<p>The program's answer for an access key ID is cached (see below) but+is checked with the program again after 5 minutes even if the access key+ID is in constant use, so revoking an access key ID in the program takes+effect within 5 minutes. A rotated secret takes effect on the first+request signed with it.</p>+<p>The <code>client_ip</code> key holds the IP address the client+connected from, without a port number. It can be used to restrict logins+to certain networks, or to log authentication attempts centrally. It is+omitted if the client has no IP address, for example when connecting+over a unix socket. Note that if rclone is behind a reverse proxy this+will be the address of the reverse proxy and not the original+client.</p><p>And as an example return this on STDOUT</p>-<div class="sourceCode" id="cb527"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb527-1"><a href="#cb527-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>-<span id="cb527-2"><a href="#cb527-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"type"</span><span class="fu">:</span> <span class="st">"sftp"</span><span class="fu">,</span></span>-<span id="cb527-3"><a href="#cb527-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"_root"</span><span class="fu">:</span> <span class="st">""</span><span class="fu">,</span></span>-<span id="cb527-4"><a href="#cb527-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"_obscure"</span><span class="fu">:</span> <span class="st">"pass"</span><span class="fu">,</span></span>-<span id="cb527-5"><a href="#cb527-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"user"</span><span class="fu">:</span> <span class="st">"me"</span><span class="fu">,</span></span>-<span id="cb527-6"><a href="#cb527-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">"pass"</span><span class="fu">:</span> <span class="st">"mypassword"</span><span class="fu">,</span></span>-<span id="cb527-7"><a href="#cb527-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"host"</span><span class="fu">:</span> <span class="st">"sftp.example.com"</span></span>-<span id="cb527-8"><a href="#cb527-8" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<div class="sourceCode" id="cb545"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb545-1"><a href="#cb545-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb545-2"><a href="#cb545-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"type"</span><span class="fu">:</span> <span class="st">"sftp"</span><span class="fu">,</span></span>+<span id="cb545-3"><a href="#cb545-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"_root"</span><span class="fu">:</span> <span class="st">""</span><span class="fu">,</span></span>+<span id="cb545-4"><a href="#cb545-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"_obscure"</span><span class="fu">:</span> <span class="st">"pass"</span><span class="fu">,</span></span>+<span id="cb545-5"><a href="#cb545-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"user"</span><span class="fu">:</span> <span class="st">"me"</span><span class="fu">,</span></span>+<span id="cb545-6"><a href="#cb545-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">"pass"</span><span class="fu">:</span> <span class="st">"mypassword"</span><span class="fu">,</span></span>+<span id="cb545-7"><a href="#cb545-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"host"</span><span class="fu">:</span> <span class="st">"sftp.example.com"</span></span>+<span id="cb545-8"><a href="#cb545-8" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div><p>This would mean that an SFTP backend would be created on the fly forthe <code>user</code> and <code>pass</code>/<code>public_key</code>returned in the output to the host given. Note that since@@ -11128,16 +11686,23 @@ make the <code>user</code> be <code>[email protected]</code> and then setthe <code>host</code> to <code>example.com</code> in the output and theuser to <code>user</code>. For security you'd probably want to restrictthe <code>host</code> to a limited list.</p>-<p>An internal cache of backends is keyed on the <code>user</code> and a-hash of the <code>pass</code> or <code>public_key</code>. This means-that if a user's password or public-key changes, or the proxy returns-different config parameters (eg a rotated <code>api_key</code>), a fresh-backend will be created on the next request rather than the cached one-being reused.</p>+<p>An internal cache of backends is keyed on the <code>user</code>, the+kind of credential presented, a hash of the <code>pass</code> or+<code>public_key</code>, and the <code>client_ip</code>. This means that+if a user's password or public-key changes, the client connects from a+new IP address, or the proxy returns different config parameters (eg a+rotated <code>api_key</code>), a fresh backend will be created on the+next request rather than the cached one being reused. A backend created+for a public key login is never reused for a password login, so a public+key cannot be sent as a password to skip the program.</p>+<p>A cached answer is checked with the program again after 5 minutes+even if the user is in constant use, so revoking a user in the program+takes effect within 5 minutes. For <code>serve sftp</code> this applies+to new logins - an existing connection is not closed.</p><p>This can be used to build general purpose proxies to any kind ofbackend that rclone supports.</p><pre><code>rclone serve http remote:path [flags]</code></pre>-<h2 id="options-82">Options</h2>+<h2 id="options-83">Options</h2><pre><code> --addr stringArray IPaddress:Port or :Port to bind server to (default 127.0.0.1:8080)--allow-origin string Origin which cross-domain request (CORS) can be executed from--auth-proxy string A program to use to create the backend from the auth@@ -11196,7 +11761,7 @@ backend that rclone supports.</p><p>Options shared with other commands are described next. See the <ahref="https://rclone.org/flags/">global flags page</a> for globaloptions not listed here.</p>-<h3 id="filter-options-27">Filter Options</h3>+<h3 id="filter-options-28">Filter Options</h3><p>Flags for filtering directory listings</p><pre class="text"><code> --delete-excluded Delete files on dest excluded from sync--exclude stringArray Exclude files matching pattern@@ -11222,7 +11787,7 @@ options not listed here.</p>--metadata-include-from stringArray Read metadata include patterns from file (use - to read from stdin)--min-age Duration Only transfer files older than this in s or suffix ms|s|m|h|d|w|M|y (default off)--min-size SizeSuffix Only transfer files bigger than this in KiB or suffix B|K|M|G|T|P (default off)</code></pre>-<h2 id="see-also-82">See Also</h2>+<h2 id="see-also-83">See Also</h2><!-- markdownlint-capture --><!-- markdownlint-disable ul-style line-length --><ul>@@ -11232,7 +11797,7 @@ options not listed here.</p><!-- markdownlint-restore --><h1 id="rclone-serve-nfs">rclone serve nfs</h1><p>Serve the remote as an NFS mount</p>-<h2 id="synopsis-76">Synopsis</h2>+<h2 id="synopsis-77">Synopsis</h2><p>Create an NFS server that serves the given remote over thenetwork.</p><p>This implements an NFSv3 server to serve any rclone remote via@@ -11281,20 +11846,20 @@ default is <code>1000000</code>, but consider lowering this limit if theserver's system resource usage causes problems. This is only used by the<code>memory</code> type cache.</p><p>To serve NFS over the network use following command:</p>-<div class="sourceCode" id="cb531"><pre-class="sourceCode sh"><code class="sourceCode bash"><span id="cb531-1"><a href="#cb531-1" aria-hidden="true" tabindex="-1"></a><span class="ex">rclone</span> serve nfs remote: <span class="at">--addr</span> 0.0.0.0:<span class="va">$PORT</span> <span class="at">--vfs-cache-mode</span><span class="op">=</span>full</span></code></pre></div>+<div class="sourceCode" id="cb549"><pre+class="sourceCode sh"><code class="sourceCode bash"><span id="cb549-1"><a href="#cb549-1" aria-hidden="true" tabindex="-1"></a><span class="ex">rclone</span> serve nfs remote: <span class="at">--addr</span> 0.0.0.0:<span class="va">$PORT</span> <span class="at">--vfs-cache-mode</span><span class="op">=</span>full</span></code></pre></div><p>This specifies a port that can be used in the mount command. To mountthe server under Linux/macOS, use the following command:</p>-<div class="sourceCode" id="cb532"><pre-class="sourceCode sh"><code class="sourceCode bash"><span id="cb532-1"><a href="#cb532-1" aria-hidden="true" tabindex="-1"></a><span class="fu">mount</span> <span class="at">-t</span> nfs <span class="at">-o</span> port=<span class="va">$PORT</span>,mountport=<span class="va">$PORT</span>,tcp <span class="va">$HOSTNAME</span>:/ path/to/mountpoint</span></code></pre></div>+<div class="sourceCode" id="cb550"><pre+class="sourceCode sh"><code class="sourceCode bash"><span id="cb550-1"><a href="#cb550-1" aria-hidden="true" tabindex="-1"></a><span class="fu">mount</span> <span class="at">-t</span> nfs <span class="at">-o</span> port=<span class="va">$PORT</span>,mountport=<span class="va">$PORT</span>,tcp <span class="va">$HOSTNAME</span>:/ path/to/mountpoint</span></code></pre></div><p>Where <code>$PORT</code> is the same port number used in the<code>serve nfs</code> command and <code>$HOSTNAME</code> is the networkaddress of the machine that <code>serve nfs</code> was run on.</p><p>NFS clients can also mount a subdirectory of the served remote byincluding it in the mount path. For example to mount only the<code>photos/2024</code> subdirectory:</p>-<div class="sourceCode" id="cb533"><pre-class="sourceCode sh"><code class="sourceCode bash"><span id="cb533-1"><a href="#cb533-1" aria-hidden="true" tabindex="-1"></a><span class="fu">mount</span> <span class="at">-t</span> nfs <span class="at">-o</span> port=<span class="va">$PORT</span>,mountport=<span class="va">$PORT</span>,tcp <span class="va">$HOSTNAME</span>:/photos/2024 path/to/mountpoint</span></code></pre></div>+<div class="sourceCode" id="cb551"><pre+class="sourceCode sh"><code class="sourceCode bash"><span id="cb551-1"><a href="#cb551-1" aria-hidden="true" tabindex="-1"></a><span class="fu">mount</span> <span class="at">-t</span> nfs <span class="at">-o</span> port=<span class="va">$PORT</span>,mountport=<span class="va">$PORT</span>,tcp <span class="va">$HOSTNAME</span>:/photos/2024 path/to/mountpoint</span></code></pre></div><p>The subpath is resolved within the served remote and must refer to anexisting directory (not a file or a symlink). Subpath mounts are aconvenience equivalent to mounting <code>/</code> and changing@@ -11346,7 +11911,7 @@ that will be used to buffer data in advance.</p>memory at all times. The buffered data is bound to one open file andwon't be shared.</p><p>This flag is a upper limit for the used memory per open file. The-buffer will only use memory for data that is downloaded but not not yet+buffer will only use memory for data that is downloaded but not yetread. If the buffer is empty, only a small amount of memory will beused.</p><p>The maximum memory used by rclone for buffering can be up to@@ -11389,11 +11954,12 @@ will start with files that haven't been accessed for the longest. Thiscache flushing strategy is efficient and more relevant files are likelyto remain cached.</p><p>The <code>--vfs-cache-max-age</code> will evict files from the cache-after the set time since last access has passed. The default value of 1-hour will start evicting files from cache that haven't been accessed for-1 hour. When a cached file is accessed the 1 hour timer is reset to 0-and will wait for 1 more hour before evicting. Specify the time with-standard notation, s, m, h, d, w .</p>+after the set time since last access has passed; it is based on access+time, not on when the file was first added to the cache. The default+value of 1 hour will start evicting files from cache that haven't been+accessed for 1 hour. When a cached file is accessed the 1 hour timer is+reset to 0 and will wait for 1 more hour before evicting. Specify the+time with standard notation, s, m, h, d, w .</p><p>You <strong>should not</strong> run two copies of rclone using thesame VFS cache with the same or overlapping remotes if using<code>--vfs-cache-mode > off</code>. This can potentially cause data@@ -11519,8 +12085,8 @@ id="vfs-read-chunk-streams-0-13"><code>--vfs-read-chunk-streams</code><p>Rclone reads <code>--vfs-read-chunk-streams</code> chunks of size<code>--vfs-read-chunk-size</code> concurrently. The size for each readwill stay constant.</p>-<p>This improves performance performance massively on high latency links-or very high bandwidth links to high performance object stores.</p>+<p>This improves performance massively on high latency links or very+high bandwidth links to high performance object stores.</p><p>Some experimentation will be needed to find the optimum values of<code>--vfs-read-chunk-size</code> and<code>--vfs-read-chunk-streams</code> as these will depend on the@@ -11700,7 +12266,7 @@ total 1048578and if there is an error reading the metadata the error will be returnedas <code>{"error":"error string"}</code>.</p><pre><code>rclone serve nfs remote:path [flags]</code></pre>-<h2 id="options-83">Options</h2>+<h2 id="options-84">Options</h2><pre><code> --addr string IPaddress:Port or :Port to bind server to--dir-cache-time Duration Time to cache directory entries for (default 5m0s)--dir-perms FileMode Directory permissions (default 777)@@ -11742,7 +12308,7 @@ as <code>{"error":"error string"}</code>.</p><p>Options shared with other commands are described next. See the <ahref="https://rclone.org/flags/">global flags page</a> for globaloptions not listed here.</p>-<h3 id="filter-options-28">Filter Options</h3>+<h3 id="filter-options-29">Filter Options</h3><p>Flags for filtering directory listings</p><pre class="text"><code> --delete-excluded Delete files on dest excluded from sync--exclude stringArray Exclude files matching pattern@@ -11768,7 +12334,7 @@ options not listed here.</p>--metadata-include-from stringArray Read metadata include patterns from file (use - to read from stdin)--min-age Duration Only transfer files older than this in s or suffix ms|s|m|h|d|w|M|y (default off)--min-size SizeSuffix Only transfer files bigger than this in KiB or suffix B|K|M|G|T|P (default off)</code></pre>-<h2 id="see-also-83">See Also</h2>+<h2 id="see-also-84">See Also</h2><!-- markdownlint-capture --><!-- markdownlint-disable ul-style line-length --><ul>@@ -11778,7 +12344,7 @@ options not listed here.</p><!-- markdownlint-restore --><h1 id="rclone-serve-restic">rclone serve restic</h1><p>Serve the remote for restic's REST API.</p>-<h2 id="synopsis-77">Synopsis</h2>+<h2 id="synopsis-78">Synopsis</h2><p>Run a basic web server to serve a remote over restic's REST backendAPI over HTTP. This allows restic to use rclone as a data storagemechanism for cloud providers that restic does not support directly.</p>@@ -11867,6 +12433,14 @@ the server will accept in the HTTP header.</p>all responses, will overriding existing values. The flag may be repeatedto add multiple headers. Use the format<code>Header-Name: value</code>.</p>+<p><code>--allow-origin</code> enables cross-origin (CORS) requests from+browsers by setting the <code>Access-Control-Allow-Origin</code> header+on every response to the origin given, e.g.+<code>--allow-origin https://app.example.com</code>. Browser CORS+preflight OPTIONS requests are answered by rclone without needing+authentication as browsers never send credentials with them. Setting+this to <code>*</code> allows any origin which is a security risk on an+authenticated server.</p><p><code>--baseurl</code> controls the URL prefix that rclone servesfrom. By default rclone will serve from the root. If you used<code>--baseurl "/rclone"</code> then rclone would serve from a URL@@ -11937,12 +12511,22 @@ BCrypt for basic authentication. Bcrypt is recommended.</p><pre class="console"><code>touch htpasswdhtpasswd -B htpasswd userhtpasswd -B htpasswd anotherUser</code></pre>+<p>Use the same hash format and cost for every entry in the file. If the+entries differ, the time taken to reject a wrong password can reveal+which usernames exist.</p><p>The password file can be updated while rclone is running.</p>+<p>When authentication is enabled every request must be authenticated,+including HTTP OPTIONS requests, as the answer to an OPTIONS request can+reveal whether a path exists. Health checks or monitoring which use+unauthenticated OPTIONS requests will need to send credentials. The+exception is browser CORS preflight requests when+<code>--allow-origin</code> is set, which are answered without+authentication as browsers never send credentials with them.</p><p>Use <code>--realm</code> to set the authentication realm.</p><p>Use <code>--salt</code> to change the password hashing salt from thedefault.</p><pre><code>rclone serve restic remote:path [flags]</code></pre>-<h2 id="options-84">Options</h2>+<h2 id="options-85">Options</h2><pre><code> --addr stringArray IPaddress:Port or :Port to bind server to (default 127.0.0.1:8080)--allow-origin string Origin which cross-domain request (CORS) can be executed from--append-only Disallow deletion of repository data@@ -11967,7 +12551,7 @@ default.</p>--user-from-header string User name from a defined HTTP header</code></pre><p>See the <a href="https://rclone.org/flags/">global flags page</a> forglobal options not listed here.</p>-<h2 id="see-also-84">See Also</h2>+<h2 id="see-also-85">See Also</h2><!-- markdownlint-capture --><!-- markdownlint-disable ul-style line-length --><ul>@@ -11977,7 +12561,7 @@ global options not listed here.</p><!-- markdownlint-restore --><h1 id="rclone-serve-s3">rclone serve s3</h1><p>Serve remote:path over s3.</p>-<h2 id="synopsis-78">Synopsis</h2>+<h2 id="synopsis-79">Synopsis</h2><p><code>serve s3</code> implements a basic s3 server that serves aremote via s3. This can be viewed with an s3 client, or you can make an<a href="https://rclone.org/s3/">s3 type remote</a> to read and write to@@ -11992,6 +12576,21 @@ docs</a>).</p><p><code>--auth-key</code> can be repeated for multiple auth pairs. If<code>--auth-key</code> is not provided then <code>serve s3</code> willallow anonymous access.</p>+<p>Alternatively <code>--auth-proxy</code> can be used to look up the+secret for each access key ID and choose the backend it maps to (see <a+href="#auth-proxy">Auth Proxy</a> below). When an auth proxy is in use+<code>--auth-key</code> is ignored and every request must be signed with+the secret the proxy returns for its access key ID. Each access key ID+is a separate user, whose multipart uploads and object metadata are+private to it, even if the proxy maps several access key IDs to the same+backend.</p>+<p>As on AWS, every <code>x-amz-*</code> header in an authenticated+request must be in its signed headers list; a request with an unsigned+<code>x-amz-*</code> header is refused with <code>AccessDenied</code>.+S3 clients sign these headers as a matter of course, but a header added+after signing (for example by a proxy in front of the server, or+<code>x-amz-meta-*</code> headers sent with a presigned URL which only+signs the host) is refused.</p><p>Like all rclone flags <code>--auth-key</code> can be set viaenvironment variables, in this case <code>RCLONE_AUTH_KEY</code>. Sincethis flag can be repeated, the input to <code>RCLONE_AUTH_KEY</code> is@@ -12015,7 +12614,15 @@ bucket name as a part of the hostname (such as mybucket.local)</p><p>Use <code>--etag-hash</code> if you want to change the hash uses forthe <code>ETag</code>. Note that using anything other than<code>MD5</code> (the default) is likely to cause problems for S3-clients which rely on the Etag being the MD5.</p>+clients which rely on the Etag being the MD5. Use+<code>--etag-hash auto</code> to use the best hash the backend supports+- with <code>--auth-proxy</code> that of each user's backend.</p>+<p>The <code>ETag</code> of an object made by a multipart upload is its+hash, like any other object, rather than the <code>-N</code> suffixed+<code>ETag</code> AWS S3 gives these. If the remote doesn't have the MD5+of objects, <code>HeadObject</code> and listings only return the MD5 of+objects uploaded through <code>serve s3</code> while the object's+metadata is kept (see below).</p><h2 id="quickstart-1">Quickstart</h2><p>For a simple set up, to serve <code>remote:path</code> over s3, runthe server like this:</p>@@ -12024,9 +12631,9 @@ the server like this:</p>with a command like this:</p><pre class="console"><code>rclone serve s3 --auth-key ACCESS_KEY_ID,SECRET_ACCESS_KEY local:/path/to/folder</code></pre><p>The <code>rclone.conf</code> for the server could look like this:</p>-<div class="sourceCode" id="cb561"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb561-1"><a href="#cb561-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[local]</span></span>-<span id="cb561-2"><a href="#cb561-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> local</span></span></code></pre></div>+<div class="sourceCode" id="cb579"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb579-1"><a href="#cb579-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[local]</span></span>+<span id="cb579-2"><a href="#cb579-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> local</span></span></code></pre></div><p>The <code>local</code> configuration is optional though. If you runthe server with a <code>remote:path</code> like<code>/path/to/folder</code> (without the <code>local:</code> prefix and@@ -12035,35 +12642,81 @@ default configuration, which will be visible as a warning in the logs.But it will run nonetheless.</p><p>This will be compatible with an rclone (client) remote configurationwhich is defined like this:</p>-<div class="sourceCode" id="cb562"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb562-1"><a href="#cb562-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[serves3]</span></span>-<span id="cb562-2"><a href="#cb562-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>-<span id="cb562-3"><a href="#cb562-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Rclone</span></span>-<span id="cb562-4"><a href="#cb562-4" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> http://127.0.0.1:8080/</span></span>-<span id="cb562-5"><a href="#cb562-5" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> ACCESS_KEY_ID</span></span>-<span id="cb562-6"><a href="#cb562-6" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> SECRET_ACCESS_KEY</span></span></code></pre></div>+<div class="sourceCode" id="cb580"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb580-1"><a href="#cb580-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[serves3]</span></span>+<span id="cb580-2"><a href="#cb580-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>+<span id="cb580-3"><a href="#cb580-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Rclone</span></span>+<span id="cb580-4"><a href="#cb580-4" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> http://127.0.0.1:8080/</span></span>+<span id="cb580-5"><a href="#cb580-5" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> ACCESS_KEY_ID</span></span>+<span id="cb580-6"><a href="#cb580-6" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> SECRET_ACCESS_KEY</span></span></code></pre></div>+<h2 id="object-uploads-put">Object uploads (PUT)</h2>+<p>A <code>PutObject</code> upload only ever changes the object at its+key atomically, on success, a failed or interrupted PUT neither removes+nor overwrites the object already stored at the key, and never leaves a+partial object visible at it.</p>+<p>Remotes that upload atomically (e.g. object stores such as+<code>s3</code>) are streamed straight to the destination. On remotes+where a partial upload would otherwise be visible (e.g.+<code>local</code>), and whenever <code>--vfs-cache-mode</code> is+<code>writes</code> or above, the upload is written to a temporary+object that is renamed into place on success; these remotes need to+support a server-side move or copy for this (nearly all do - without+move or copy the upload is written directly and a failed PUT may leave a+partial object at the key). If <code>serve s3</code> is killed part-way+through an upload the temporary object (named with a leading+<code>.rclone_temp_put_</code>) may be left behind; it is hidden from S3+listings but must be removed manually.</p><h2 id="multipart-uploads">Multipart uploads</h2>-<p>By default <code>serve s3</code> <strong>streams</strong> each-multipart upload, in part-number order, into a single-<code>PutStream</code> upload to the underlying remote, so the whole-file is never buffered in memory - memory use stays bounded by the parts-in flight. The remote then performs its own internal upload (for example-its own multipart upload, still with bounded memory). This works for any-remote that supports <code>PutStream</code>, which is nearly all of-them, including through <code>crypt</code>.</p>-<p>The upload is atomic so the destination object only ever changes on a+<p>Multipart uploads are written, in part-number order, to a temporary+object which is renamed into place, server-side, on completion, so the+upload is atomic. The object at the key only ever changes on asuccessful completion. A failed or aborted upload never affects any-object already stored under that name. Remotes that upload atomically-already (object stores such as <code>s3</code>) are streamed straight to-the destination. On remotes where a partial upload would otherwise be-visible (such as <code>local</code>), the parts are streamed to a-temporary object that is moved into place, server-side, on completion;-these remotes therefore also need to support a server-side move or-copy.</p>+object already stored under that name and a partly-uploaded object never+becomes visible under it.</p>+<p>With the default <code>--vfs-cache-mode off</code>+<code>serve s3</code> <strong>streams</strong> each multipart upload, in+part-number order, into a single streaming upload to the underlying+remote, so the whole file is never buffered in memory: only parts which+arrive out of order are buffered, within the limits described below. The+remote then performs its own internal upload (for example its own+multipart upload, still with bounded memory). Remotes that don't support+streaming uploads (those that must know the file size before the upload+starts, such as <code>onedrive</code>, <code>pcloud</code>,+<code>jottacloud</code>, <code>mailru</code>, <code>opendrive</code>,+<code>putio</code>, <code>protondrive</code> and <code>zoho</code>) have+the parts spooled to a temporary file on <strong>local disk</strong>+instead, and uploaded with the size then known on completion, so they+need local disk space for the largest objects in flight rather than+memory.</p>+<p>With <code>--vfs-cache-mode writes</code> (or <code>full</code>) the+parts are written to a temporary file in the VFS cache and uploaded by+the VFS write-back - see <a+href="#multipart-uploads-and-the-vfs-cache">Multipart uploads and the+VFS cache</a> below.</p>+<p>The rename into place needs the remote to support a server-side move+or copy, which nearly all do. It is a cheap rename on most remotes, but+on object stores without a real rename (such as <code>s3</code> itself)+the move is performed as a server-side copy and delete of the whole+object, which can take time and API calls for large objects. Concurrent+multipart uploads of the same key (which S3 permits) are safe. Each+writes its own temporary object and the last to complete wins.</p>+<p>On the few remotes that support neither server side move nor copy,+the parts are written straight to the destination object instead and+never buffered in memory. This is at some cost in atomicity - the+incomplete object is visible under its final name while the upload is in+flight, as it also is for a plain object PUT on such remotes, and+concurrent multipart uploads of the same key write to the same object+and can interleave. A failed or aborted upload still leaves any+pre-existing object untouched provided the remote uploads atomically and+the VFS cache is off; on a remote where partial uploads are visible it+may leave partial data at the key (like a plain PUT there), and with+<code>--vfs-cache-mode writes</code> (or <code>full</code>) a write to+the cache cannot be abandoned, so an aborted upload's partial data is+written back to the remote as if it had completed.</p><p><strong>Features</strong></p><ul>-<li>The whole object is never buffered in memory; memory use is bounded-by the parts in flight, not the upload size.</li>+<li>The whole object is never buffered in memory; only parts arriving+out of order are, within fixed limits, whatever the upload size.</li><li>Parts can be any size. Clients that don't produce uniform-sizedparts work fine - for example PostgreSQL backup tools such as<strong>pgBarman</strong> and <strong>pgBackRest</strong>, which flush@@ -12073,123 +12726,547 @@ chunk size plus a variable overshoot.</li>is encrypted as one continuous stream.</li><li>The destination object only ever changes atomically, on completion:an aborted or failed upload leaves any pre-existing object of the same-name untouched, and a partly-uploaded object never becomes visible.</li>-<li>Backend-agnostic - it only needs the remote to support-<code>PutStream</code> (plus a server-side move or copy on remotes that-don't upload atomically).</li>+name untouched, and a partly-uploaded object never becomes visible+(except on the few remotes with no server-side move or copy, as+above).</li>+<li>Multipart uploads go through the VFS like any other upload, so they+show in rclone's transfer stats and obey <code>--bwlimit</code>.</li>+<li>Backend-agnostic - it only needs the remote to support a server-side+move or copy for the rename into place, which nearly all do; a remote+without streaming upload support spools to local disk as above.</li></ul><p><strong>Limitations</strong></p><ul><li>Parts must arrive in ascending, contiguous part-number order (1, 2,-3, ...). Parts the client uploads concurrently or out of order are-buffered until their turn. The memory used for this buffering is capped,-per upload, by <code>--multipart-streaming-buffer-limit</code> (default-<code>256M</code>, <code>0</code> for no limit): a part that would take-the buffer over the limit is stalled until the stream drains, so a-client that uploads faster than the remote can accept sees backpressure-rather than unbounded server memory use. Since a stalled part holds its-HTTP request open, clients whose upload concurrency times chunk size-exceeds the limit may need a longer read timeout when the remote is-slow. Non-contiguous part numbers are rejected on completion. Configure-the client to upload in part order, ideally with low concurrency, for-the lowest memory use.</li>+3, ...). The part the stream needs next is written straight through as+it arrives, so it is never buffered, whatever its size. Parts the client+uploads concurrently or out of order are buffered until their turn. The+memory used for this buffering is capped, per upload, by+<code>--multipart-streaming-buffer-limit</code> (default+<code>256M</code>, <code>0</code> for no limit), and across all the+uploads of each user (each access key) by+<code>--multipart-streaming-buffer-total</code> (default+<code>1G</code>, <code>0</code> for no limit), counting each buffered+part in the whole 1 MiB memory pages it occupies: a part that would take+the buffer over either limit is stalled until the stream drains (or, if+it is bigger than the limits, until it is the next part), so a client+that uploads faster than the remote can accept sees backpressure rather+than unbounded server memory use. A stalled part holds its HTTP request+open, and one stalled for more than a minute is failed with a+<code>SlowDown</code> error, which S3 clients retry. Clients whose+upload concurrency times part size exceeds the limits will see these+stalls and retries when the remote is slower than the client.+Non-contiguous part numbers are rejected on completion. Configure the+client to upload in part order, ideally with low concurrency, for the+lowest memory use.</li><li>A part uploaded again before completion - typically a clientretrying after a timeout - is accepted: if the earlier copy is stillbuffered it is replaced, and if it has already been streamed an-identical re-upload is a no-op. What isn't possible is replacing a part-that has already been streamed with <em>different</em> content - that is-rejected. A failure in the stream to the remote itself still aborts the-whole upload and the client must start it again. (The remote's own-upload still retries its internal chunks.)</li>+identical re-upload is a no-op (a re-upload which arrives while the part+is being streamed waits for it to finish first). If a part fails part+way through being streamed (for example the client's connection drops)+the client's retry of it carries on where it stopped, provided it has+the same content. What isn't possible is replacing a part that has+already been streamed, in whole or in part, with <em>different</em>+content - that is rejected. A part which arrives in full but is then+rejected, for example because it doesn't match the+<code>Content-MD5</code> the client declared, has already been streamed,+so the whole upload is failed and the client must start it again. So is+a failure in the stream to the remote itself. (The remote's own upload+still retries its internal chunks.)</li>+<li>A <code>CompleteMultipartUpload</code> which fails because the parts+it lists don't match those uploaded, or there is a gap in the part+numbers, leaves the upload open, as on S3, so the client can upload any+missing parts and complete it again. If writing the completed object to+the remote fails, the upload is failed and must be started again.</li><li>Parts are serialised into one stream, so ingest from the client is-effectively single-threaded, although the remote's own upload still runs-concurrently.</li>-<li>On remotes that don't upload atomically (such as-<code>local</code>), the completed object is moved into place with a-server-side operation. This is a cheap rename on most such remotes. On-these remotes, if <code>serve s3</code> is killed part-way through an-upload the temporary object (named with a leading-<code>.rclone_multipart_upload_</code>) may be left behind; it is hidden+effectively single-threaded. When streaming, the remote's own upload+runs concurrently with the parts arriving; with the local disk spool or+the VFS cache the upload to the remote only starts on completion.</li>+<li>If <code>serve s3</code> is killed part-way through an upload the+temporary object (named with a leading+<code>.rclone_temp_multipart_</code>) may be left behind; it is hiddenfrom S3 listings but must be removed manually.</li></ul>+<h3 id="multipart-uploads-and-the-vfs-cache">Multipart uploads and the+VFS cache</h3>+<p>With <code>--vfs-cache-mode writes</code> (or <code>full</code>)+multipart uploads do not stream to the remote at all. The parts are+written, in part-number order, to a temporary file in the VFS cache. On+completion the file is renamed into place and uploaded by the VFS+write-back, exactly like a plain object PUT. This needs no streaming+upload support from the remote. The rename normally happens in the cache+before the upload has started, but the VFS requires the remote to+support a server-side move or copy to rename files at all (and uses one+if the temporary file has already been written back, e.g. with+<code>--vfs-write-back 0</code>). On remotes without either, the parts+are written to the cache directly under the final key instead: the+upload still never touches memory, but it loses its atomicity - the+in-flight upload is visible at the key, and an aborted upload cannot be+abandoned once in the cache, so its partial data is written back to the+remote as if it were a completed object.</p>+<p>Remotes that benefit from <code>--vfs-cache-mode writes</code>:</p>+<ul>+<li><strong>Remotes over slow or unreliable links.</strong> A failure in+a streamed upload aborts the whole multipart upload and the client must+start again from the first part; a failed write-back upload is retried+by the VFS (see <code>--vfs-cache-max-age</code> and friends) without+the client being involved. Ingest from the client also runs at local+disk speed rather than being throttled to the remote's pace.</li>+<li><strong>Workloads that read back or overwrite what they just+wrote.</strong> The completed object stays in the cache, so subsequent+<code>GET</code>/<code>HEAD</code> requests are served locally, and+plain PUTs and multipart uploads to the same key go through the same+cache entry so the last write wins regardless of upload style.</li>+</ul>+<p>The trade-offs of the VFS cache:</p>+<ul>+<li>The whole object lands on local disk, so the cache+(<code>--cache-dir</code>) needs space for the largest objects in+flight; <code>--vfs-cache-max-size</code> cannot evict files which are+still being uploaded.</li>+<li>The <code>200 OK</code> for <code>CompleteMultipartUpload</code>+means the data is safely in the <strong>local cache</strong>, not yet on+the remote - the same durability the cache gives plain PUTs. If an+acknowledgement must mean the data has reached the remote (for example+WAL archiving), use the default <code>--vfs-cache-mode off</code>.</li>+<li>The upload to the remote only starts on completion, rather than+overlapping with the parts arriving, so the data reaches the remote+later than with streaming.</li>+<li>If <code>serve s3</code> is killed part-way through an upload, the+temporary file survives in the cache and the VFS cache recovery uploads+it to the remote on restart as a temporary object (named with a leading+<code>.rclone_temp_multipart_</code>); as with the streaming path, it is+hidden from S3 listings but must be removed manually.</li>+</ul>+<h3 id="cleaning-up-temporary-objects">Cleaning up temporary+objects</h3>+<p>If <code>serve s3</code> is killed part-way through an upload it can+leave a temporary object behind, named with a leading+<code>.rclone_temp_</code>. This whole prefix is reserved: any object+whose name (the last <code>/</code>-separated segment of its key) starts+with <code>.rclone_temp_</code> is hidden from S3 listings, so don't+give real objects such names - an existing object with such a name+disappears from listings (though it stays accessible directly by its+key: only listings hide reserved names, <code>GET</code>,+<code>HEAD</code> and <code>DELETE</code> of the exact key still work).+A temporary object never holds acknowledged data - uploads whose+temporary object survived were never confirmed to the client - so old+ones are safe to delete:</p>+<pre><code>rclone delete --min-age 24h --include ".rclone_temp_*" remote:path</code></pre>+<p>The <code>--min-age</code> protects uploads which are still in+progress: make sure it is longer than your longest upload, especially if+several <code>serve s3</code> instances share the same remote.</p>+<p>rclone v1.75 named its temporary multipart objects+<code>.rclone_multipart_upload_*</code>; leftovers from an older server+are also hidden from listings and can be cleaned up the same way.</p>+<h3 id="abandoned-uploads">Abandoned uploads</h3>+<p>A client which starts a multipart upload and vanishes without either+completing or aborting it would otherwise hold on to its resources+forever.</p>+<p>An incomplete multipart upload which has had no activity for+<code>--multipart-expiry</code> (default <code>24h</code>) is therefore+aborted and cleaned up, exactly as if the client had called+<code>AbortMultipartUpload</code>, and a <code>NOTICE</code> is+logged.</p>+<p>An upload with a part still being received is never expired, however+slowly the part is arriving, and each completed part restarts the clock,+so the expiry only needs to outlast the client's pauses <em>between</em>+parts, not the whole upload.</p>+<p>Late operations on an expired upload fail with+<code>NoSuchUpload</code>, as they do on real S3 when a lifecycle rule+has aborted the upload. Set <code>--multipart-expiry 0</code> to keep+incomplete uploads forever - but note that abandoned uploads are then+never cleaned up and go on counting towards+<code>--multipart-max-uploads</code> below, so enough of them will stop+the user starting any new multipart upload until the server is+restarted.</p>+<p>Each multipart upload in progress holds server resources - for+example the buffers the remote's own upload uses once the first part has+arrived - so each user (each access key) can have no more than+<code>--multipart-max-uploads</code> (default <code>1000</code>,+<code>0</code> for no limit) in progress at once. Starting another fails+with a <code>SlowDown</code> error, which S3 clients retry, and a+one-off <code>NOTICE</code> is logged.</p><h3 id="disabling-streaming">Disabling streaming</h3>-<p>If you pass <code>--disable-multipart-streaming</code>, or the remote-doesn't support <code>PutStream</code> (or doesn't upload atomically and-can't move or copy server-side), multipart uploads are instead-<strong>buffered in memory</strong> by the underlying S3 library: every-part is held in memory and the whole object is written out in one go-when the upload completes (the previous behaviour). This removes the+<p>If you pass <code>--disable-multipart-streaming</code>, multipart+uploads are instead <strong>buffered in memory</strong> by the+underlying S3 library: every part is held in memory and the whole object+is written out in one go when the upload completes. This removes thein-order/contiguous-part restriction above, so parts can be uploaded inany order, but <strong>memory use grows with the size of theupload</strong>, so it is only suitable for small objects. A one-off-<code>NOTICE</code> is logged the first time this happens.</p>-<p>Alternatively, if the client is an rclone <code>s3</code> remote-(like the <code>[serves3]</code> example above), you can set-<code>use_multipart_uploads = false</code> on it so it uploads each-object as a single stream and skips multipart uploads altogether.</p>+<code>NOTICE</code> is logged the first time this happens. This flag is+the only thing that makes multipart uploads buffer in memory - it is+never done because of missing remote capabilities. The limits on+multipart uploads above+(<code>--multipart-streaming-buffer-limit</code>,+<code>--multipart-streaming-buffer-total</code> and+<code>--multipart-max-uploads</code>) do not apply to uploads buffered+in memory, so don't use this flag if the server is open to clients you+don't trust. Consider <code>--vfs-cache-mode writes</code> instead,+which buffers the upload in the VFS cache on disk and takes precedence+over <code>--disable-multipart-streaming</code>.</p>+<h2 id="memory-use">Memory use</h2>+<p><code>serve s3</code> never buffers whole objects in memory (except+with <code>--disable-multipart-streaming</code>, above), and limits the+memory each kind of request can use, so that a client, even a hostile+one, can't make it use unbounded memory.</p>+<ul>+<li><strong>Object data.</strong> <code>PutObject</code> bodies and+multipart parts arriving in order are streamed straight through to the+remote. Parts arriving out of order are buffered, up to+<code>--multipart-streaming-buffer-limit</code> per upload and+<code>--multipart-streaming-buffer-total</code> per user.</li>+<li><strong>Uploads in progress.</strong> Each user can have at most+<code>--multipart-max-uploads</code> multipart uploads in progress. An+upload holds little memory until its first part arrives. After that,+with <code>--vfs-cache-mode off</code>, it holds a+<code>--streaming-upload-cutoff</code> sized buffer plus whatever the+remote's own upload buffers, which for many remotes is up to their chunk+size times their upload concurrency (for example+<code>--s3-chunk-size</code> times+<code>--s3-upload-concurrency</code>). Plain <code>PutObject</code>+uploads use the same while they are in progress.</li>+<li><strong>Metadata.</strong> The metadata of at most+<code>--metadata-max-objects</code> objects of each user is kept, each+limited to 2 KB.</li>+<li><strong>Request bodies.</strong> The XML bodies of requests such as+<code>CompleteMultipartUpload</code> and <code>DeleteObjects</code> are+limited to 10 MiB, and <code>DeleteObjects</code> to 1000 keys. A+browser form (<code>POST</code>) upload holds at most 1 MiB in memory,+spooling the rest to a temporary file, and is limited to 5 GB.</li>+<li><strong>Listings.</strong> Each page of a <code>ListObjects</code>+listing reads only the directories needed to fill it, and returns at+most 1000 keys. Directory listings are cached by the <a+href="#vfs-directory-cache">VFS directory cache</a>.</li>+</ul>+<p>The largest of these is usually the remote's upload buffers, which+can be up to <code>--multipart-max-uploads</code> times the buffers of+one upload for each user if clients start many uploads and send part of+each. The limits apply to each user separately, so that one user can't+stop others uploading, which means the memory use grows with the number+of users. To limit it, lower <code>--multipart-max-uploads</code>, the+remote's chunk size or upload concurrency, or set+<code>--max-buffer-memory</code> to limit the memory used by the buffers+of all uploads to remotes which use rclone's buffer pool (including+<code>s3</code>, <code>azureblob</code> and <code>b2</code>). Buffered+multipart parts come from the same pool, so if you set+<code>--max-buffer-memory</code> make it comfortably bigger than+<code>--multipart-streaming-buffer-total</code> times the number of+users uploading at once, otherwise buffered parts can use up all the+memory the remote needs to upload the parts which would free them.</p>+<p>The number of connections is not limited, so if the server is open to+clients you don't trust, use <code>--auth-key</code> or+<code>--auth-proxy</code> and consider putting it behind a reverse proxy+which limits connections.</p>+<h2 id="design-goals">Design goals</h2>+<p>These goals explain many of the limitations below.</p>+<ul>+<li><strong>Objects are plain files on the remote.</strong> A bucket is+a directory in the root and a key is a path in it, so the remote stays+usable by other tools, such as <code>rclone mount</code> or+<code>rclone sync</code>. So keys must be paths (see <a+href="#object-names">Object names</a>).</li>+<li><strong>No database.</strong> <code>serve s3</code> stores nothing+but the objects. What the remote can't hold, such as user metadata,+tags, the MD5 <code>ETag</code> of an object on a remote without MD5 and+the directory markers put, is kept in memory as a best effort, and lost+on restart.</li>+<li><strong>Works with any rclone remote.</strong> What+<code>serve s3</code> can do depends on what the remote supports, such+as the hashes used for <code>ETag</code>s and modification times, rather+than needing a particular backend.</li>+<li><strong>Bounded memory.</strong> Uploads, including multipart+uploads, are streamed to the remote and the memory each request can use+is limited (see <a href="#memory-use">Memory use</a>). Nothing is+written to local disk unless the VFS cache is used or the remote can't+stream uploads (see <a href="#multipart-uploads">Multipart+uploads</a>).</li>+<li><strong>Refuse rather than ignore.</strong> Unsupported features are+refused, or answered as S3 answers when the feature isn't configured, so+a client is never told something was done which wasn't (see <a+href="#compatibility">Compatibility</a>).</li>+<li><strong>A storage gateway, not a full S3.</strong>+<code>serve s3</code> implements the parts of the S3 API clients use to+store and fetch objects. Versioning, ACLs, bucket policies, lifecycle+rules and the like are out of scope.</li>+<li><strong>Compatible with rclone's s3 backend.</strong> This includes+its directory markers, which the tests check.</li>+</ul><h2 id="bugs">Bugs</h2>-<p>Multipart server side copies do not work (see <a-href="https://github.com/rclone/rclone/issues/7454">#7454</a>). These-take a very long time and eventually fail. The default threshold for-multipart server side copies is 5G which is the maximum it can be, so-files above this side will fail to be server side copied.</p><p>For a current list of <code>serve s3</code> bugs see the <ahref="https://github.com/rclone/rclone/labels/serve%20s3">serve s3</a>bug category on GitHub.</p><h2 id="limitations-2">Limitations</h2><p><code>serve s3</code> will treat all directories in the root asbuckets and ignore all files in the root. You can use-<code>CreateBucket</code> to create folders under the root, but you-can't create empty folders under other folders not in the root.</p>+<code>CreateBucket</code> to create folders under the root, and put a+directory marker to create an empty folder below them (see <a+href="#object-names">Object names</a>).</p><p>When using <code>PutObject</code> or <code>DeleteObject</code>,rclone will automatically create or clean up empty folders. If you don'twant to clean up empty folders automatically, use-<code>--no-cleanup</code>.</p>-<p>When using <code>ListObjects</code>, rclone will use <code>/</code>-when the delimiter is empty. This reduces backend requests with no-effect on most operations, but if the delimiter is something other than-<code>/</code> and empty, rclone will do a full recursive search of the-backend, which can take some time.</p>-<p>Versioning is not currently supported.</p>-<p>Metadata will only be saved in memory other than the rclone+<code>--directory-markers</code> (see <a href="#object-names">Object+names</a>). The deprecated <code>--no-cleanup</code> does the same.</p>+<p>When using <code>ListObjects</code> with the delimiter+<code>/</code>, rclone lists the directory the prefix is in, showing its+subdirectories as common prefixes, including empty ones. Without a+delimiter the listing is recursive: rclone walks the directory tree+below the prefix, reading only as many directories as it needs to fill+each page, and lists empty directories as their directory markers (see+below). Any other delimiter is supported too, but needs the whole+directory tree below the prefix walking, unless the delimiter is in the+directory names.</p>+<p>Versioning is not supported (see <a href="#design-goals">Design+goals</a>).</p>+<p>Conditional writes, <code>PutObject</code> and+<code>CompleteMultipartUpload</code> with <code>If-None-Match</code> or+<code>If-Match</code>, are supported. The conditions are checked before+the object is written, and again as it is moved into place when it is+written to a temporary object first. This happens for multipart uploads+and, for <code>PutObject</code>, with remotes which can upload partial+files (such as local disk) or with <code>--vfs-cache-mode minimal</code>+or above. The second check is atomic against other writes through+<code>serve s3</code>, but neither is atomic against changes made to the+remote some other way. If the remote doesn't support the+<code>--etag-hash</code>, objects have no <code>ETag</code> so+<code>If-Match</code> with an <code>ETag</code> always fails.</p>+<p>Metadata will only be saved in memory (see <a+href="#design-goals">Design goals</a>) other than the rclone<code>mtime</code> metadata which will be set as the modification time-of the file.</p>+of the file. The metadata of at most <code>--metadata-max-objects</code>+objects (default <code>100000</code>, <code>0</code> for no limit) of+each user (each access key) is kept; after that the metadata of the+user's least recently used object is forgotten. The metadata of an+object is ignored once the object has been changed some other way, for+example by another <code>--auth-proxy</code> user or directly on the+backend. Object tags are kept with the metadata in the same way.</p><h2 id="object-names">Object names</h2><p><code>serve s3</code> stores objects as files in the backend, soobject keys are mapped to file paths rather than treated as the opaquestrings AWS S3 allows. Keys must be in canonical path form: keys thatcontain <code>..</code> or <code>.</code> path segments, repeated-slashes (<code>//</code>), or a leading or trailing slash are rejected-with a <code>400 Bad Request</code> (<code>InvalidArgument</code>)-instead of being normalised, since normalising them could alias two-distinct keys to the same file or resolve a key outside its bucket. This-matches the behaviour of other S3 servers such as MinIO.</p>-<h2 id="supported-operations">Supported operations</h2>-<p><code>serve s3</code> currently supports the following-operations.</p>-<ul>-<li>Bucket-<ul>-<li><code>ListBuckets</code></li>-<li><code>CreateBucket</code></li>-<li><code>DeleteBucket</code></li>-</ul></li>-<li>Object-<ul>-<li><code>HeadObject</code></li>-<li><code>ListObjects</code></li>-<li><code>GetObject</code></li>-<li><code>PutObject</code></li>-<li><code>DeleteObject</code></li>-<li><code>DeleteObjects</code></li>-<li><code>CreateMultipartUpload</code></li>-<li><code>CompleteMultipartUpload</code></li>-<li><code>AbortMultipartUpload</code></li>-<li><code>CopyObject</code></li>-<li><code>UploadPart</code></li>-</ul></li>-</ul>-<p>Other operations will return error <code>Unimplemented</code>.</p>+slashes (<code>//</code>), or a leading slash are rejected with a+<code>400 Bad Request</code> (<code>InvalidArgument</code>) instead of+being normalised, since normalising them could alias two distinct keys+to the same file or resolve a key outside its bucket. This matches the+behaviour of other S3 servers such as MinIO.</p>+<p>A key ending in <code>/</code>, such as <code>dir/</code>, is a+directory marker, standing for the directory <code>dir</code>. Putting+an empty directory marker creates the directory (one with data is+rejected with <code>400 Bad Request</code>), and deleting it removes the+directory if it is empty. Directory markers are found by+<code>HeadObject</code>, <code>GetObject</code> and+<code>CopyObject</code> for any directory, but only the markers of empty+directories, and of the directories a marker was put for, are listed as+objects, as other directories may not have been made by putting a+marker. The markers put are remembered in memory with the metadata (see+<a href="#limitations">Limitations</a>), so after a restart, or once+forgotten, the marker of a directory holding other keys is no longer+listed. While a marker is remembered its directory isn't removed when+the last key below it is deleted.</p>+<p>With <code>--directory-markers</code> every directory is listed as a+directory marker, so the markers are listed after a restart too, as+needed by clients which rely on them, such as rclone's s3 backend with+<code>--s3-directory-markers</code>. As every directory has a marker, a+directory isn't removed when the last key below it is deleted. Deleting+the marker of a directory which holds other keys stops it being listed,+and the directory is then removed with the last key below it, but this+is only remembered in memory, like the markers put without+<code>--directory-markers</code>.</p>+<p>As keys are file paths, a key can't be both an object and a directory+holding other objects, for example <code>a/b</code> and+<code>a/b/c</code>. Writing an object which would need this fails with+<code>400 Bad Request</code> (<code>InvalidRequest</code>).</p>+<h2 id="compatibility">Compatibility</h2>+<p><code>serve s3</code> implements the parts of the S3 API that S3+clients commonly use to store and fetch objects. Features it doesn't+support are refused with <code>501 NotImplemented</code>, or answered as+AWS S3 answers for a bucket without that feature configured, rather than+being silently ignored.</p>+<table>+<colgroup>+<col style="width: 33%" />+<col style="width: 42%" />+<col style="width: 23%" />+</colgroup>+<thead>+<tr class="header">+<th>Feature</th>+<th>Supported</th>+<th>Notes</th>+</tr>+</thead>+<tbody>+<tr class="odd">+<td><code>ListBuckets</code>, <code>CreateBucket</code>,+<code>DeleteBucket</code>, <code>HeadBucket</code>,+<code>GetBucketLocation</code></td>+<td>Yes</td>+<td>Buckets are the directories in the root of the remote.</td>+</tr>+<tr class="even">+<td><code>ListObjects</code>, <code>ListObjectsV2</code></td>+<td>Yes</td>+<td>Prefixes, delimiters, markers, <code>start-after</code>,+continuation tokens, <code>fetch-owner</code> and+<code>encoding-type=url</code> all work. Delimiters other than+<code>/</code> need the whole directory tree walking.</td>+</tr>+<tr class="odd">+<td><code>HeadObject</code>, <code>GetObject</code></td>+<td>Yes</td>+<td>Including <code>Range</code>, <code>partNumber</code>, the+<code>response-*</code> header overrides and the <code>If-Match</code>,+<code>If-None-Match</code>, <code>If-Modified-Since</code> and+<code>If-Unmodified-Since</code> conditions.</td>+</tr>+<tr class="even">+<td><code>PutObject</code></td>+<td>Yes</td>+<td>Including <code>Content-MD5</code>, <code>aws-chunked</code> uploads+and uploads with no <code>Content-Length</code>.</td>+</tr>+<tr class="odd">+<td><code>CopyObject</code></td>+<td>Yes</td>+<td>Including the metadata and tagging directives and the+<code>x-amz-copy-source-if-*</code> conditions.</td>+</tr>+<tr class="even">+<td><code>DeleteObject</code>, <code>DeleteObjects</code></td>+<td>Yes</td>+<td></td>+</tr>+<tr class="odd">+<td>Multipart uploads</td>+<td>Yes</td>+<td><code>CreateMultipartUpload</code>, <code>UploadPart</code>,+<code>UploadPartCopy</code>, <code>CompleteMultipartUpload</code>,+<code>AbortMultipartUpload</code>, <code>ListMultipartUploads</code> and+<code>ListParts</code>. The <code>ETag</code> of the object is its hash,+not the <code>-N</code> suffixed one AWS gives. The 5 MiB minimum part+size isn't enforced.</td>+</tr>+<tr class="even">+<td>Conditional writes</td>+<td>Partial</td>+<td><code>If-Match</code> and <code>If-None-Match</code> on+<code>PutObject</code> and <code>CompleteMultipartUpload</code> are+checked, but not atomically against changes made to the remote some+other way (see <a href="#limitations">Limitations</a>).</td>+</tr>+<tr class="odd">+<td>Authentication</td>+<td>Partial</td>+<td>AWS Signature Version 4, in headers, presigned URLs and browser form+(<code>POST</code>) uploads. Signature Version 2 isn't supported.</td>+</tr>+<tr class="even">+<td>Anonymous access</td>+<td>Partial</td>+<td>All requests are allowed without <code>--auth-key</code> or+<code>--auth-proxy</code>, and none are with them, as buckets and+objects can't be made public.</td>+</tr>+<tr class="odd">+<td>Metadata</td>+<td>Partial</td>+<td><code>x-amz-meta-*</code>, the storage class and the standard+headers such as <code>Content-Type</code> are kept in memory only (see+<a href="#limitations">Limitations</a>), except the modification time.+The storage class is only recorded, so doesn't change how the object is+stored.</td>+</tr>+<tr class="even">+<td>Directory markers</td>+<td>Partial</td>+<td>Keys ending in <code>/</code> are stored as directories, so can't+hold data. A marker put for a directory holding other keys is only+listed while it is remembered in memory, unless+<code>--directory-markers</code> is set (see <a+href="#object-names">Object names</a>).</td>+</tr>+<tr class="odd">+<td>Checksums</td>+<td>Partial</td>+<td>The <code>x-amz-checksum-*</code> of a <code>PutObject</code>+(CRC32, CRC32C, CRC64NVME, SHA1 or SHA256) is checked and returned with+the metadata. The <code>Content-MD5</code> and+<code>x-amz-checksum-*</code> of the bodies of other requests, such as+<code>DeleteObjects</code>, are checked. The checksums of multipart+uploads aren't supported.</td>+</tr>+<tr class="even">+<td><code>GetObjectAttributes</code></td>+<td>Partial</td>+<td>Everything but the checksums of multipart uploads.</td>+</tr>+<tr class="odd">+<td>Object tagging</td>+<td>Partial</td>+<td><code>GetObjectTagging</code>, <code>PutObjectTagging</code> and+<code>DeleteObjectTagging</code> work, but tags are kept in memory with+the metadata.</td>+</tr>+<tr class="even">+<td>Bucket owner</td>+<td>No</td>+<td><code>serve s3</code> has no AWS account IDs, so+<code>x-amz-expected-bucket-owner</code> and+<code>x-amz-source-expected-bucket-owner</code> are ignored.</td>+</tr>+<tr class="odd">+<td>ACLs</td>+<td>Partial</td>+<td><code>GetBucketAcl</code> and <code>GetObjectAcl</code> give the+owner full control. Only the <code>private</code> ACL can be set.</td>+</tr>+<tr class="even">+<td>CORS</td>+<td>Partial</td>+<td>Preflight (<code>OPTIONS</code>) requests are answered, allowing any+origin. CORS configuration (<code>PutBucketCors</code>) isn't+supported.</td>+</tr>+<tr class="odd">+<td>Server side encryption</td>+<td>No</td>+<td>SSE-S3 (<code>AES256</code>) is accepted but the data isn't+encrypted. SSE-KMS and SSE-C are refused.</td>+</tr>+<tr class="even">+<td>Versioning</td>+<td>No</td>+<td>Buckets are reported as unversioned and+<code>ListObjectVersions</code> lists each object as version+<code>null</code>. Enabling versioning is refused.</td>+</tr>+<tr class="odd">+<td>Object lock, retention and legal hold</td>+<td>No</td>+<td></td>+</tr>+<tr class="even">+<td>Bucket configuration</td>+<td>No</td>+<td>Bucket policies, tagging, lifecycle, logging, encryption, website,+replication, notifications and the like. Getting them fails as for a+bucket without them configured.</td>+</tr>+<tr class="odd">+<td><code>SelectObjectContent</code>, <code>RestoreObject</code>,+torrents</td>+<td>No</td>+<td></td>+</tr>+</tbody>+</table>+<p>The <a href="https://github.com/ceph/s3-tests">ceph s3-tests</a> can+be run against <code>serve s3</code> to check its compatibility. See <a+href="https://github.com/rclone/rclone/tree/master/cmd/serve/s3/s3tests">cmd/serve/s3/s3tests</a>+for how.</p><h3 id="authentication-4">Authentication</h3><p>By default this will serve files without needing a login.</p><p>You can either use an htpasswd file which can take lots of users, or@@ -12212,7 +13289,17 @@ BCrypt for basic authentication. Bcrypt is recommended.</p><pre class="console"><code>touch htpasswdhtpasswd -B htpasswd userhtpasswd -B htpasswd anotherUser</code></pre>+<p>Use the same hash format and cost for every entry in the file. If the+entries differ, the time taken to reject a wrong password can reveal+which usernames exist.</p><p>The password file can be updated while rclone is running.</p>+<p>When authentication is enabled every request must be authenticated,+including HTTP OPTIONS requests, as the answer to an OPTIONS request can+reveal whether a path exists. Health checks or monitoring which use+unauthenticated OPTIONS requests will need to send credentials. The+exception is browser CORS preflight requests when+<code>--allow-origin</code> is set, which are answered without+authentication as browsers never send credentials with them.</p><p>Use <code>--realm</code> to set the authentication realm.</p><p>Use <code>--salt</code> to change the password hashing salt from thedefault.</p>@@ -12240,6 +13327,14 @@ the server will accept in the HTTP header.</p>all responses, will overriding existing values. The flag may be repeatedto add multiple headers. Use the format<code>Header-Name: value</code>.</p>+<p><code>--allow-origin</code> enables cross-origin (CORS) requests from+browsers by setting the <code>Access-Control-Allow-Origin</code> header+on every response to the origin given, e.g.+<code>--allow-origin https://app.example.com</code>. Browser CORS+preflight OPTIONS requests are answered by rclone without needing+authentication as browsers never send credentials with them. Setting+this to <code>*</code> allows any origin which is a security risk on an+authenticated server.</p><p><code>--baseurl</code> controls the URL prefix that rclone servesfrom. By default rclone will serve from the root. If you used<code>--baseurl "/rclone"</code> then rclone would serve from a URL@@ -12326,7 +13421,7 @@ that will be used to buffer data in advance.</p>memory at all times. The buffered data is bound to one open file andwon't be shared.</p><p>This flag is a upper limit for the used memory per open file. The-buffer will only use memory for data that is downloaded but not not yet+buffer will only use memory for data that is downloaded but not yetread. If the buffer is empty, only a small amount of memory will beused.</p><p>The maximum memory used by rclone for buffering can be up to@@ -12369,11 +13464,12 @@ will start with files that haven't been accessed for the longest. Thiscache flushing strategy is efficient and more relevant files are likelyto remain cached.</p><p>The <code>--vfs-cache-max-age</code> will evict files from the cache-after the set time since last access has passed. The default value of 1-hour will start evicting files from cache that haven't been accessed for-1 hour. When a cached file is accessed the 1 hour timer is reset to 0-and will wait for 1 more hour before evicting. Specify the time with-standard notation, s, m, h, d, w .</p>+after the set time since last access has passed; it is based on access+time, not on when the file was first added to the cache. The default+value of 1 hour will start evicting files from cache that haven't been+accessed for 1 hour. When a cached file is accessed the 1 hour timer is+reset to 0 and will wait for 1 more hour before evicting. Specify the+time with standard notation, s, m, h, d, w .</p><p>You <strong>should not</strong> run two copies of rclone using thesame VFS cache with the same or overlapping remotes if using<code>--vfs-cache-mode > off</code>. This can potentially cause data@@ -12499,8 +13595,8 @@ id="vfs-read-chunk-streams-0-15"><code>--vfs-read-chunk-streams</code><p>Rclone reads <code>--vfs-read-chunk-streams</code> chunks of size<code>--vfs-read-chunk-size</code> concurrently. The size for each readwill stay constant.</p>-<p>This improves performance performance massively on high latency links-or very high bandwidth links to high performance object stores.</p>+<p>This improves performance massively on high latency links or very+high bandwidth links to high performance object stores.</p><p>Some experimentation will be needed to find the optimum values of<code>--vfs-read-chunk-size</code> and<code>--vfs-read-chunk-streams</code> as these will depend on the@@ -12679,8 +13775,121 @@ total 1048578<p>If the file has no metadata it will be returned as <code>{}</code>and if there is an error reading the metadata the error will be returnedas <code>{"error":"error string"}</code>.</p>+<h2 id="auth-proxy-2">Auth Proxy</h2>+<p>If you supply the parameter+<code>--auth-proxy /path/to/program</code> then rclone will use that+program to generate backends on the fly which then are used to+authenticate incoming requests. This uses a simple JSON based protocol+with input on STDIN and output on STDOUT.</p>+<p><strong>PLEASE NOTE:</strong> <code>--auth-proxy</code> and+<code>--authorized-keys</code> cannot be used together, if+<code>--auth-proxy</code> is set the authorized keys option will be+ignored.</p>+<p>There is an example program <a+href="https://github.com/rclone/rclone/blob/master/bin/test_proxy.py">bin/test_proxy.py</a>+in the rclone source code.</p>+<p>The program's job is to take a <code>user</code> and+<code>pass</code> on the input and turn those into the config for a+backend on STDOUT in JSON format. This config will have any default+parameters for the backend added, but it won't use configuration from+environment variables or command line options - it is the job of the+proxy program to make a complete config.</p>+<p>This config generated must have this extra parameter</p>+<ul>+<li><code>_root</code> - root to use for the backend</li>+</ul>+<p>And it may have these parameters</p>+<ul>+<li><code>_obscure</code> - comma separated strings for parameters to+obscure</li>+<li><code>_secret_access_key</code> - the secret for S3 access key auth+(see below)</li>+</ul>+<p>If password authentication was used by the client, input to the proxy+process (on STDIN) would look similar to this:</p>+<div class="sourceCode" id="cb597"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb597-1"><a href="#cb597-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb597-2"><a href="#cb597-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"user"</span><span class="fu">:</span> <span class="st">"me"</span><span class="fu">,</span></span>+<span id="cb597-3"><a href="#cb597-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"pass"</span><span class="fu">:</span> <span class="st">"mypassword"</span><span class="fu">,</span></span>+<span id="cb597-4"><a href="#cb597-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"client_ip"</span><span class="fu">:</span> <span class="st">"192.168.1.1"</span></span>+<span id="cb597-5"><a href="#cb597-5" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<p>If public-key authentication was used by the client, input to the+proxy process (on STDIN) would look similar to this:</p>+<div class="sourceCode" id="cb598"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb598-1"><a href="#cb598-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb598-2"><a href="#cb598-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"user"</span><span class="fu">:</span> <span class="st">"me"</span><span class="fu">,</span></span>+<span id="cb598-3"><a href="#cb598-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"public_key"</span><span class="fu">:</span> <span class="st">"AAAAB3NzaC1yc2EAAAADAQABAAABAQDuwESFdAe14hVS6omeyX7edc...JQdf"</span><span class="fu">,</span></span>+<span id="cb598-4"><a href="#cb598-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"client_ip"</span><span class="fu">:</span> <span class="st">"192.168.1.1"</span></span>+<span id="cb598-5"><a href="#cb598-5" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<p>If the client authenticated with an S3 access key+(<code>rclone serve s3</code>), the client never sends its secret, only+a signature made with it, so the input contains just the access key ID+as the <code>user</code> with no <code>pass</code> or+<code>public_key</code>:</p>+<div class="sourceCode" id="cb599"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb599-1"><a href="#cb599-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb599-2"><a href="#cb599-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"user"</span><span class="fu">:</span> <span class="st">"AKIAIOSFODNN7EXAMPLE"</span><span class="fu">,</span></span>+<span id="cb599-3"><a href="#cb599-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"client_ip"</span><span class="fu">:</span> <span class="st">"192.168.1.1"</span></span>+<span id="cb599-4"><a href="#cb599-4" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<p>In this case the program must look up the secret access key for that+access key ID and return it in the <code>_secret_access_key</code> field+of the output. Rclone then uses that secret to verify the signature on+the request, refusing the request if it does not match. This means the+proxy program is the source of truth for both the credentials and the+backend they map to. If the program does not return+<code>_secret_access_key</code> or returns it empty the request is+refused.</p>+<p>The program's answer for an access key ID is cached (see below) but+is checked with the program again after 5 minutes even if the access key+ID is in constant use, so revoking an access key ID in the program takes+effect within 5 minutes. A rotated secret takes effect on the first+request signed with it.</p>+<p>The <code>client_ip</code> key holds the IP address the client+connected from, without a port number. It can be used to restrict logins+to certain networks, or to log authentication attempts centrally. It is+omitted if the client has no IP address, for example when connecting+over a unix socket. Note that if rclone is behind a reverse proxy this+will be the address of the reverse proxy and not the original+client.</p>+<p>And as an example return this on STDOUT</p>+<div class="sourceCode" id="cb600"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb600-1"><a href="#cb600-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb600-2"><a href="#cb600-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"type"</span><span class="fu">:</span> <span class="st">"sftp"</span><span class="fu">,</span></span>+<span id="cb600-3"><a href="#cb600-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"_root"</span><span class="fu">:</span> <span class="st">""</span><span class="fu">,</span></span>+<span id="cb600-4"><a href="#cb600-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"_obscure"</span><span class="fu">:</span> <span class="st">"pass"</span><span class="fu">,</span></span>+<span id="cb600-5"><a href="#cb600-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"user"</span><span class="fu">:</span> <span class="st">"me"</span><span class="fu">,</span></span>+<span id="cb600-6"><a href="#cb600-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">"pass"</span><span class="fu">:</span> <span class="st">"mypassword"</span><span class="fu">,</span></span>+<span id="cb600-7"><a href="#cb600-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"host"</span><span class="fu">:</span> <span class="st">"sftp.example.com"</span></span>+<span id="cb600-8"><a href="#cb600-8" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<p>This would mean that an SFTP backend would be created on the fly for+the <code>user</code> and <code>pass</code>/<code>public_key</code>+returned in the output to the host given. Note that since+<code>_obscure</code> is set to <code>pass</code>, rclone will obscure+the <code>pass</code> parameter before creating the backend (which is+required for sftp backends).</p>+<p>The program can manipulate the supplied <code>user</code> in any way,+for example to make proxy to many different sftp backends, you could+make the <code>user</code> be <code>[email protected]</code> and then set+the <code>host</code> to <code>example.com</code> in the output and the+user to <code>user</code>. For security you'd probably want to restrict+the <code>host</code> to a limited list.</p>+<p>An internal cache of backends is keyed on the <code>user</code>, the+kind of credential presented, a hash of the <code>pass</code> or+<code>public_key</code>, and the <code>client_ip</code>. This means that+if a user's password or public-key changes, the client connects from a+new IP address, or the proxy returns different config parameters (eg a+rotated <code>api_key</code>), a fresh backend will be created on the+next request rather than the cached one being reused. A backend created+for a public key login is never reused for a password login, so a public+key cannot be sent as a password to skip the program.</p>+<p>A cached answer is checked with the program again after 5 minutes+even if the user is in constant use, so revoking a user in the program+takes effect within 5 minutes. For <code>serve sftp</code> this applies+to new logins - an existing connection is not closed.</p>+<p>This can be used to build general purpose proxies to any kind of+backend that rclone supports.</p><pre><code>rclone serve s3 remote:path [flags]</code></pre>-<h2 id="options-85">Options</h2>+<h2 id="options-86">Options</h2><pre><code> --addr stringArray IPaddress:Port or :Port to bind server to (default 127.0.0.1:8080)--allow-origin string Origin which cross-domain request (CORS) can be executed from--auth-key stringArray Set key pair for v4 authorization: access_key_id,secret_access_key@@ -12690,7 +13899,8 @@ as <code>{"error":"error string"}</code>.</p>--client-ca string Client certificate authority to verify clients with--dir-cache-time Duration Time to cache directory entries for (default 5m0s)--dir-perms FileMode Directory permissions (default 777)- --disable-multipart-streaming Buffer multipart uploads in memory instead of streaming them to the backend (see the Multipart uploads docs section)+ --directory-markers List every directory as a directory marker object+ --disable-multipart-streaming Buffer multipart uploads in memory instead of streaming them to the backend--etag-hash string Which hash to use for the ETag, or auto or blank for off (default "MD5")--file-perms FileMode File permissions (default 666)--force-path-style If true use path style access if false use virtual hosted style (default true)@@ -12700,10 +13910,14 @@ as <code>{"error":"error string"}</code>.</p>--key string TLS PEM Private key--link-perms FileMode Link permissions (default 666)--max-header-bytes int Maximum size of request header (default 4096)+ --metadata-max-objects int Maximum number of objects of a user whose metadata is kept in memory, 0 for unlimited (default 100000)--min-tls-version string Minimum TLS version that is acceptable (default "tls1.0")- --multipart-streaming-buffer-limit SizeSuffix Maximum memory buffered per streamed multipart upload for parts arriving out of order, 0 for unlimited (see the Multipart uploads docs section) (default 256Mi)+ --multipart-expiry Duration Abort incomplete multipart uploads idle for longer than this, 0 to keep forever (default 1d)+ --multipart-max-uploads int Maximum number of streamed multipart uploads a user can have in progress at once, 0 for unlimited (default 1000)+ --multipart-streaming-buffer-limit SizeSuffix Maximum memory buffered per streamed multipart upload for parts arriving out of order, 0 for unlimited (default 256Mi)+ --multipart-streaming-buffer-total SizeSuffix Maximum memory buffered by all the streamed multipart uploads of a user for parts arriving out of order, 0 for unlimited (default 1Gi)--no-checksum Don't compare checksums on up/download- --no-cleanup Not to cleanup empty folder after object is deleted+ --no-cleanup Deprecated: use --directory-markers instead--no-modtime Don't read/write the modification time (can speed things up)--no-seek Don't allow seeking in files--pass string Password for authentication@@ -12742,7 +13956,7 @@ as <code>{"error":"error string"}</code>.</p><p>Options shared with other commands are described next. See the <ahref="https://rclone.org/flags/">global flags page</a> for globaloptions not listed here.</p>-<h3 id="filter-options-29">Filter Options</h3>+<h3 id="filter-options-30">Filter Options</h3><p>Flags for filtering directory listings</p><pre class="text"><code> --delete-excluded Delete files on dest excluded from sync--exclude stringArray Exclude files matching pattern@@ -12768,7 +13982,7 @@ options not listed here.</p>--metadata-include-from stringArray Read metadata include patterns from file (use - to read from stdin)--min-age Duration Only transfer files older than this in s or suffix ms|s|m|h|d|w|M|y (default off)--min-size SizeSuffix Only transfer files bigger than this in KiB or suffix B|K|M|G|T|P (default off)</code></pre>-<h2 id="see-also-85">See Also</h2>+<h2 id="see-also-86">See Also</h2><!-- markdownlint-capture --><!-- markdownlint-disable ul-style line-length --><ul>@@ -12778,7 +13992,7 @@ options not listed here.</p><!-- markdownlint-restore --><h1 id="rclone-serve-sftp">rclone serve sftp</h1><p>Serve the remote over SFTP.</p>-<h2 id="synopsis-79">Synopsis</h2>+<h2 id="synopsis-80">Synopsis</h2><p>Run an SFTP server to serve a remote over SFTP. This can be used withan SFTP client or you can make a remote of type <a href="/sftp">sftp</a>to use with it.</p>@@ -12876,7 +14090,7 @@ that will be used to buffer data in advance.</p>memory at all times. The buffered data is bound to one open file andwon't be shared.</p><p>This flag is a upper limit for the used memory per open file. The-buffer will only use memory for data that is downloaded but not not yet+buffer will only use memory for data that is downloaded but not yetread. If the buffer is empty, only a small amount of memory will beused.</p><p>The maximum memory used by rclone for buffering can be up to@@ -12919,11 +14133,12 @@ will start with files that haven't been accessed for the longest. Thiscache flushing strategy is efficient and more relevant files are likelyto remain cached.</p><p>The <code>--vfs-cache-max-age</code> will evict files from the cache-after the set time since last access has passed. The default value of 1-hour will start evicting files from cache that haven't been accessed for-1 hour. When a cached file is accessed the 1 hour timer is reset to 0-and will wait for 1 more hour before evicting. Specify the time with-standard notation, s, m, h, d, w .</p>+after the set time since last access has passed; it is based on access+time, not on when the file was first added to the cache. The default+value of 1 hour will start evicting files from cache that haven't been+accessed for 1 hour. When a cached file is accessed the 1 hour timer is+reset to 0 and will wait for 1 more hour before evicting. Specify the+time with standard notation, s, m, h, d, w .</p><p>You <strong>should not</strong> run two copies of rclone using thesame VFS cache with the same or overlapping remotes if using<code>--vfs-cache-mode > off</code>. This can potentially cause data@@ -13049,8 +14264,8 @@ id="vfs-read-chunk-streams-0-17"><code>--vfs-read-chunk-streams</code><p>Rclone reads <code>--vfs-read-chunk-streams</code> chunks of size<code>--vfs-read-chunk-size</code> concurrently. The size for each readwill stay constant.</p>-<p>This improves performance performance massively on high latency links-or very high bandwidth links to high performance object stores.</p>+<p>This improves performance massively on high latency links or very+high bandwidth links to high performance object stores.</p><p>Some experimentation will be needed to find the optimum values of<code>--vfs-read-chunk-size</code> and<code>--vfs-read-chunk-streams</code> as these will depend on the@@ -13229,7 +14444,7 @@ total 1048578<p>If the file has no metadata it will be returned as <code>{}</code>and if there is an error reading the metadata the error will be returnedas <code>{"error":"error string"}</code>.</p>-<h2 id="auth-proxy-2">Auth Proxy</h2>+<h2 id="auth-proxy-3">Auth Proxy</h2><p>If you supply the parameter<code>--auth-proxy /path/to/program</code> then rclone will use thatprogram to generate backends on the fly which then are used to@@ -13252,35 +14467,69 @@ proxy program to make a complete config.</p><ul><li><code>_root</code> - root to use for the backend</li></ul>-<p>And it may have this parameter</p>+<p>And it may have these parameters</p><ul><li><code>_obscure</code> - comma separated strings for parameters toobscure</li>+<li><code>_secret_access_key</code> - the secret for S3 access key auth+(see below)</li></ul><p>If password authentication was used by the client, input to the proxyprocess (on STDIN) would look similar to this:</p>-<div class="sourceCode" id="cb596"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb596-1"><a href="#cb596-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>-<span id="cb596-2"><a href="#cb596-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"user"</span><span class="fu">:</span> <span class="st">"me"</span><span class="fu">,</span></span>-<span id="cb596-3"><a href="#cb596-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"pass"</span><span class="fu">:</span> <span class="st">"mypassword"</span></span>-<span id="cb596-4"><a href="#cb596-4" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<div class="sourceCode" id="cb619"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb619-1"><a href="#cb619-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb619-2"><a href="#cb619-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"user"</span><span class="fu">:</span> <span class="st">"me"</span><span class="fu">,</span></span>+<span id="cb619-3"><a href="#cb619-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"pass"</span><span class="fu">:</span> <span class="st">"mypassword"</span><span class="fu">,</span></span>+<span id="cb619-4"><a href="#cb619-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"client_ip"</span><span class="fu">:</span> <span class="st">"192.168.1.1"</span></span>+<span id="cb619-5"><a href="#cb619-5" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div><p>If public-key authentication was used by the client, input to theproxy process (on STDIN) would look similar to this:</p>-<div class="sourceCode" id="cb597"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb597-1"><a href="#cb597-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>-<span id="cb597-2"><a href="#cb597-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"user"</span><span class="fu">:</span> <span class="st">"me"</span><span class="fu">,</span></span>-<span id="cb597-3"><a href="#cb597-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"public_key"</span><span class="fu">:</span> <span class="st">"AAAAB3NzaC1yc2EAAAADAQABAAABAQDuwESFdAe14hVS6omeyX7edc...JQdf"</span></span>-<span id="cb597-4"><a href="#cb597-4" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<div class="sourceCode" id="cb620"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb620-1"><a href="#cb620-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb620-2"><a href="#cb620-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"user"</span><span class="fu">:</span> <span class="st">"me"</span><span class="fu">,</span></span>+<span id="cb620-3"><a href="#cb620-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"public_key"</span><span class="fu">:</span> <span class="st">"AAAAB3NzaC1yc2EAAAADAQABAAABAQDuwESFdAe14hVS6omeyX7edc...JQdf"</span><span class="fu">,</span></span>+<span id="cb620-4"><a href="#cb620-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"client_ip"</span><span class="fu">:</span> <span class="st">"192.168.1.1"</span></span>+<span id="cb620-5"><a href="#cb620-5" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<p>If the client authenticated with an S3 access key+(<code>rclone serve s3</code>), the client never sends its secret, only+a signature made with it, so the input contains just the access key ID+as the <code>user</code> with no <code>pass</code> or+<code>public_key</code>:</p>+<div class="sourceCode" id="cb621"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb621-1"><a href="#cb621-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb621-2"><a href="#cb621-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"user"</span><span class="fu">:</span> <span class="st">"AKIAIOSFODNN7EXAMPLE"</span><span class="fu">,</span></span>+<span id="cb621-3"><a href="#cb621-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"client_ip"</span><span class="fu">:</span> <span class="st">"192.168.1.1"</span></span>+<span id="cb621-4"><a href="#cb621-4" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<p>In this case the program must look up the secret access key for that+access key ID and return it in the <code>_secret_access_key</code> field+of the output. Rclone then uses that secret to verify the signature on+the request, refusing the request if it does not match. This means the+proxy program is the source of truth for both the credentials and the+backend they map to. If the program does not return+<code>_secret_access_key</code> or returns it empty the request is+refused.</p>+<p>The program's answer for an access key ID is cached (see below) but+is checked with the program again after 5 minutes even if the access key+ID is in constant use, so revoking an access key ID in the program takes+effect within 5 minutes. A rotated secret takes effect on the first+request signed with it.</p>+<p>The <code>client_ip</code> key holds the IP address the client+connected from, without a port number. It can be used to restrict logins+to certain networks, or to log authentication attempts centrally. It is+omitted if the client has no IP address, for example when connecting+over a unix socket. Note that if rclone is behind a reverse proxy this+will be the address of the reverse proxy and not the original+client.</p><p>And as an example return this on STDOUT</p>-<div class="sourceCode" id="cb598"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb598-1"><a href="#cb598-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>-<span id="cb598-2"><a href="#cb598-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"type"</span><span class="fu">:</span> <span class="st">"sftp"</span><span class="fu">,</span></span>-<span id="cb598-3"><a href="#cb598-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"_root"</span><span class="fu">:</span> <span class="st">""</span><span class="fu">,</span></span>-<span id="cb598-4"><a href="#cb598-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"_obscure"</span><span class="fu">:</span> <span class="st">"pass"</span><span class="fu">,</span></span>-<span id="cb598-5"><a href="#cb598-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"user"</span><span class="fu">:</span> <span class="st">"me"</span><span class="fu">,</span></span>-<span id="cb598-6"><a href="#cb598-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">"pass"</span><span class="fu">:</span> <span class="st">"mypassword"</span><span class="fu">,</span></span>-<span id="cb598-7"><a href="#cb598-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"host"</span><span class="fu">:</span> <span class="st">"sftp.example.com"</span></span>-<span id="cb598-8"><a href="#cb598-8" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<div class="sourceCode" id="cb622"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb622-1"><a href="#cb622-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb622-2"><a href="#cb622-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"type"</span><span class="fu">:</span> <span class="st">"sftp"</span><span class="fu">,</span></span>+<span id="cb622-3"><a href="#cb622-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"_root"</span><span class="fu">:</span> <span class="st">""</span><span class="fu">,</span></span>+<span id="cb622-4"><a href="#cb622-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"_obscure"</span><span class="fu">:</span> <span class="st">"pass"</span><span class="fu">,</span></span>+<span id="cb622-5"><a href="#cb622-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"user"</span><span class="fu">:</span> <span class="st">"me"</span><span class="fu">,</span></span>+<span id="cb622-6"><a href="#cb622-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">"pass"</span><span class="fu">:</span> <span class="st">"mypassword"</span><span class="fu">,</span></span>+<span id="cb622-7"><a href="#cb622-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"host"</span><span class="fu">:</span> <span class="st">"sftp.example.com"</span></span>+<span id="cb622-8"><a href="#cb622-8" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div><p>This would mean that an SFTP backend would be created on the fly forthe <code>user</code> and <code>pass</code>/<code>public_key</code>returned in the output to the host given. Note that since@@ -13293,16 +14542,23 @@ make the <code>user</code> be <code>[email protected]</code> and then setthe <code>host</code> to <code>example.com</code> in the output and theuser to <code>user</code>. For security you'd probably want to restrictthe <code>host</code> to a limited list.</p>-<p>An internal cache of backends is keyed on the <code>user</code> and a-hash of the <code>pass</code> or <code>public_key</code>. This means-that if a user's password or public-key changes, or the proxy returns-different config parameters (eg a rotated <code>api_key</code>), a fresh-backend will be created on the next request rather than the cached one-being reused.</p>+<p>An internal cache of backends is keyed on the <code>user</code>, the+kind of credential presented, a hash of the <code>pass</code> or+<code>public_key</code>, and the <code>client_ip</code>. This means that+if a user's password or public-key changes, the client connects from a+new IP address, or the proxy returns different config parameters (eg a+rotated <code>api_key</code>), a fresh backend will be created on the+next request rather than the cached one being reused. A backend created+for a public key login is never reused for a password login, so a public+key cannot be sent as a password to skip the program.</p>+<p>A cached answer is checked with the program again after 5 minutes+even if the user is in constant use, so revoking a user in the program+takes effect within 5 minutes. For <code>serve sftp</code> this applies+to new logins - an existing connection is not closed.</p><p>This can be used to build general purpose proxies to any kind ofbackend that rclone supports.</p><pre><code>rclone serve sftp remote:path [flags]</code></pre>-<h2 id="options-86">Options</h2>+<h2 id="options-87">Options</h2><pre><code> --addr string IPaddress:Port or :Port to bind server to (default "localhost:2022")--auth-proxy string A program to use to create the backend from the auth--authorized-keys string Authorized keys file (default "~/.ssh/authorized_keys")@@ -13348,7 +14604,7 @@ backend that rclone supports.</p><p>Options shared with other commands are described next. See the <ahref="https://rclone.org/flags/">global flags page</a> for globaloptions not listed here.</p>-<h3 id="filter-options-30">Filter Options</h3>+<h3 id="filter-options-31">Filter Options</h3><p>Flags for filtering directory listings</p><pre class="text"><code> --delete-excluded Delete files on dest excluded from sync--exclude stringArray Exclude files matching pattern@@ -13374,7 +14630,7 @@ options not listed here.</p>--metadata-include-from stringArray Read metadata include patterns from file (use - to read from stdin)--min-age Duration Only transfer files older than this in s or suffix ms|s|m|h|d|w|M|y (default off)--min-size SizeSuffix Only transfer files bigger than this in KiB or suffix B|K|M|G|T|P (default off)</code></pre>-<h2 id="see-also-86">See Also</h2>+<h2 id="see-also-87">See Also</h2><!-- markdownlint-capture --><!-- markdownlint-disable ul-style line-length --><ul>@@ -13384,7 +14640,7 @@ options not listed here.</p><!-- markdownlint-restore --><h1 id="rclone-serve-webdav">rclone serve webdav</h1><p>Serve remote:path over WebDAV.</p>-<h2 id="synopsis-80">Synopsis</h2>+<h2 id="synopsis-81">Synopsis</h2><p>Run a basic WebDAV server to serve a remote over HTTP via the WebDAVprotocol. This can be viewed with a WebDAV client, through a webbrowser, or you can make a remote of type WebDAV to read and write@@ -13471,6 +14727,14 @@ the server will accept in the HTTP header.</p>all responses, will overriding existing values. The flag may be repeatedto add multiple headers. Use the format<code>Header-Name: value</code>.</p>+<p><code>--allow-origin</code> enables cross-origin (CORS) requests from+browsers by setting the <code>Access-Control-Allow-Origin</code> header+on every response to the origin given, e.g.+<code>--allow-origin https://app.example.com</code>. Browser CORS+preflight OPTIONS requests are answered by rclone without needing+authentication as browsers never send credentials with them. Setting+this to <code>*</code> allows any origin which is a security risk on an+authenticated server.</p><p><code>--baseurl</code> controls the URL prefix that rclone servesfrom. By default rclone will serve from the root. If you used<code>--baseurl "/rclone"</code> then rclone would serve from a URL@@ -13567,55 +14831,101 @@ asc).</td><td style="text-align: left;">Currently unused.</td></tr><tr class="even">+<td style="text-align: left;">.Path</td>+<td style="text-align: left;"></td>+<td style="text-align: left;">The path of the directory from the root+with leading and trailing slashes, e.g. '/dir/'.</td>+</tr>+<tr class="odd">+<td style="text-align: left;">.IsRoot</td>+<td style="text-align: left;"></td>+<td style="text-align: left;">Boolean for if this is the root+directory.</td>+</tr>+<tr class="even">+<td style="text-align: left;">.UpLink</td>+<td style="text-align: left;"></td>+<td style="text-align: left;">The link to the parent directory.</td>+</tr>+<tr class="odd">+<td style="text-align: left;">.Static</td>+<td style="text-align: left;"></td>+<td style="text-align: left;">Boolean for if the listing is being+written as a static page by rclone index.</td>+</tr>+<tr class="even">+<td style="text-align: left;">.NumDirs</td>+<td style="text-align: left;"></td>+<td style="text-align: left;">The number of directories in the+listing.</td>+</tr>+<tr class="odd">+<td style="text-align: left;">.NumFiles</td>+<td style="text-align: left;"></td>+<td style="text-align: left;">The number of files in the listing.</td>+</tr>+<tr class="even">+<td style="text-align: left;">.TotalSize</td>+<td style="text-align: left;"></td>+<td style="text-align: left;">The total size in bytes of the files in+the listing.</td>+</tr>+<tr class="odd"><td style="text-align: left;">.Breadcrumb</td><td style="text-align: left;"></td><td style="text-align: left;">Allows for creating a relativenavigation.</td></tr>-<tr class="odd">+<tr class="even"><td style="text-align: left;"></td><td style="text-align: left;">.Link</td><td style="text-align: left;">The link of the Text relative to theroot.</td></tr>-<tr class="even">+<tr class="odd"><td style="text-align: left;"></td><td style="text-align: left;">.Text</td><td style="text-align: left;">The Name of the directory.</td></tr>-<tr class="odd">+<tr class="even"><td style="text-align: left;">.Entries</td><td style="text-align: left;"></td><td style="text-align: left;">Information about a specificfile/directory.</td></tr>-<tr class="even">+<tr class="odd"><td style="text-align: left;"></td><td style="text-align: left;">.URL</td><td style="text-align: left;">The url of an entry.</td></tr>-<tr class="odd">+<tr class="even"><td style="text-align: left;"></td><td style="text-align: left;">.Leaf</td><td style="text-align: left;">Currently same as '.URL' but intended tobe just the name.</td></tr>-<tr class="even">+<tr class="odd"><td style="text-align: left;"></td><td style="text-align: left;">.IsDir</td><td style="text-align: left;">Boolean for if an entry is a directory ornot.</td></tr>-<tr class="odd">+<tr class="even"><td style="text-align: left;"></td><td style="text-align: left;">.Size</td><td style="text-align: left;">Size in bytes of the entry.</td></tr>-<tr class="even">+<tr class="odd"><td style="text-align: left;"></td><td style="text-align: left;">.ModTime</td><td style="text-align: left;">The UTC timestamp of an entry.</td></tr>+<tr class="even">+<td style="text-align: left;"></td>+<td style="text-align: left;">.MimeType</td>+<td style="text-align: left;">The MIME type of an entry guessed from its+name, or 'inode/directory'.</td>+</tr></tbody></table><p>The server also makes the following functions available so that they@@ -13678,7 +14988,17 @@ BCrypt for basic authentication. Bcrypt is recommended.</p><pre class="console"><code>touch htpasswdhtpasswd -B htpasswd userhtpasswd -B htpasswd anotherUser</code></pre>+<p>Use the same hash format and cost for every entry in the file. If the+entries differ, the time taken to reject a wrong password can reveal+which usernames exist.</p><p>The password file can be updated while rclone is running.</p>+<p>When authentication is enabled every request must be authenticated,+including HTTP OPTIONS requests, as the answer to an OPTIONS request can+reveal whether a path exists. Health checks or monitoring which use+unauthenticated OPTIONS requests will need to send credentials. The+exception is browser CORS preflight requests when+<code>--allow-origin</code> is set, which are answered without+authentication as browsers never send credentials with them.</p><p>Use <code>--realm</code> to set the authentication realm.</p><p>Use <code>--salt</code> to change the password hashing salt from thedefault.</p>@@ -13720,7 +15040,7 @@ that will be used to buffer data in advance.</p>memory at all times. The buffered data is bound to one open file andwon't be shared.</p><p>This flag is a upper limit for the used memory per open file. The-buffer will only use memory for data that is downloaded but not not yet+buffer will only use memory for data that is downloaded but not yetread. If the buffer is empty, only a small amount of memory will beused.</p><p>The maximum memory used by rclone for buffering can be up to@@ -13763,11 +15083,12 @@ will start with files that haven't been accessed for the longest. Thiscache flushing strategy is efficient and more relevant files are likelyto remain cached.</p><p>The <code>--vfs-cache-max-age</code> will evict files from the cache-after the set time since last access has passed. The default value of 1-hour will start evicting files from cache that haven't been accessed for-1 hour. When a cached file is accessed the 1 hour timer is reset to 0-and will wait for 1 more hour before evicting. Specify the time with-standard notation, s, m, h, d, w .</p>+after the set time since last access has passed; it is based on access+time, not on when the file was first added to the cache. The default+value of 1 hour will start evicting files from cache that haven't been+accessed for 1 hour. When a cached file is accessed the 1 hour timer is+reset to 0 and will wait for 1 more hour before evicting. Specify the+time with standard notation, s, m, h, d, w .</p><p>You <strong>should not</strong> run two copies of rclone using thesame VFS cache with the same or overlapping remotes if using<code>--vfs-cache-mode > off</code>. This can potentially cause data@@ -13893,8 +15214,8 @@ id="vfs-read-chunk-streams-0-19"><code>--vfs-read-chunk-streams</code><p>Rclone reads <code>--vfs-read-chunk-streams</code> chunks of size<code>--vfs-read-chunk-size</code> concurrently. The size for each readwill stay constant.</p>-<p>This improves performance performance massively on high latency links-or very high bandwidth links to high performance object stores.</p>+<p>This improves performance massively on high latency links or very+high bandwidth links to high performance object stores.</p><p>Some experimentation will be needed to find the optimum values of<code>--vfs-read-chunk-size</code> and<code>--vfs-read-chunk-streams</code> as these will depend on the@@ -14073,7 +15394,7 @@ total 1048578<p>If the file has no metadata it will be returned as <code>{}</code>and if there is an error reading the metadata the error will be returnedas <code>{"error":"error string"}</code>.</p>-<h2 id="auth-proxy-3">Auth Proxy</h2>+<h2 id="auth-proxy-4">Auth Proxy</h2><p>If you supply the parameter<code>--auth-proxy /path/to/program</code> then rclone will use thatprogram to generate backends on the fly which then are used to@@ -14096,35 +15417,69 @@ proxy program to make a complete config.</p><ul><li><code>_root</code> - root to use for the backend</li></ul>-<p>And it may have this parameter</p>+<p>And it may have these parameters</p><ul><li><code>_obscure</code> - comma separated strings for parameters toobscure</li>+<li><code>_secret_access_key</code> - the secret for S3 access key auth+(see below)</li></ul><p>If password authentication was used by the client, input to the proxyprocess (on STDIN) would look similar to this:</p>-<div class="sourceCode" id="cb621"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb621-1"><a href="#cb621-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>-<span id="cb621-2"><a href="#cb621-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"user"</span><span class="fu">:</span> <span class="st">"me"</span><span class="fu">,</span></span>-<span id="cb621-3"><a href="#cb621-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"pass"</span><span class="fu">:</span> <span class="st">"mypassword"</span></span>-<span id="cb621-4"><a href="#cb621-4" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<div class="sourceCode" id="cb645"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb645-1"><a href="#cb645-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb645-2"><a href="#cb645-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"user"</span><span class="fu">:</span> <span class="st">"me"</span><span class="fu">,</span></span>+<span id="cb645-3"><a href="#cb645-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"pass"</span><span class="fu">:</span> <span class="st">"mypassword"</span><span class="fu">,</span></span>+<span id="cb645-4"><a href="#cb645-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"client_ip"</span><span class="fu">:</span> <span class="st">"192.168.1.1"</span></span>+<span id="cb645-5"><a href="#cb645-5" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div><p>If public-key authentication was used by the client, input to theproxy process (on STDIN) would look similar to this:</p>-<div class="sourceCode" id="cb622"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb622-1"><a href="#cb622-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>-<span id="cb622-2"><a href="#cb622-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"user"</span><span class="fu">:</span> <span class="st">"me"</span><span class="fu">,</span></span>-<span id="cb622-3"><a href="#cb622-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"public_key"</span><span class="fu">:</span> <span class="st">"AAAAB3NzaC1yc2EAAAADAQABAAABAQDuwESFdAe14hVS6omeyX7edc...JQdf"</span></span>-<span id="cb622-4"><a href="#cb622-4" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<div class="sourceCode" id="cb646"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb646-1"><a href="#cb646-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb646-2"><a href="#cb646-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"user"</span><span class="fu">:</span> <span class="st">"me"</span><span class="fu">,</span></span>+<span id="cb646-3"><a href="#cb646-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"public_key"</span><span class="fu">:</span> <span class="st">"AAAAB3NzaC1yc2EAAAADAQABAAABAQDuwESFdAe14hVS6omeyX7edc...JQdf"</span><span class="fu">,</span></span>+<span id="cb646-4"><a href="#cb646-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"client_ip"</span><span class="fu">:</span> <span class="st">"192.168.1.1"</span></span>+<span id="cb646-5"><a href="#cb646-5" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<p>If the client authenticated with an S3 access key+(<code>rclone serve s3</code>), the client never sends its secret, only+a signature made with it, so the input contains just the access key ID+as the <code>user</code> with no <code>pass</code> or+<code>public_key</code>:</p>+<div class="sourceCode" id="cb647"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb647-1"><a href="#cb647-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb647-2"><a href="#cb647-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"user"</span><span class="fu">:</span> <span class="st">"AKIAIOSFODNN7EXAMPLE"</span><span class="fu">,</span></span>+<span id="cb647-3"><a href="#cb647-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"client_ip"</span><span class="fu">:</span> <span class="st">"192.168.1.1"</span></span>+<span id="cb647-4"><a href="#cb647-4" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<p>In this case the program must look up the secret access key for that+access key ID and return it in the <code>_secret_access_key</code> field+of the output. Rclone then uses that secret to verify the signature on+the request, refusing the request if it does not match. This means the+proxy program is the source of truth for both the credentials and the+backend they map to. If the program does not return+<code>_secret_access_key</code> or returns it empty the request is+refused.</p>+<p>The program's answer for an access key ID is cached (see below) but+is checked with the program again after 5 minutes even if the access key+ID is in constant use, so revoking an access key ID in the program takes+effect within 5 minutes. A rotated secret takes effect on the first+request signed with it.</p>+<p>The <code>client_ip</code> key holds the IP address the client+connected from, without a port number. It can be used to restrict logins+to certain networks, or to log authentication attempts centrally. It is+omitted if the client has no IP address, for example when connecting+over a unix socket. Note that if rclone is behind a reverse proxy this+will be the address of the reverse proxy and not the original+client.</p><p>And as an example return this on STDOUT</p>-<div class="sourceCode" id="cb623"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb623-1"><a href="#cb623-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>-<span id="cb623-2"><a href="#cb623-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"type"</span><span class="fu">:</span> <span class="st">"sftp"</span><span class="fu">,</span></span>-<span id="cb623-3"><a href="#cb623-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"_root"</span><span class="fu">:</span> <span class="st">""</span><span class="fu">,</span></span>-<span id="cb623-4"><a href="#cb623-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"_obscure"</span><span class="fu">:</span> <span class="st">"pass"</span><span class="fu">,</span></span>-<span id="cb623-5"><a href="#cb623-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"user"</span><span class="fu">:</span> <span class="st">"me"</span><span class="fu">,</span></span>-<span id="cb623-6"><a href="#cb623-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">"pass"</span><span class="fu">:</span> <span class="st">"mypassword"</span><span class="fu">,</span></span>-<span id="cb623-7"><a href="#cb623-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"host"</span><span class="fu">:</span> <span class="st">"sftp.example.com"</span></span>-<span id="cb623-8"><a href="#cb623-8" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<div class="sourceCode" id="cb648"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb648-1"><a href="#cb648-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb648-2"><a href="#cb648-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"type"</span><span class="fu">:</span> <span class="st">"sftp"</span><span class="fu">,</span></span>+<span id="cb648-3"><a href="#cb648-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"_root"</span><span class="fu">:</span> <span class="st">""</span><span class="fu">,</span></span>+<span id="cb648-4"><a href="#cb648-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"_obscure"</span><span class="fu">:</span> <span class="st">"pass"</span><span class="fu">,</span></span>+<span id="cb648-5"><a href="#cb648-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"user"</span><span class="fu">:</span> <span class="st">"me"</span><span class="fu">,</span></span>+<span id="cb648-6"><a href="#cb648-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">"pass"</span><span class="fu">:</span> <span class="st">"mypassword"</span><span class="fu">,</span></span>+<span id="cb648-7"><a href="#cb648-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"host"</span><span class="fu">:</span> <span class="st">"sftp.example.com"</span></span>+<span id="cb648-8"><a href="#cb648-8" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div><p>This would mean that an SFTP backend would be created on the fly forthe <code>user</code> and <code>pass</code>/<code>public_key</code>returned in the output to the host given. Note that since@@ -14137,16 +15492,23 @@ make the <code>user</code> be <code>[email protected]</code> and then setthe <code>host</code> to <code>example.com</code> in the output and theuser to <code>user</code>. For security you'd probably want to restrictthe <code>host</code> to a limited list.</p>-<p>An internal cache of backends is keyed on the <code>user</code> and a-hash of the <code>pass</code> or <code>public_key</code>. This means-that if a user's password or public-key changes, or the proxy returns-different config parameters (eg a rotated <code>api_key</code>), a fresh-backend will be created on the next request rather than the cached one-being reused.</p>+<p>An internal cache of backends is keyed on the <code>user</code>, the+kind of credential presented, a hash of the <code>pass</code> or+<code>public_key</code>, and the <code>client_ip</code>. This means that+if a user's password or public-key changes, the client connects from a+new IP address, or the proxy returns different config parameters (eg a+rotated <code>api_key</code>), a fresh backend will be created on the+next request rather than the cached one being reused. A backend created+for a public key login is never reused for a password login, so a public+key cannot be sent as a password to skip the program.</p>+<p>A cached answer is checked with the program again after 5 minutes+even if the user is in constant use, so revoking a user in the program+takes effect within 5 minutes. For <code>serve sftp</code> this applies+to new logins - an existing connection is not closed.</p><p>This can be used to build general purpose proxies to any kind ofbackend that rclone supports.</p><pre><code>rclone serve webdav remote:path [flags]</code></pre>-<h2 id="options-87">Options</h2>+<h2 id="options-88">Options</h2><pre><code> --addr stringArray IPaddress:Port or :Port to bind server to (default 127.0.0.1:8080)--allow-origin string Origin which cross-domain request (CORS) can be executed from--auth-proxy string A program to use to create the backend from the auth@@ -14206,7 +15568,7 @@ backend that rclone supports.</p><p>Options shared with other commands are described next. See the <ahref="https://rclone.org/flags/">global flags page</a> for globaloptions not listed here.</p>-<h3 id="filter-options-31">Filter Options</h3>+<h3 id="filter-options-32">Filter Options</h3><p>Flags for filtering directory listings</p><pre class="text"><code> --delete-excluded Delete files on dest excluded from sync--exclude stringArray Exclude files matching pattern@@ -14232,7 +15594,7 @@ options not listed here.</p>--metadata-include-from stringArray Read metadata include patterns from file (use - to read from stdin)--min-age Duration Only transfer files older than this in s or suffix ms|s|m|h|d|w|M|y (default off)--min-size SizeSuffix Only transfer files bigger than this in KiB or suffix B|K|M|G|T|P (default off)</code></pre>-<h2 id="see-also-87">See Also</h2>+<h2 id="see-also-88">See Also</h2><!-- markdownlint-capture --><!-- markdownlint-disable ul-style line-length --><ul>@@ -14242,7 +15604,7 @@ options not listed here.</p><!-- markdownlint-restore --><h1 id="rclone-settier">rclone settier</h1><p>Changes storage class/tier of objects in remote.</p>-<h2 id="synopsis-81">Synopsis</h2>+<h2 id="synopsis-82">Synopsis</h2><p>Changes storage tier or class at remote if supported. Few cloudstorage services provides different storage classes on objects, forexample AWS S3 and Glacier, Azure Blob storage - Hot, Cool and Archive,@@ -14259,11 +15621,11 @@ similarly S3 to Glacier makes object inaccessible.true</p>tiered</p><pre class="console"><code>rclone settier tier remote:path/dir</code></pre><pre><code>rclone settier tier remote:path [flags]</code></pre>-<h2 id="options-88">Options</h2>+<h2 id="options-89">Options</h2><pre><code> -h, --help help for settier</code></pre><p>See the <a href="https://rclone.org/flags/">global flags page</a> forglobal options not listed here.</p>-<h2 id="see-also-88">See Also</h2>+<h2 id="see-also-89">See Also</h2><!-- markdownlint-capture --><!-- markdownlint-disable ul-style line-length --><ul>@@ -14273,7 +15635,7 @@ for rclone commands, flags and backends.</li><!-- markdownlint-restore --><h1 id="rclone-test">rclone test</h1><p>Run a test command</p>-<h2 id="synopsis-82">Synopsis</h2>+<h2 id="synopsis-83">Synopsis</h2><p>Rclone test is used to run test commands.</p><p>Select which test command you want with the subcommand, eg</p><pre class="console"><code>rclone test memory remote:</code></pre>@@ -14281,11 +15643,11 @@ for rclone commands, flags and backends.</li>help.</p><p><strong>NB</strong> Be careful running these commands, they may dostrange things so reading their documentation first is recommended.</p>-<h2 id="options-89">Options</h2>+<h2 id="options-90">Options</h2><pre><code> -h, --help help for test</code></pre><p>See the <a href="https://rclone.org/flags/">global flags page</a> forglobal options not listed here.</p>-<h2 id="see-also-89">See Also</h2>+<h2 id="see-also-90">See Also</h2><!-- markdownlint-capture --><!-- markdownlint-disable ul-style line-length --><ul>@@ -14314,12 +15676,12 @@ speed</a> - Run a speed test to the remote</li><h1 id="rclone-test-changenotify">rclone test changenotify</h1><p>Log any change notify requests for the remote passed in.</p><pre><code>rclone test changenotify remote: [flags]</code></pre>-<h2 id="options-90">Options</h2>+<h2 id="options-91">Options</h2><pre><code> -h, --help help for changenotify--poll-interval Duration Time to wait between polling for changes (default 10s)</code></pre><p>See the <a href="https://rclone.org/flags/">global flags page</a> forglobal options not listed here.</p>-<h2 id="see-also-90">See Also</h2>+<h2 id="see-also-91">See Also</h2><!-- markdownlint-capture --><!-- markdownlint-disable ul-style line-length --><ul>@@ -14329,17 +15691,17 @@ Run a test command</li><!-- markdownlint-restore --><h1 id="rclone-test-histogram">rclone test histogram</h1><p>Makes a histogram of file name characters.</p>-<h2 id="synopsis-83">Synopsis</h2>+<h2 id="synopsis-84">Synopsis</h2><p>This command outputs JSON which shows the histogram of charactersused in filenames in the remote:path specified.</p><p>The data doesn't contain any identifying information but is usefulfor the rclone developers when developing filename compression.</p><pre><code>rclone test histogram [remote:path] [flags]</code></pre>-<h2 id="options-91">Options</h2>+<h2 id="options-92">Options</h2><pre><code> -h, --help help for histogram</code></pre><p>See the <a href="https://rclone.org/flags/">global flags page</a> forglobal options not listed here.</p>-<h2 id="see-also-91">See Also</h2>+<h2 id="see-also-92">See Also</h2><!-- markdownlint-capture --><!-- markdownlint-disable ul-style line-length --><ul>@@ -14349,7 +15711,7 @@ Run a test command</li><!-- markdownlint-restore --><h1 id="rclone-test-info">rclone test info</h1><p>Discovers file name or other limitations for paths.</p>-<h2 id="synopsis-84">Synopsis</h2>+<h2 id="synopsis-85">Synopsis</h2><p>Discovers what filenames and upload methods are possible to write tothe paths passed in and how long they can be. It can take some time. Itwill write test files into the remote:path passed in. It outputs a bit@@ -14357,7 +15719,7 @@ of go code for each one.</p><p><strong>NB</strong> this can create undeletable files and otherhazards - use with care!</p><pre><code>rclone test info [remote:path]+ [flags]</code></pre>-<h2 id="options-92">Options</h2>+<h2 id="options-93">Options</h2><pre><code> --all Run all tests--check-base32768 Check can store all possible base32768 characters--check-control Check control characters@@ -14370,7 +15732,7 @@ hazards - use with care!</p>--write-json string Write results to file</code></pre><p>See the <a href="https://rclone.org/flags/">global flags page</a> forglobal options not listed here.</p>-<h2 id="see-also-92">See Also</h2>+<h2 id="see-also-93">See Also</h2><!-- markdownlint-capture --><!-- markdownlint-disable ul-style line-length --><ul>@@ -14381,7 +15743,7 @@ Run a test command</li><h1 id="rclone-test-makefile">rclone test makefile</h1><p>Make files with random contents of the size given</p><pre><code>rclone test makefile <size> [<file>]+ [flags]</code></pre>-<h2 id="options-93">Options</h2>+<h2 id="options-94">Options</h2><pre><code> --ascii Fill files with random ASCII printable bytes only--chargen Fill files with a ASCII chargen pattern-h, --help help for makefile@@ -14391,7 +15753,7 @@ Run a test command</li>--zero Fill files with ASCII 0x00</code></pre><p>See the <a href="https://rclone.org/flags/">global flags page</a> forglobal options not listed here.</p>-<h2 id="see-also-93">See Also</h2>+<h2 id="see-also-94">See Also</h2><!-- markdownlint-capture --><!-- markdownlint-disable ul-style line-length --><ul>@@ -14402,7 +15764,7 @@ Run a test command</li><h1 id="rclone-test-makefiles">rclone test makefiles</h1><p>Make a random file hierarchy in a directory</p><pre><code>rclone test makefiles <dir> [flags]</code></pre>-<h2 id="options-94">Options</h2>+<h2 id="options-95">Options</h2><pre><code> --ascii Fill files with random ASCII printable bytes only--chargen Fill files with a ASCII chargen pattern--files int Number of files to create (default 1000)@@ -14420,7 +15782,7 @@ Run a test command</li>--zero Fill files with ASCII 0x00</code></pre><p>See the <a href="https://rclone.org/flags/">global flags page</a> forglobal options not listed here.</p>-<h2 id="see-also-94">See Also</h2>+<h2 id="see-also-95">See Also</h2><!-- markdownlint-capture --><!-- markdownlint-disable ul-style line-length --><ul>@@ -14432,11 +15794,11 @@ Run a test command</li><p>Load all the objects at remote:path into memory and report memorystats.</p><pre><code>rclone test memory remote:path [flags]</code></pre>-<h2 id="options-95">Options</h2>+<h2 id="options-96">Options</h2><pre><code> -h, --help help for memory</code></pre><p>See the <a href="https://rclone.org/flags/">global flags page</a> forglobal options not listed here.</p>-<h2 id="see-also-95">See Also</h2>+<h2 id="see-also-96">See Also</h2><!-- markdownlint-capture --><!-- markdownlint-disable ul-style line-length --><ul>@@ -14446,12 +15808,12 @@ Run a test command</li><!-- markdownlint-restore --><h1 id="rclone-test-speed">rclone test speed</h1><p>Run a speed test to the remote</p>-<h2 id="synopsis-85">Synopsis</h2>+<h2 id="synopsis-86">Synopsis</h2><p>Run a speed test to the remote.</p><p>This command runs a series of uploads and downloads to the remote,measuring and printing the speed of each test using varying file sizesand numbers of files.</p>-<p>Test time can be innaccurate with small file caps and large files. As+<p>Test time can be inaccurate with small file caps and large files. Asit uses the results of an initial test to determine how many files touse in each subsequent test.</p><p>It is recommended to use -q flag for a simpler output. e.g.:</p>@@ -14462,7 +15824,7 @@ on a clean exit.</p><p>You can use the --json flag to only print the results in JSONformat.</p><pre><code>rclone test speed <remote> [flags]</code></pre>-<h2 id="options-96">Options</h2>+<h2 id="options-97">Options</h2><pre><code> --ascii Fill files with random ASCII printable bytes only--chargen Fill files with a ASCII chargen pattern--file-cap int Maximum number of files to use in each test (default 100)@@ -14478,7 +15840,7 @@ format.</p>--zero Fill files with ASCII 0x00</code></pre><p>See the <a href="https://rclone.org/flags/">global flags page</a> forglobal options not listed here.</p>-<h2 id="see-also-96">See Also</h2>+<h2 id="see-also-97">See Also</h2><!-- markdownlint-capture --><!-- markdownlint-disable ul-style line-length --><ul>@@ -14488,7 +15850,7 @@ Run a test command</li><!-- markdownlint-restore --><h1 id="rclone-touch">rclone touch</h1><p>Create new file or change file modification time.</p>-<h2 id="synopsis-86">Synopsis</h2>+<h2 id="synopsis-87">Synopsis</h2><p>Set the modification time on file(s) as specified by remote:path tohave the current time.</p><p>If remote:path does not exist then a zero sized file will be created,@@ -14503,9 +15865,11 @@ or the <code>--interactive</code>/<code>-i</code> flag. This will touchto that time instead of the current time. Times may be specified as oneof:</p><ul>-<li>'YYMMDD' - e.g. 17.10.30</li>+<li>'YYMMDD' - e.g. 171030</li>+<li>'YYYY-MM-DD' - e.g. 2006-01-02</li><li>'YYYY-MM-DDTHH:MM:SS' - e.g. 2006-01-02T15:04:05</li><li>'YYYY-MM-DDTHH:MM:SS.SSS' - e.g. 2006-01-02T15:04:05.123456789</li>+<li>'YYYY-MM-DDTHH:MM:SSZ' - e.g. 2006-01-02T15:04:05Z</li></ul><p>Note that value of <code>--timestamp</code> is in UTC. If you wantlocal time then add the <code>--localtime</code> flag.</p>@@ -14513,7 +15877,7 @@ local time then add the <code>--localtime</code> flag.</p><code>--metadata-set</code>. For example: rclone touch remote:path -M--metadata-set key=value</p><pre><code>rclone touch remote:path [flags]</code></pre>-<h2 id="options-97">Options</h2>+<h2 id="options-98">Options</h2><pre><code> -h, --help help for touch--localtime Use localtime for timestamp, not UTC-C, --no-create Do not create the file if it does not exist (implied with --recursive)@@ -14522,12 +15886,12 @@ local time then add the <code>--localtime</code> flag.</p><p>Options shared with other commands are described next. See the <ahref="https://rclone.org/flags/">global flags page</a> for globaloptions not listed here.</p>-<h3 id="important-options-18">Important Options</h3>+<h3 id="important-options-19">Important Options</h3><p>Important flags useful for most commands</p><pre class="text"><code> -n, --dry-run Do a trial run with no permanent changes-i, --interactive Enable interactive mode-v, --verbose count Print lots more stuff (repeat for more)</code></pre>-<h3 id="filter-options-32">Filter Options</h3>+<h3 id="filter-options-33">Filter Options</h3><p>Flags for filtering directory listings</p><pre class="text"><code> --delete-excluded Delete files on dest excluded from sync--exclude stringArray Exclude files matching pattern@@ -14553,11 +15917,11 @@ options not listed here.</p>--metadata-include-from stringArray Read metadata include patterns from file (use - to read from stdin)--min-age Duration Only transfer files older than this in s or suffix ms|s|m|h|d|w|M|y (default off)--min-size SizeSuffix Only transfer files bigger than this in KiB or suffix B|K|M|G|T|P (default off)</code></pre>-<h3 id="listing-options-21">Listing Options</h3>+<h3 id="listing-options-22">Listing Options</h3><p>Flags for listing directories</p><pre class="text"><code> --default-time Time Time to show if modtime is unknown for files and directories (default 2000-01-01T00:00:00Z)--fast-list Use recursive list if available; uses more memory but fewer transactions</code></pre>-<h2 id="see-also-97">See Also</h2>+<h2 id="see-also-98">See Also</h2><!-- markdownlint-capture --><!-- markdownlint-disable ul-style line-length --><ul>@@ -14567,7 +15931,7 @@ for rclone commands, flags and backends.</li><!-- markdownlint-restore --><h1 id="rclone-tree">rclone tree</h1><p>List the contents of the remote in a tree like fashion.</p>-<h2 id="synopsis-87">Synopsis</h2>+<h2 id="synopsis-88">Synopsis</h2><p>Lists the contents of a remote in a similar way to the unix treecommand.</p><p>For example</p>@@ -14591,7 +15955,7 @@ options as they conflict with rclone's short options.</p><p>For a more interactive navigation of the remote see the <ahref="https://rclone.org/commands/rclone_ncdu/">ncdu</a> command.</p><pre><code>rclone tree remote:path [flags]</code></pre>-<h2 id="options-98">Options</h2>+<h2 id="options-99">Options</h2><pre><code> -a, --all All files are listed (list . files too)-d, --dirs-only List directories only--dirsfirst List directories before files (-U disables)@@ -14614,7 +15978,7 @@ href="https://rclone.org/commands/rclone_ncdu/">ncdu</a> command.</p><p>Options shared with other commands are described next. See the <ahref="https://rclone.org/flags/">global flags page</a> for globaloptions not listed here.</p>-<h3 id="filter-options-33">Filter Options</h3>+<h3 id="filter-options-34">Filter Options</h3><p>Flags for filtering directory listings</p><pre class="text"><code> --delete-excluded Delete files on dest excluded from sync--exclude stringArray Exclude files matching pattern@@ -14640,11 +16004,11 @@ options not listed here.</p>--metadata-include-from stringArray Read metadata include patterns from file (use - to read from stdin)--min-age Duration Only transfer files older than this in s or suffix ms|s|m|h|d|w|M|y (default off)--min-size SizeSuffix Only transfer files bigger than this in KiB or suffix B|K|M|G|T|P (default off)</code></pre>-<h3 id="listing-options-22">Listing Options</h3>+<h3 id="listing-options-23">Listing Options</h3><p>Flags for listing directories</p><pre class="text"><code> --default-time Time Time to show if modtime is unknown for files and directories (default 2000-01-01T00:00:00Z)--fast-list Use recursive list if available; uses more memory but fewer transactions</code></pre>-<h2 id="see-also-98">See Also</h2>+<h2 id="see-also-99">See Also</h2><!-- markdownlint-capture --><!-- markdownlint-disable ul-style line-length --><ul>@@ -14812,6 +16176,11 @@ must be cautious to use characters that are possible to write in all ofthem. This is mostly a problem on Windows, where the consoletraditionally uses a non-Unicode character set - defined by theso-called "code page".</p>+<p>If you want to configure a remote using environment variables then+its name becomes part of the environment variable name, which in+practice restricts the characters you can use in it. See the <a+href="#config-file">Config file</a> section under Environment variables+for details.</p><p>Do not use single character names on Windows as it creates ambiguitywith Windows drives' names, e.g.: remote called <code>C</code> isindistinguishable from <code>C</code> drive. Rclone will always assume@@ -14837,11 +16206,11 @@ infrastructure without a proper certificate. You could supply the<code>--no-check-certificate</code> flag to rclone, but this will affect<strong>all</strong> the remotes. To make it just affect this remote youuse an override. You could put this in the config file:</p>-<div class="sourceCode" id="cb682"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb682-1"><a href="#cb682-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[remote]</span></span>-<span id="cb682-2"><a href="#cb682-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> XXX</span></span>-<span id="cb682-3"><a href="#cb682-3" aria-hidden="true" tabindex="-1"></a><span class="dt">...</span></span>-<span id="cb682-4"><a href="#cb682-4" aria-hidden="true" tabindex="-1"></a><span class="dt">override.no_check_certificate </span><span class="ot">=</span><span class="st"> </span><span class="kw">true</span></span></code></pre></div>+<div class="sourceCode" id="cb707"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb707-1"><a href="#cb707-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[remote]</span></span>+<span id="cb707-2"><a href="#cb707-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> XXX</span></span>+<span id="cb707-3"><a href="#cb707-3" aria-hidden="true" tabindex="-1"></a><span class="dt">...</span></span>+<span id="cb707-4"><a href="#cb707-4" aria-hidden="true" tabindex="-1"></a><span class="dt">override.no_check_certificate </span><span class="ot">=</span><span class="st"> </span><span class="kw">true</span></span></code></pre></div><p>or use it in the connection string<code>remote,override.no_check_certificate=true:</code> (or just<code>remote,override.no_check_certificate:</code>).</p>@@ -14852,6 +16221,7 @@ gets <code>-</code> replaced with <code>_</code> and gets anconfig is only applied during the remote creation. Here is a nonexhaustive list of ones which might be useful:</p><ul>+<li><code>allow_https_downgrade</code></li><li><code>bind_addr</code></li><li><code>ca_cert</code></li><li><code>client_cert</code></li>@@ -14885,11 +16255,11 @@ as an <code>override</code>. For example, say you have a remote whereyou would always like to use the <code>--checksum</code> flag. You couldsupply the <code>--checksum</code> flag to rclone on every command line,but instead you could put this in the config file:</p>-<div class="sourceCode" id="cb683"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb683-1"><a href="#cb683-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[remote]</span></span>-<span id="cb683-2"><a href="#cb683-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> XXX</span></span>-<span id="cb683-3"><a href="#cb683-3" aria-hidden="true" tabindex="-1"></a><span class="dt">...</span></span>-<span id="cb683-4"><a href="#cb683-4" aria-hidden="true" tabindex="-1"></a><span class="dt">global.checksum </span><span class="ot">=</span><span class="st"> </span><span class="kw">true</span></span></code></pre></div>+<div class="sourceCode" id="cb708"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb708-1"><a href="#cb708-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[remote]</span></span>+<span id="cb708-2"><a href="#cb708-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> XXX</span></span>+<span id="cb708-3"><a href="#cb708-3" aria-hidden="true" tabindex="-1"></a><span class="dt">...</span></span>+<span id="cb708-4"><a href="#cb708-4" aria-hidden="true" tabindex="-1"></a><span class="dt">global.checksum </span><span class="ot">=</span><span class="st"> </span><span class="kw">true</span></span></code></pre></div><p>or use it in the connection string<code>remote,global.checksum=true:</code> (or just<code>remote,global.checksum:</code>). This is equivalent to using the@@ -14925,13 +16295,13 @@ shell.</p><h3 id="windows-1">Windows</h3><p>If your names have spaces in you need to put them in <code>"</code>,e.g.</p>-<div class="sourceCode" id="cb686"><pre-class="sourceCode bat"><code class="sourceCode dosbat"><span id="cb686-1"><a href="#cb686-1" aria-hidden="true" tabindex="-1"></a>rclone copy <span class="st">"E:\folder name\folder name\folder name"</span> remote:backup</span></code></pre></div>+<div class="sourceCode" id="cb711"><pre+class="sourceCode bat"><code class="sourceCode dosbat"><span id="cb711-1"><a href="#cb711-1" aria-hidden="true" tabindex="-1"></a>rclone copy <span class="st">"E:\folder name\folder name\folder name"</span> remote:backup</span></code></pre></div><p>If you are using the root directory on its own then don't quote it(see <a href="https://github.com/rclone/rclone/issues/464">#464</a> forwhy), e.g.</p>-<div class="sourceCode" id="cb687"><pre-class="sourceCode bat"><code class="sourceCode dosbat"><span id="cb687-1"><a href="#cb687-1" aria-hidden="true" tabindex="-1"></a>rclone copy E:\ remote:backup</span></code></pre></div>+<div class="sourceCode" id="cb712"><pre+class="sourceCode bat"><code class="sourceCode dosbat"><span id="cb712-1"><a href="#cb712-1" aria-hidden="true" tabindex="-1"></a>rclone copy E:\ remote:backup</span></code></pre></div><h2 id="copying-files-or-directories-with-in-the-names">Copying files ordirectories with <code>:</code> in the names</h2><p>rclone uses <code>:</code> to mark a remote name. This is, however, a@@ -15151,7 +16521,117 @@ will take precedence if supplied in the metadata over reading the<code>Content-Type</code> or modification time of the source object.</p><p>Hashes are not included in system metadata as there is a well definedway of reading those already.</p>-<h2 id="options-99">Options</h2>+<h2 id="performance">Performance</h2>+<p>Rclone's defaults are fairly conservative. They are chosen to work on+small machines, slow links and with cloud providers which rate limit+heavily. If you have plenty of bandwidth, memory and a fast storage+system then rclone can usually be made to run a great deal faster by+adjusting a few flags.</p>+<p>The right values depend on the size of your files, your network and+the storage systems at each end, so treat the numbers below as starting+points. Change one thing at a time and measure the result.</p>+<h3 id="measuring">Measuring</h3>+<p>Use <code>rclone copy</code> with <a+href="#p-progress">-P/--progress</a> to watch the transfer rate while+you are experimenting. For unattended runs log the stats instead, e.g.+<code>--stats 1m --stats-one-line --stats-log-level NOTICE</code>.</p>+<p>A transfer is always limited by something: the network bandwidth, the+rate at which the storage system will accept requests, disk IO, CPU or+memory. Tuning is working out which limit you are hitting and raising it+until you hit the next one. Watch the network, disk, CPU and memory use+of the machine rclone is running on as well as rclone's stats.</p>+<h3 id="lots-of-files">Lots of files</h3>+<p>The time taken to transfer small files is dominated by the round trip+time of the requests rather than by the bandwidth, so the way to make+them go faster is to do more of them at once.</p>+<p><a href="#transfers-int">--transfers</a> sets the number of files+transferred in parallel. The default is 4. Double this until the+transfer rate stops increasing or the storage system starts returning+errors. Values of 32 to 64 are common and values in the hundreds work+well with the large object storage systems (e.g. s3, azureblob,+googlecloudstorage, b2) when transferring lots of small files.</p>+<p><a href="#checkers-int">--checkers</a> sets the number of directory+listings and file comparisons done in parallel. The default is 8.+Raising this speeds up syncs where most of the files are already present+on the destination and stops the checking being the bottleneck when+<code>--transfers</code> is large. 16 to 64 is a reasonable range.</p>+<p>Not all storage systems like lots of parallel requests. If you see+rate limiting errors (e.g. HTTP 429 or 503 errors) or lots of low level+retries in the log then you have gone too far, so reduce+<code>--transfers</code> and <code>--checkers</code> or use <a+href="#tpslimit-float">--tpslimit</a> to cap the request rate. Local+disks, especially spinning ones, and network file systems may get slower+rather than faster with too much parallelism.</p>+<h3 id="large-files">Large files</h3>+<p>A single stream often won't fill a fast link, so rclone can transfer+a large file as several chunks in parallel. These are called+multi-thread transfers and are used for files bigger than <a+href="#multi-thread-cutoff">--multi-thread-cutoff</a> on backends which+support them.</p>+<p>The number of chunks transferred in parallel is set by <a+href="#multi-thread-streams-int">--multi-thread-streams</a> (default 4)+and this works for all backends so increase this first to make large+files transfer faster.</p>+<p>The size of the chunks is set by the backend, usually with a+<code>--backend-chunk-size</code> flag, e.g.+<code>--s3-chunk-size</code>. Backends without a natural chunk size+(e.g. local, smb) use <a+href="#multi-thread-chunk-size-sizesuffix">--multi-thread-chunk-size</a>.+Larger chunks mean fewer requests which increases throughput on big+files. See the docs for the backend you are using for details.</p>+<p>Chunked uploads are usually buffered in memory, so in the worst case+rclone will use</p>+<pre class="text"><code>--transfers * upload concurrency * chunk size</code></pre>+<p>of memory for buffers. For example <code>--transfers 256</code> with+an upload concurrency of 8 and a chunk size of 16 MiB could use up to 32+GiB. When you raise <code>--transfers</code> to speed up small files you+may therefore need to lower the upload concurrency, or set <a+href="#max-buffer-memory">--max-buffer-memory</a> which limits the total+memory used for buffers and makes transfers wait when the limit is+reached.</p>+<p><a href="#use-mmap">--use-mmap</a> makes rclone return buffer memory+to the OS as soon as it has finished with it which is helpful when using+lots of transfers or large chunks.</p>+<h3 id="a-mixture-of-large-and-small-files">A mixture of large and small+files</h3>+<p>By default rclone transfers files in roughly the order it finds them.+With a mixture of file sizes this means that at times all the transfers+are large files, which fill the bandwidth but make few requests, and at+other times all the transfers are small files, which make lots of+requests but use little of the bandwidth.</p>+<p>Using <a href="#order-by-string">--order-by</a>+<code>size,mixed,50</code> makes half of the transfers take the smallest+files in the backlog and half of them take the largest. This keeps both+large and small files moving at once so the bandwidth and the request+rate are both kept full for the whole of the run.</p>+<p>This works best if rclone has lots of files to choose from so use it+with a big <a href="#max-backlog-int">--max-backlog</a>, e.g.+<code>--max-backlog 2000000</code>. Each item in the backlog uses in the+order of 1 KiB of memory.</p>+<h3 id="doing-less-work">Doing less work</h3>+<p>Fewer requests to the storage system means a faster transfer.</p>+<ul>+<li>If the source and destination are on the same remote then rclone+will use <a href="#server-side-copy">server-side copy</a> where it is+supported so the data doesn't need to be downloaded and uploaded+again.</li>+<li><a href="#fast-list">--fast-list</a> lists bucket-based remotes with+fewer transactions at the cost of using more memory. This may be faster+or slower than the default so it is worth testing.</li>+<li><a href="#no-traverse">--no-traverse</a> stops rclone listing the+destination which helps when copying a small number of files into a+large destination.</li>+<li><a href="#size-only">--size-only</a> or <a+href="#c-checksum">--checksum</a> avoid reading the modification time.+On some object storage systems (e.g. s3, swift) this needs an extra+request for each file. Alternatively use <a+href="#use-server-modtime">--use-server-modtime</a> with+<code>--update</code>.</li>+<li><a href="#no-check-dest">--no-check-dest</a> skips checking the+destination entirely. Only use this if you know that none of the files+are there already.</li>+</ul>+<h2 id="options-100">Options</h2><p>Rclone has a number of options to control its behaviour. These aredocumented below, and in the <a href="/flags">flags</a> page.</p><p>Options that take parameters can have the values passed in two ways,@@ -15221,7 +16701,7 @@ valid:</p><li><code>M</code> - Months</li><li><code>y</code> - Years</li></ul>-<p>Examples: "10", "300ms", "-1.5h" or "2h45m".</p>+<p>Examples: "10", "300ms", "-1.5h", "2h45m" or "1d12h".</p><h3 id="size-options">Size options</h3><p>Options that take a SizeSuffix parameter can be specified as aninteger value, which will then be assumed to represent a KiB value@@ -15232,6 +16712,17 @@ and <code>P</code> for PiB. These are the binary units, e.g. 1, 2**10,2**20, 2**30 respectively.</p><p>See also <a href="#human-readable">--human-readable</a>.</p><h2 id="main-options">Main options</h2>+<h3 id="allow-https-downgrade">--allow-https-downgrade</h3>+<p>By default rclone refuses to follow a redirect from an+<code>https://</code> URL to a plain <code>http://</code> URL, as any+credentials or other headers in the request would then be sent+unencrypted, and the response could be altered by anyone on the network+path. The request fails with an error which mentions this flag.</p>+<p>Set <code>--allow-https-downgrade</code> to follow such redirects,+for example when downloading from a mirror which redirects to plain+HTTP. Some backends still refuse to follow them, as doing so would send+their credentials in cleartext.</p>+<p>This option defaults to <code>false</code>.</p><h3 id="backup-dir-string">--backup-dir string</h3><p>When using <ahref="https://rclone.org/commands/rclone_sync/">sync</a>, <a@@ -15325,7 +16816,7 @@ unlimited.</p>week. So this example:</p><pre class="text"><code>--bwlimit "Mon-00:00,512 12:00,1M Sun-20:00,off"</code></pre><p>Is equivalent to this:</p>-<pre class="text"><code>--bwlimit "Mon-00:00,512Mon-12:00,1M Tue-12:00,1M Wed-12:00,1M Thu-12:00,1M Fri-12:00,1M Sat-12:00,1M Sun-12:00,1M Sun-20:00,off"</code></pre>+<pre class="text"><code>--bwlimit "Mon-00:00,512 Mon-12:00,1M Tue-12:00,1M Wed-12:00,1M Thu-12:00,1M Fri-12:00,1M Sat-12:00,1M Sun-12:00,1M Sun-20:00,off"</code></pre><p>Bandwidth limit apply to the data transfer for all backends. For mostbackends the directory listing bandwidth is also included (exceptionsbeing the non HTTP backends, <code>ftp</code>, <code>sftp</code> and@@ -15360,7 +16851,7 @@ effect at the start of the transfer.</p><code>--transfer</code> will use this much memory for buffering.</p><p>When using <code>mount</code> or <code>cmount</code> each open filedescriptor will use this much memory for buffering. See the <a-href="https://rclone.org/commands/rclone_mount/#file-buffering">mount</a>+href="https://rclone.org/commands/rclone_mount/#vfs-file-buffering">mount</a>documentation for more details.</p><p>Set to <code>0</code> to disable the buffering for the minimum memoryusage.</p>@@ -15544,11 +17035,11 @@ value is the internal lowercase name as returned by command<code>rclone help backends</code>. Comments are indicated by<code>;</code> or <code>#</code> at the beginning of a line.</p><p>Example:</p>-<div class="sourceCode" id="cb701"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb701-1"><a href="#cb701-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[megaremote]</span></span>-<span id="cb701-2"><a href="#cb701-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> mega</span></span>-<span id="cb701-3"><a href="#cb701-3" aria-hidden="true" tabindex="-1"></a><span class="dt">user </span><span class="ot">=</span><span class="st"> [email protected]</span></span>-<span id="cb701-4"><a href="#cb701-4" aria-hidden="true" tabindex="-1"></a><span class="dt">pass </span><span class="ot">=</span><span class="st"> PDPcQVVjVtzFY-GTdDFozqBhTdsPg3qH</span></span></code></pre></div>+<div class="sourceCode" id="cb727"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb727-1"><a href="#cb727-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[megaremote]</span></span>+<span id="cb727-2"><a href="#cb727-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> mega</span></span>+<span id="cb727-3"><a href="#cb727-3" aria-hidden="true" tabindex="-1"></a><span class="dt">user </span><span class="ot">=</span><span class="st"> [email protected]</span></span>+<span id="cb727-4"><a href="#cb727-4" aria-hidden="true" tabindex="-1"></a><span class="dt">pass </span><span class="ot">=</span><span class="st"> PDPcQVVjVtzFY-GTdDFozqBhTdsPg3qH</span></span></code></pre></div><p>Note that passwords are in <ahref="https://rclone.org/commands/rclone_obscure/">obscured</a> form.Also, many storage systems uses token-based authentication instead of@@ -15608,8 +17099,8 @@ interactive|skip|first|newest|oldest|largest|smallest|rename|list</h3>href="https://rclone.org/commands/rclone_dedupe/">dedupe</a> command formore information as to what these options mean.</p><h3 id="default-time-time">--default-time Time</h3>-<p>If a file or directory does have a modification time rclone can read-then rclone will display this fixed time instead.</p>+<p>If a file or directory does not have a modification time rclone can+read then rclone will display this fixed time instead.</p><p>The default is <code>2000-01-01 00:00:00 UTC</code>. This can beconfigured in any of the ways shown in <a href="#time-options">timeoptions</a>.</p>@@ -15739,6 +17230,11 @@ downloads use <code>--header-download</code>.</p>supported by <code>--header-upload</code> and<code>--header-download</code> so may be used as a workaround for thosewith care.</p>+<p>The headers are only sent to the host of the original request. If the+server redirects to another host (including a subdomain or a different+port), or redirects an <code>https</code> request to <code>http</code>,+the headers are not sent to it, or to any further hop in that redirect+chain. Rclone logs this at debug level (<code>-vv</code>).</p><pre class="console"><code>rclone ls remote:test --header "X-Rclone: Foo" --header "X-LetMeIn: Yes"</code></pre><h3 id="header-download-stringarray">--header-download stringArray</h3><p>Add an HTTP header for all download transactions. The flag can be@@ -15892,6 +17388,11 @@ the transfer fails.</p><code>--inplace</code> hard links (Unix only) will be broken. And if youdo use <code>--inplace</code> you won't be able to update in useexecutables.</p>+<p>On macOS, if rclone clones a file (see <a+href="https://rclone.org/local/#local-no-clone">--local-no-clone</a>)+and the destination already exists, the existing file is replaced with a+new one, even with <code>--inplace</code>, so hard links to it will be+broken. Use <code>--local-no-clone</code> to keep them.</p><p>Note also that versions of rclone prior to v1.63.0 behave as if the<code>--inplace</code> flag is always supplied.</p><h3 id="interactive">-i, --interactive</h3>@@ -15935,7 +17436,7 @@ empty.</p>like symlinks under Windows). Ignored files won't be copied, moved ordeleted in a sync.</p><p>If you supply this flag then rclone will copy symbolic links from any-supported backend backend, and store them as text files, with a+supported backend, and store them as text files, with a<code>.rclonelink</code> suffix in the destination.</p><p>The text file will contain the target of the symbolic link.</p><p>The <code>--links</code> / <code>-l</code> flag enables this feature@@ -16069,49 +17570,49 @@ complete log file is not strictly valid JSON and needs a parser that canhandle it.</p><p>The JSON logs will be printed on a single line, but are shownexpanded here for clarity.</p>-<div class="sourceCode" id="cb714"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb714-1"><a href="#cb714-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>-<span id="cb714-2"><a href="#cb714-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"time"</span><span class="fu">:</span> <span class="st">"2025-05-13T17:30:51.036237518+01:00"</span><span class="fu">,</span></span>-<span id="cb714-3"><a href="#cb714-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"level"</span><span class="fu">:</span> <span class="st">"debug"</span><span class="fu">,</span></span>-<span id="cb714-4"><a href="#cb714-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"msg"</span><span class="fu">:</span> <span class="st">"4 go routines active</span><span class="ch">\n</span><span class="st">"</span><span class="fu">,</span></span>-<span id="cb714-5"><a href="#cb714-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"source"</span><span class="fu">:</span> <span class="st">"cmd/cmd.go:298"</span></span>-<span id="cb714-6"><a href="#cb714-6" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<div class="sourceCode" id="cb740"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb740-1"><a href="#cb740-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb740-2"><a href="#cb740-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"time"</span><span class="fu">:</span> <span class="st">"2025-05-13T17:30:51.036237518+01:00"</span><span class="fu">,</span></span>+<span id="cb740-3"><a href="#cb740-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"level"</span><span class="fu">:</span> <span class="st">"debug"</span><span class="fu">,</span></span>+<span id="cb740-4"><a href="#cb740-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"msg"</span><span class="fu">:</span> <span class="st">"4 go routines active</span><span class="ch">\n</span><span class="st">"</span><span class="fu">,</span></span>+<span id="cb740-5"><a href="#cb740-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"source"</span><span class="fu">:</span> <span class="st">"cmd/cmd.go:298"</span></span>+<span id="cb740-6"><a href="#cb740-6" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div><p>Completed data transfer logs will have extra <code>size</code>information. Logs which are about a particular object will have<code>object</code> and <code>objectType</code> fields also.</p>-<div class="sourceCode" id="cb715"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb715-1"><a href="#cb715-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>-<span id="cb715-2"><a href="#cb715-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"time"</span><span class="fu">:</span> <span class="st">"2025-05-13T17:38:05.540846352+01:00"</span><span class="fu">,</span></span>-<span id="cb715-3"><a href="#cb715-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"level"</span><span class="fu">:</span> <span class="st">"info"</span><span class="fu">,</span></span>-<span id="cb715-4"><a href="#cb715-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"msg"</span><span class="fu">:</span> <span class="st">"Copied (new) to: file2.txt"</span><span class="fu">,</span></span>-<span id="cb715-5"><a href="#cb715-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"size"</span><span class="fu">:</span> <span class="dv">6</span><span class="fu">,</span></span>-<span id="cb715-6"><a href="#cb715-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">"object"</span><span class="fu">:</span> <span class="st">"file.txt"</span><span class="fu">,</span></span>-<span id="cb715-7"><a href="#cb715-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"objectType"</span><span class="fu">:</span> <span class="st">"*local.Object"</span><span class="fu">,</span></span>-<span id="cb715-8"><a href="#cb715-8" aria-hidden="true" tabindex="-1"></a> <span class="dt">"source"</span><span class="fu">:</span> <span class="st">"operations/copy.go:368"</span></span>-<span id="cb715-9"><a href="#cb715-9" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<div class="sourceCode" id="cb741"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb741-1"><a href="#cb741-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb741-2"><a href="#cb741-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"time"</span><span class="fu">:</span> <span class="st">"2025-05-13T17:38:05.540846352+01:00"</span><span class="fu">,</span></span>+<span id="cb741-3"><a href="#cb741-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"level"</span><span class="fu">:</span> <span class="st">"info"</span><span class="fu">,</span></span>+<span id="cb741-4"><a href="#cb741-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"msg"</span><span class="fu">:</span> <span class="st">"Copied (new) to: file2.txt"</span><span class="fu">,</span></span>+<span id="cb741-5"><a href="#cb741-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"size"</span><span class="fu">:</span> <span class="dv">6</span><span class="fu">,</span></span>+<span id="cb741-6"><a href="#cb741-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">"object"</span><span class="fu">:</span> <span class="st">"file.txt"</span><span class="fu">,</span></span>+<span id="cb741-7"><a href="#cb741-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"objectType"</span><span class="fu">:</span> <span class="st">"*local.Object"</span><span class="fu">,</span></span>+<span id="cb741-8"><a href="#cb741-8" aria-hidden="true" tabindex="-1"></a> <span class="dt">"source"</span><span class="fu">:</span> <span class="st">"operations/copy.go:368"</span></span>+<span id="cb741-9"><a href="#cb741-9" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div><p>Stats logs will contain a <code>stats</code> field which is the sameas returned from the rc call <ahref="https://rclone.org/rc/#core-stats">core/stats</a>.</p>-<div class="sourceCode" id="cb716"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb716-1"><a href="#cb716-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>-<span id="cb716-2"><a href="#cb716-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"time"</span><span class="fu">:</span> <span class="st">"2025-05-13T17:38:05.540912847+01:00"</span><span class="fu">,</span></span>-<span id="cb716-3"><a href="#cb716-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"level"</span><span class="fu">:</span> <span class="st">"info"</span><span class="fu">,</span></span>-<span id="cb716-4"><a href="#cb716-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"msg"</span><span class="fu">:</span> <span class="st">"...text version of the stats..."</span><span class="fu">,</span></span>-<span id="cb716-5"><a href="#cb716-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"stats"</span><span class="fu">:</span> <span class="fu">{</span></span>-<span id="cb716-6"><a href="#cb716-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">"bytes"</span><span class="fu">:</span> <span class="dv">6</span><span class="fu">,</span></span>-<span id="cb716-7"><a href="#cb716-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"checks"</span><span class="fu">:</span> <span class="dv">0</span><span class="fu">,</span></span>-<span id="cb716-8"><a href="#cb716-8" aria-hidden="true" tabindex="-1"></a> <span class="dt">"deletedDirs"</span><span class="fu">:</span> <span class="dv">0</span><span class="fu">,</span></span>-<span id="cb716-9"><a href="#cb716-9" aria-hidden="true" tabindex="-1"></a> <span class="dt">"deletes"</span><span class="fu">:</span> <span class="dv">0</span><span class="fu">,</span></span>-<span id="cb716-10"><a href="#cb716-10" aria-hidden="true" tabindex="-1"></a> <span class="dt">"elapsedTime"</span><span class="fu">:</span> <span class="fl">0.000904825</span><span class="fu">,</span></span>-<span id="cb716-11"><a href="#cb716-11" aria-hidden="true" tabindex="-1"></a> <span class="er">...truncated</span> <span class="er">for</span> <span class="er">clarity...</span></span>-<span id="cb716-12"><a href="#cb716-12" aria-hidden="true" tabindex="-1"></a> <span class="dt">"totalBytes"</span><span class="fu">:</span> <span class="dv">6</span><span class="fu">,</span></span>-<span id="cb716-13"><a href="#cb716-13" aria-hidden="true" tabindex="-1"></a> <span class="dt">"totalChecks"</span><span class="fu">:</span> <span class="dv">0</span><span class="fu">,</span></span>-<span id="cb716-14"><a href="#cb716-14" aria-hidden="true" tabindex="-1"></a> <span class="dt">"totalTransfers"</span><span class="fu">:</span> <span class="dv">1</span><span class="fu">,</span></span>-<span id="cb716-15"><a href="#cb716-15" aria-hidden="true" tabindex="-1"></a> <span class="dt">"transferTime"</span><span class="fu">:</span> <span class="fl">0.000882794</span><span class="fu">,</span></span>-<span id="cb716-16"><a href="#cb716-16" aria-hidden="true" tabindex="-1"></a> <span class="dt">"transfers"</span><span class="fu">:</span> <span class="dv">1</span></span>-<span id="cb716-17"><a href="#cb716-17" aria-hidden="true" tabindex="-1"></a> <span class="fu">},</span></span>-<span id="cb716-18"><a href="#cb716-18" aria-hidden="true" tabindex="-1"></a> <span class="dt">"source"</span><span class="fu">:</span> <span class="st">"accounting/stats.go:569"</span></span>-<span id="cb716-19"><a href="#cb716-19" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<div class="sourceCode" id="cb742"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb742-1"><a href="#cb742-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb742-2"><a href="#cb742-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"time"</span><span class="fu">:</span> <span class="st">"2025-05-13T17:38:05.540912847+01:00"</span><span class="fu">,</span></span>+<span id="cb742-3"><a href="#cb742-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"level"</span><span class="fu">:</span> <span class="st">"info"</span><span class="fu">,</span></span>+<span id="cb742-4"><a href="#cb742-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"msg"</span><span class="fu">:</span> <span class="st">"...text version of the stats..."</span><span class="fu">,</span></span>+<span id="cb742-5"><a href="#cb742-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"stats"</span><span class="fu">:</span> <span class="fu">{</span></span>+<span id="cb742-6"><a href="#cb742-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">"bytes"</span><span class="fu">:</span> <span class="dv">6</span><span class="fu">,</span></span>+<span id="cb742-7"><a href="#cb742-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"checks"</span><span class="fu">:</span> <span class="dv">0</span><span class="fu">,</span></span>+<span id="cb742-8"><a href="#cb742-8" aria-hidden="true" tabindex="-1"></a> <span class="dt">"deletedDirs"</span><span class="fu">:</span> <span class="dv">0</span><span class="fu">,</span></span>+<span id="cb742-9"><a href="#cb742-9" aria-hidden="true" tabindex="-1"></a> <span class="dt">"deletes"</span><span class="fu">:</span> <span class="dv">0</span><span class="fu">,</span></span>+<span id="cb742-10"><a href="#cb742-10" aria-hidden="true" tabindex="-1"></a> <span class="dt">"elapsedTime"</span><span class="fu">:</span> <span class="fl">0.000904825</span><span class="fu">,</span></span>+<span id="cb742-11"><a href="#cb742-11" aria-hidden="true" tabindex="-1"></a> <span class="er">...truncated</span> <span class="er">for</span> <span class="er">clarity...</span></span>+<span id="cb742-12"><a href="#cb742-12" aria-hidden="true" tabindex="-1"></a> <span class="dt">"totalBytes"</span><span class="fu">:</span> <span class="dv">6</span><span class="fu">,</span></span>+<span id="cb742-13"><a href="#cb742-13" aria-hidden="true" tabindex="-1"></a> <span class="dt">"totalChecks"</span><span class="fu">:</span> <span class="dv">0</span><span class="fu">,</span></span>+<span id="cb742-14"><a href="#cb742-14" aria-hidden="true" tabindex="-1"></a> <span class="dt">"totalTransfers"</span><span class="fu">:</span> <span class="dv">1</span><span class="fu">,</span></span>+<span id="cb742-15"><a href="#cb742-15" aria-hidden="true" tabindex="-1"></a> <span class="dt">"transferTime"</span><span class="fu">:</span> <span class="fl">0.000882794</span><span class="fu">,</span></span>+<span id="cb742-16"><a href="#cb742-16" aria-hidden="true" tabindex="-1"></a> <span class="dt">"transfers"</span><span class="fu">:</span> <span class="dv">1</span></span>+<span id="cb742-17"><a href="#cb742-17" aria-hidden="true" tabindex="-1"></a> <span class="fu">},</span></span>+<span id="cb742-18"><a href="#cb742-18" aria-hidden="true" tabindex="-1"></a> <span class="dt">"source"</span><span class="fu">:</span> <span class="st">"accounting/stats.go:569"</span></span>+<span id="cb742-19"><a href="#cb742-19" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div><h3 id="low-level-retries-int">--low-level-retries int</h3><p>This controls the number of low level retries rclone does.</p><p>A low level retry is used to retry a failing operation - typically@@ -16266,63 +17767,63 @@ known.</li><li><code>Metadata</code> is the backend specific metadata as describedin the backend docs.</li></ul>-<div class="sourceCode" id="cb718"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb718-1"><a href="#cb718-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>-<span id="cb718-2"><a href="#cb718-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"SrcFs"</span><span class="fu">:</span> <span class="st">"gdrive:"</span><span class="fu">,</span></span>-<span id="cb718-3"><a href="#cb718-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"SrcFsType"</span><span class="fu">:</span> <span class="st">"drive"</span><span class="fu">,</span></span>-<span id="cb718-4"><a href="#cb718-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"DstFs"</span><span class="fu">:</span> <span class="st">"newdrive:user"</span><span class="fu">,</span></span>-<span id="cb718-5"><a href="#cb718-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"DstFsType"</span><span class="fu">:</span> <span class="st">"onedrive"</span><span class="fu">,</span></span>-<span id="cb718-6"><a href="#cb718-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Remote"</span><span class="fu">:</span> <span class="st">"test.txt"</span><span class="fu">,</span></span>-<span id="cb718-7"><a href="#cb718-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Size"</span><span class="fu">:</span> <span class="dv">6</span><span class="fu">,</span></span>-<span id="cb718-8"><a href="#cb718-8" aria-hidden="true" tabindex="-1"></a> <span class="dt">"MimeType"</span><span class="fu">:</span> <span class="st">"text/plain; charset=utf-8"</span><span class="fu">,</span></span>-<span id="cb718-9"><a href="#cb718-9" aria-hidden="true" tabindex="-1"></a> <span class="dt">"ModTime"</span><span class="fu">:</span> <span class="st">"2022-10-11T17:53:10.286745272+01:00"</span><span class="fu">,</span></span>-<span id="cb718-10"><a href="#cb718-10" aria-hidden="true" tabindex="-1"></a> <span class="dt">"IsDir"</span><span class="fu">:</span> <span class="kw">false</span><span class="fu">,</span></span>-<span id="cb718-11"><a href="#cb718-11" aria-hidden="true" tabindex="-1"></a> <span class="dt">"ID"</span><span class="fu">:</span> <span class="st">"xyz"</span><span class="fu">,</span></span>-<span id="cb718-12"><a href="#cb718-12" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Metadata"</span><span class="fu">:</span> <span class="fu">{</span></span>-<span id="cb718-13"><a href="#cb718-13" aria-hidden="true" tabindex="-1"></a> <span class="dt">"btime"</span><span class="fu">:</span> <span class="st">"2022-10-11T16:53:11Z"</span><span class="fu">,</span></span>-<span id="cb718-14"><a href="#cb718-14" aria-hidden="true" tabindex="-1"></a> <span class="dt">"content-type"</span><span class="fu">:</span> <span class="st">"text/plain; charset=utf-8"</span><span class="fu">,</span></span>-<span id="cb718-15"><a href="#cb718-15" aria-hidden="true" tabindex="-1"></a> <span class="dt">"mtime"</span><span class="fu">:</span> <span class="st">"2022-10-11T17:53:10.286745272+01:00"</span><span class="fu">,</span></span>-<span id="cb718-16"><a href="#cb718-16" aria-hidden="true" tabindex="-1"></a> <span class="dt">"owner"</span><span class="fu">:</span> <span class="st">"[email protected]"</span><span class="fu">,</span></span>-<span id="cb718-17"><a href="#cb718-17" aria-hidden="true" tabindex="-1"></a> <span class="dt">"permissions"</span><span class="fu">:</span> <span class="st">"..."</span><span class="fu">,</span></span>-<span id="cb718-18"><a href="#cb718-18" aria-hidden="true" tabindex="-1"></a> <span class="dt">"description"</span><span class="fu">:</span> <span class="st">"my nice file"</span><span class="fu">,</span></span>-<span id="cb718-19"><a href="#cb718-19" aria-hidden="true" tabindex="-1"></a> <span class="dt">"starred"</span><span class="fu">:</span> <span class="st">"false"</span></span>-<span id="cb718-20"><a href="#cb718-20" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>-<span id="cb718-21"><a href="#cb718-21" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<div class="sourceCode" id="cb744"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb744-1"><a href="#cb744-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb744-2"><a href="#cb744-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"SrcFs"</span><span class="fu">:</span> <span class="st">"gdrive:"</span><span class="fu">,</span></span>+<span id="cb744-3"><a href="#cb744-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"SrcFsType"</span><span class="fu">:</span> <span class="st">"drive"</span><span class="fu">,</span></span>+<span id="cb744-4"><a href="#cb744-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"DstFs"</span><span class="fu">:</span> <span class="st">"newdrive:user"</span><span class="fu">,</span></span>+<span id="cb744-5"><a href="#cb744-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"DstFsType"</span><span class="fu">:</span> <span class="st">"onedrive"</span><span class="fu">,</span></span>+<span id="cb744-6"><a href="#cb744-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Remote"</span><span class="fu">:</span> <span class="st">"test.txt"</span><span class="fu">,</span></span>+<span id="cb744-7"><a href="#cb744-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Size"</span><span class="fu">:</span> <span class="dv">6</span><span class="fu">,</span></span>+<span id="cb744-8"><a href="#cb744-8" aria-hidden="true" tabindex="-1"></a> <span class="dt">"MimeType"</span><span class="fu">:</span> <span class="st">"text/plain; charset=utf-8"</span><span class="fu">,</span></span>+<span id="cb744-9"><a href="#cb744-9" aria-hidden="true" tabindex="-1"></a> <span class="dt">"ModTime"</span><span class="fu">:</span> <span class="st">"2022-10-11T17:53:10.286745272+01:00"</span><span class="fu">,</span></span>+<span id="cb744-10"><a href="#cb744-10" aria-hidden="true" tabindex="-1"></a> <span class="dt">"IsDir"</span><span class="fu">:</span> <span class="kw">false</span><span class="fu">,</span></span>+<span id="cb744-11"><a href="#cb744-11" aria-hidden="true" tabindex="-1"></a> <span class="dt">"ID"</span><span class="fu">:</span> <span class="st">"xyz"</span><span class="fu">,</span></span>+<span id="cb744-12"><a href="#cb744-12" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Metadata"</span><span class="fu">:</span> <span class="fu">{</span></span>+<span id="cb744-13"><a href="#cb744-13" aria-hidden="true" tabindex="-1"></a> <span class="dt">"btime"</span><span class="fu">:</span> <span class="st">"2022-10-11T16:53:11Z"</span><span class="fu">,</span></span>+<span id="cb744-14"><a href="#cb744-14" aria-hidden="true" tabindex="-1"></a> <span class="dt">"content-type"</span><span class="fu">:</span> <span class="st">"text/plain; charset=utf-8"</span><span class="fu">,</span></span>+<span id="cb744-15"><a href="#cb744-15" aria-hidden="true" tabindex="-1"></a> <span class="dt">"mtime"</span><span class="fu">:</span> <span class="st">"2022-10-11T17:53:10.286745272+01:00"</span><span class="fu">,</span></span>+<span id="cb744-16"><a href="#cb744-16" aria-hidden="true" tabindex="-1"></a> <span class="dt">"owner"</span><span class="fu">:</span> <span class="st">"[email protected]"</span><span class="fu">,</span></span>+<span id="cb744-17"><a href="#cb744-17" aria-hidden="true" tabindex="-1"></a> <span class="dt">"permissions"</span><span class="fu">:</span> <span class="st">"..."</span><span class="fu">,</span></span>+<span id="cb744-18"><a href="#cb744-18" aria-hidden="true" tabindex="-1"></a> <span class="dt">"description"</span><span class="fu">:</span> <span class="st">"my nice file"</span><span class="fu">,</span></span>+<span id="cb744-19"><a href="#cb744-19" aria-hidden="true" tabindex="-1"></a> <span class="dt">"starred"</span><span class="fu">:</span> <span class="st">"false"</span></span>+<span id="cb744-20"><a href="#cb744-20" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>+<span id="cb744-21"><a href="#cb744-21" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div><p>The program should then modify the input as desired and send it toSTDOUT. The returned <code>Metadata</code> field will be used in itsentirety for the destination object. Any other fields will be ignored.Note in this example we translate user names and permissions and addsomething to the description:</p>-<div class="sourceCode" id="cb719"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb719-1"><a href="#cb719-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>-<span id="cb719-2"><a href="#cb719-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Metadata"</span><span class="fu">:</span> <span class="fu">{</span></span>-<span id="cb719-3"><a href="#cb719-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"btime"</span><span class="fu">:</span> <span class="st">"2022-10-11T16:53:11Z"</span><span class="fu">,</span></span>-<span id="cb719-4"><a href="#cb719-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"content-type"</span><span class="fu">:</span> <span class="st">"text/plain; charset=utf-8"</span><span class="fu">,</span></span>-<span id="cb719-5"><a href="#cb719-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"mtime"</span><span class="fu">:</span> <span class="st">"2022-10-11T17:53:10.286745272+01:00"</span><span class="fu">,</span></span>-<span id="cb719-6"><a href="#cb719-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">"owner"</span><span class="fu">:</span> <span class="st">"[email protected]"</span><span class="fu">,</span></span>-<span id="cb719-7"><a href="#cb719-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"permissions"</span><span class="fu">:</span> <span class="st">"..."</span><span class="fu">,</span></span>-<span id="cb719-8"><a href="#cb719-8" aria-hidden="true" tabindex="-1"></a> <span class="dt">"description"</span><span class="fu">:</span> <span class="st">"my nice file [migrated from domain1]"</span><span class="fu">,</span></span>-<span id="cb719-9"><a href="#cb719-9" aria-hidden="true" tabindex="-1"></a> <span class="dt">"starred"</span><span class="fu">:</span> <span class="st">"false"</span></span>-<span id="cb719-10"><a href="#cb719-10" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>-<span id="cb719-11"><a href="#cb719-11" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<div class="sourceCode" id="cb745"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb745-1"><a href="#cb745-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb745-2"><a href="#cb745-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Metadata"</span><span class="fu">:</span> <span class="fu">{</span></span>+<span id="cb745-3"><a href="#cb745-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"btime"</span><span class="fu">:</span> <span class="st">"2022-10-11T16:53:11Z"</span><span class="fu">,</span></span>+<span id="cb745-4"><a href="#cb745-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"content-type"</span><span class="fu">:</span> <span class="st">"text/plain; charset=utf-8"</span><span class="fu">,</span></span>+<span id="cb745-5"><a href="#cb745-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"mtime"</span><span class="fu">:</span> <span class="st">"2022-10-11T17:53:10.286745272+01:00"</span><span class="fu">,</span></span>+<span id="cb745-6"><a href="#cb745-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">"owner"</span><span class="fu">:</span> <span class="st">"[email protected]"</span><span class="fu">,</span></span>+<span id="cb745-7"><a href="#cb745-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"permissions"</span><span class="fu">:</span> <span class="st">"..."</span><span class="fu">,</span></span>+<span id="cb745-8"><a href="#cb745-8" aria-hidden="true" tabindex="-1"></a> <span class="dt">"description"</span><span class="fu">:</span> <span class="st">"my nice file [migrated from domain1]"</span><span class="fu">,</span></span>+<span id="cb745-9"><a href="#cb745-9" aria-hidden="true" tabindex="-1"></a> <span class="dt">"starred"</span><span class="fu">:</span> <span class="st">"false"</span></span>+<span id="cb745-10"><a href="#cb745-10" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>+<span id="cb745-11"><a href="#cb745-11" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div><p>Metadata can be removed here too.</p><p>An example python program might look something like this to implementthe above transformations.</p>-<div class="sourceCode" id="cb720"><pre-class="sourceCode python"><code class="sourceCode python"><span id="cb720-1"><a href="#cb720-1" aria-hidden="true" tabindex="-1"></a><span class="im">import</span> sys, json</span>-<span id="cb720-2"><a href="#cb720-2" aria-hidden="true" tabindex="-1"></a></span>-<span id="cb720-3"><a href="#cb720-3" aria-hidden="true" tabindex="-1"></a>i <span class="op">=</span> json.load(sys.stdin)</span>-<span id="cb720-4"><a href="#cb720-4" aria-hidden="true" tabindex="-1"></a>metadata <span class="op">=</span> i[<span class="st">"Metadata"</span>]</span>-<span id="cb720-5"><a href="#cb720-5" aria-hidden="true" tabindex="-1"></a><span class="co"># Add tag to description</span></span>-<span id="cb720-6"><a href="#cb720-6" aria-hidden="true" tabindex="-1"></a><span class="cf">if</span> <span class="st">"description"</span> <span class="kw">in</span> metadata:</span>-<span id="cb720-7"><a href="#cb720-7" aria-hidden="true" tabindex="-1"></a> metadata[<span class="st">"description"</span>] <span class="op">+=</span> <span class="st">" [migrated from domain1]"</span></span>-<span id="cb720-8"><a href="#cb720-8" aria-hidden="true" tabindex="-1"></a><span class="cf">else</span>:</span>-<span id="cb720-9"><a href="#cb720-9" aria-hidden="true" tabindex="-1"></a> metadata[<span class="st">"description"</span>] <span class="op">=</span> <span class="st">"[migrated from domain1]"</span></span>-<span id="cb720-10"><a href="#cb720-10" aria-hidden="true" tabindex="-1"></a><span class="co"># Modify owner</span></span>-<span id="cb720-11"><a href="#cb720-11" aria-hidden="true" tabindex="-1"></a><span class="cf">if</span> <span class="st">"owner"</span> <span class="kw">in</span> metadata:</span>-<span id="cb720-12"><a href="#cb720-12" aria-hidden="true" tabindex="-1"></a> metadata[<span class="st">"owner"</span>] <span class="op">=</span> metadata[<span class="st">"owner"</span>].replace(<span class="st">"domain1.com"</span>, <span class="st">"domain2.com"</span>)</span>-<span id="cb720-13"><a href="#cb720-13" aria-hidden="true" tabindex="-1"></a>o <span class="op">=</span> { <span class="st">"Metadata"</span>: metadata }</span>-<span id="cb720-14"><a href="#cb720-14" aria-hidden="true" tabindex="-1"></a>json.dump(o, sys.stdout, indent<span class="op">=</span><span class="st">"</span><span class="ch">\t</span><span class="st">"</span>)</span></code></pre></div>+<div class="sourceCode" id="cb746"><pre+class="sourceCode python"><code class="sourceCode python"><span id="cb746-1"><a href="#cb746-1" aria-hidden="true" tabindex="-1"></a><span class="im">import</span> sys, json</span>+<span id="cb746-2"><a href="#cb746-2" aria-hidden="true" tabindex="-1"></a></span>+<span id="cb746-3"><a href="#cb746-3" aria-hidden="true" tabindex="-1"></a>i <span class="op">=</span> json.load(sys.stdin)</span>+<span id="cb746-4"><a href="#cb746-4" aria-hidden="true" tabindex="-1"></a>metadata <span class="op">=</span> i[<span class="st">"Metadata"</span>]</span>+<span id="cb746-5"><a href="#cb746-5" aria-hidden="true" tabindex="-1"></a><span class="co"># Add tag to description</span></span>+<span id="cb746-6"><a href="#cb746-6" aria-hidden="true" tabindex="-1"></a><span class="cf">if</span> <span class="st">"description"</span> <span class="kw">in</span> metadata:</span>+<span id="cb746-7"><a href="#cb746-7" aria-hidden="true" tabindex="-1"></a> metadata[<span class="st">"description"</span>] <span class="op">+=</span> <span class="st">" [migrated from domain1]"</span></span>+<span id="cb746-8"><a href="#cb746-8" aria-hidden="true" tabindex="-1"></a><span class="cf">else</span>:</span>+<span id="cb746-9"><a href="#cb746-9" aria-hidden="true" tabindex="-1"></a> metadata[<span class="st">"description"</span>] <span class="op">=</span> <span class="st">"[migrated from domain1]"</span></span>+<span id="cb746-10"><a href="#cb746-10" aria-hidden="true" tabindex="-1"></a><span class="co"># Modify owner</span></span>+<span id="cb746-11"><a href="#cb746-11" aria-hidden="true" tabindex="-1"></a><span class="cf">if</span> <span class="st">"owner"</span> <span class="kw">in</span> metadata:</span>+<span id="cb746-12"><a href="#cb746-12" aria-hidden="true" tabindex="-1"></a> metadata[<span class="st">"owner"</span>] <span class="op">=</span> metadata[<span class="st">"owner"</span>].replace(<span class="st">"domain1.com"</span>, <span class="st">"domain2.com"</span>)</span>+<span id="cb746-13"><a href="#cb746-13" aria-hidden="true" tabindex="-1"></a>o <span class="op">=</span> { <span class="st">"Metadata"</span>: metadata }</span>+<span id="cb746-14"><a href="#cb746-14" aria-hidden="true" tabindex="-1"></a>json.dump(o, sys.stdout, indent<span class="op">=</span><span class="st">"</span><span class="ch">\t</span><span class="st">"</span>)</span></code></pre></div><p>You can find this example (slightly expanded) in the rclone sourcecode at <ahref="https://github.com/rclone/rclone/blob/master/bin/test_metadata_mapper.py">bin/test_metadata_mapper.py</a>.</p>@@ -16377,7 +17878,7 @@ rclone will use multiple threads to transfer the file (defaulthref="https://rclone.org/overview/#optional-features">overview</a> as<code>MultithreadUpload</code>. (They need to implement either the<code>OpenWriterAt</code> or <code>OpenChunkWriter</code> internal-interfaces). These include include, <code>local</code>, <code>s3</code>,+interfaces). These include <code>local</code>, <code>s3</code>,<code>azureblob</code>, <code>b2</code>,<code>oracleobjectstorage</code> and <code>smb</code> at the time ofwriting.</p>@@ -16405,7 +17906,7 @@ at maximum <code>--transfers</code> *<code>--multi-thread-chunk-size</code> *<code>--multi-thread-streams</code> or specifically for the s3 backend<code>--transfers</code> * <code>--s3-chunk-size</code> *-<code>--s3-concurrency</code>. However you can use the the <a+<code>--s3-concurrency</code>. However you can use the <ahref="https://rclone.org/docs/#max-buffer-memory">--max-buffer-memory</a>flag to control the maximum memory used here.</p><p><strong>NB</strong> that this <strong>only</strong> works with@@ -16509,8 +18010,10 @@ describes what aspect is being measured:</p><li><code>name</code> - order by the full path of the files</li><li><code>modtime</code> - order by the modification date of thefiles</li>+<li><code>pattern</code> - order by a list of path glob patterns</li></ul>-<p>This can have a modifier appended with a comma:</p>+<p>The <code>size</code>, <code>name</code> and <code>modtime</code>+comparisons can have a modifier appended with a comma:</p><ul><li><code>ascending</code> or <code>asc</code> - order so that thesmallest (or oldest) is processed first</li>@@ -16530,6 +18033,13 @@ and small files - the large files are guaranteed upload threads andbandwidth and the small files will be processed continuously.</p><p>If no modifier is supplied then the order is<code>ascending</code>.</p>+<p>For <code>pattern</code>, append one or more comma-separated+patterns. Patterns use the <a+href="https://rclone.org/filtering/#patterns">filter pattern syntax</a>.+Files matching the first pattern are processed first, followed by files+matching the second pattern, and so on. Files which do not match a+pattern are processed last. Within each group, files are ordered by full+path. Patterns cannot contain a comma.</p><p>For example</p><ul><li><code>--order-by size,desc</code> - send the largest files@@ -16538,6 +18048,8 @@ first</li>first</li><li><code>--order-by name</code> - send the files with alphabetically bypath first</li>+<li><code>--order-by 'pattern,*.tar.gz,*.md5'</code> - send archives+before checksum files</li></ul><p>If the <code>--order-by</code> flag is not supplied or it is suppliedwith an empty string then the default ordering will be used which is as@@ -16576,12 +18088,15 @@ when setting the config password for the first time.</p>of arguments. If one of the arguments has a space in then enclose it in<code>"</code>, if you want a literal <code>"</code> in an argument thenenclose the argument in <code>"</code> and double the <code>"</code>.-See <a href="https://godoc.org/encoding/csv">CSV encoding</a> for more+This includes the command itself: if the path to the executable contains+a space, it must be quoted too. See <a+href="https://godoc.org/encoding/csv">CSV encoding</a> for moreinfo.</p><p>Eg</p><pre class="text"><code>--password-command "echo hello"--password-command 'echo "hello with space"'---password-command 'echo "hello with ""quotes"" and space"'</code></pre>+--password-command 'echo "hello with ""quotes"" and space"'+--password-command '"/path with a space/get-password.sh"'</code></pre><p>Note that when changing the configuration password the environmentvariable <code>RCLONE_PASSWORD_CHANGE=1</code> will be set. This can beused to distinguish initial decryption of the config file from the new@@ -17137,11 +18652,11 @@ password, in which case it will be used for decrypting theconfiguration.</p><p>You can set this for a session from a script. For unix like systemssave this to a file called <code>set-rclone-password</code>:</p>-<div class="sourceCode" id="cb726"><pre-class="sourceCode sh"><code class="sourceCode bash"><span id="cb726-1"><a href="#cb726-1" aria-hidden="true" tabindex="-1"></a><span class="co">#!/bin/echo Source this file don't run it</span></span>-<span id="cb726-2"><a href="#cb726-2" aria-hidden="true" tabindex="-1"></a></span>-<span id="cb726-3"><a href="#cb726-3" aria-hidden="true" tabindex="-1"></a><span class="bu">read</span> <span class="at">-s</span> <span class="va">RCLONE_CONFIG_PASS</span></span>-<span id="cb726-4"><a href="#cb726-4" aria-hidden="true" tabindex="-1"></a><span class="bu">export</span> <span class="va">RCLONE_CONFIG_PASS</span></span></code></pre></div>+<div class="sourceCode" id="cb752"><pre+class="sourceCode sh"><code class="sourceCode bash"><span id="cb752-1"><a href="#cb752-1" aria-hidden="true" tabindex="-1"></a><span class="co">#!/bin/echo Source this file don't run it</span></span>+<span id="cb752-2"><a href="#cb752-2" aria-hidden="true" tabindex="-1"></a></span>+<span id="cb752-3"><a href="#cb752-3" aria-hidden="true" tabindex="-1"></a><span class="bu">read</span> <span class="at">-s</span> <span class="va">RCLONE_CONFIG_PASS</span></span>+<span id="cb752-4"><a href="#cb752-4" aria-hidden="true" tabindex="-1"></a><span class="bu">export</span> <span class="va">RCLONE_CONFIG_PASS</span></span></code></pre></div><p>Then source the file when you want to use it. From the shell youwould do <code>source set-rclone-password</code>. It will then ask youfor the password and set it in the environment variable.</p>@@ -17213,11 +18728,11 @@ a password store: <code>pass init rclone</code>.</p></li><h4 id="windows-2">Windows</h4><ul><li><p>Generate and store a password</p>-<div class="sourceCode" id="cb732"><pre-class="sourceCode powershell"><code class="sourceCode powershell"><span id="cb732-1"><a href="#cb732-1" aria-hidden="true" tabindex="-1"></a><span class="fu">New-Object</span> <span class="op">-</span>TypeName PSCredential <span class="op">-</span>ArgumentList <span class="st">"rclone"</span><span class="op">,</span> <span class="op">(</span><span class="fu">ConvertTo-SecureString</span> <span class="op">-</span>String <span class="op">([</span>System<span class="op">.</span><span class="fu">Web</span><span class="op">.</span><span class="fu">Security</span><span class="op">.</span><span class="fu">Membership</span><span class="op">]::</span>GeneratePassword<span class="op">(</span>40<span class="op">,</span> 10<span class="op">))</span> <span class="op">-</span>AsPlainText <span class="op">-</span>Force<span class="op">)</span> <span class="op">|</span> <span class="fu">Export-Clixml</span> <span class="op">-</span>Path <span class="st">"rclone-credential.xml"</span></span></code></pre></div></li>+<div class="sourceCode" id="cb758"><pre+class="sourceCode powershell"><code class="sourceCode powershell"><span id="cb758-1"><a href="#cb758-1" aria-hidden="true" tabindex="-1"></a><span class="fu">New-Object</span> <span class="op">-</span>TypeName PSCredential <span class="op">-</span>ArgumentList <span class="st">"rclone"</span><span class="op">,</span> <span class="op">(</span><span class="fu">ConvertTo-SecureString</span> <span class="op">-</span>String <span class="op">([</span>System<span class="op">.</span><span class="fu">Web</span><span class="op">.</span><span class="fu">Security</span><span class="op">.</span><span class="fu">Membership</span><span class="op">]::</span>GeneratePassword<span class="op">(</span>40<span class="op">,</span> 10<span class="op">))</span> <span class="op">-</span>AsPlainText <span class="op">-</span>Force<span class="op">)</span> <span class="op">|</span> <span class="fu">Export-Clixml</span> <span class="op">-</span>Path <span class="st">"rclone-credential.xml"</span></span></code></pre></div></li><li><p>Add the password retrieval instruction</p>-<div class="sourceCode" id="cb733"><pre-class="sourceCode powershell"><code class="sourceCode powershell"><span id="cb733-1"><a href="#cb733-1" aria-hidden="true" tabindex="-1"></a><span class="op">[</span>Environment<span class="op">]::</span>SetEnvironmentVariable<span class="op">(</span><span class="st">"RCLONE_PASSWORD_COMMAND"</span><span class="op">,</span> <span class="st">"[System.Runtime.InteropServices.Marshal]::PtrToStringAuto([System.Runtime.InteropServices.Marshal]::SecureStringToBSTR((Import-Clixml -Path "</span>rclone<span class="op">-</span>credential<span class="op">.</span><span class="fu">xml</span><span class="st">").Password))"</span><span class="op">)</span></span></code></pre></div></li>+<div class="sourceCode" id="cb759"><pre+class="sourceCode powershell"><code class="sourceCode powershell"><span id="cb759-1"><a href="#cb759-1" aria-hidden="true" tabindex="-1"></a><span class="op">[</span>Environment<span class="op">]::</span>SetEnvironmentVariable<span class="op">(</span><span class="st">"RCLONE_PASSWORD_COMMAND"</span><span class="op">,</span> <span class="st">"[System.Runtime.InteropServices.Marshal]::PtrToStringAuto([System.Runtime.InteropServices.Marshal]::SecureStringToBSTR((Import-Clixml -Path "</span>rclone<span class="op">-</span>credential<span class="op">.</span><span class="fu">xml</span><span class="st">").Password))"</span><span class="op">)</span></span></code></pre></div></li></ul><h4 id="encrypt-the-config-file-all-systems">Encrypt the config file(all systems)</h4>@@ -17423,7 +18938,7 @@ reached</li><h2 id="environment-variables">Environment variables</h2><p>Rclone can be configured entirely using environment variables. Thesecan be used to set defaults for options or config file entries.</p>-<h3 id="options-100">Options</h3>+<h3 id="options-101">Options</h3><p>Every option in rclone can have its default set by environmentvariable.</p><p>To find the name of the environment variable, first, take the long@@ -17488,10 +19003,14 @@ remote basis. The names of the config items are documented in the pagefor each backend.</p><p>To find the name of the environment variable, you need to set, take<code>RCLONE_CONFIG_</code> + name of remote + <code>_</code> + name of-config file option and make it all uppercase. Note one implication here-is the remote's name must be convertible into a valid environment-variable name, so it can only contain letters, digits, or the-<code>_</code> (underscore) character.</p>+config file option and make it all uppercase. Note that the name of the+remote is only converted to uppercase - characters such as+<code>-</code> and <code>.</code> are not changed to <code>_</code> - so+a remote called <code>my-remote</code> needs+<code>RCLONE_CONFIG_MY-REMOTE_TYPE</code>. Many shells only allow+letters, digits and <code>_</code> (underscore) in environment variable+names, so it is best to use only those characters in the name of a+remote you want to configure this way.</p><p>For example, to configure an S3 remote named <code>mys3:</code>without a config file (using unix ways of setting environmentvariables):</p>@@ -17712,7 +19231,7 @@ href="#examples">examples</a> are below.</p><pre class="text"><code>* matches any sequence of non-separator (/) characters** matches any sequence of characters including / separators? matches any single non-separator (/) character-[ [ ! ] { character-range } ]+[ [ ! | ^ ] { character-range } ]character class (must be non-empty){ pattern-list }pattern alternatives@@ -18648,6 +20167,9 @@ will open.</li><li>The URL already contains the username & password, in which casethe GUI will use those values and log you in automatically.</li></ul>+<p>The rc address is passed to the GUI via the <code>url</code> query+parameter (the URL printed in the terminal); append+<code>?url=http://...</code> to point it at a different rc server.</p><h2 id="security-1">Security</h2><p>It's important to think first about what rclone has access to andwhat you might be sharing.</p>@@ -18668,7 +20190,7 @@ different ports.</li><p>If you want to access it remotely but want to avoid running a proxyand exposing ports, you can use Cloudflare Tunnels or localhost.run orTailscale (all free).</p>-<h2 id="options-101">Options</h2>+<h2 id="options-102">Options</h2><pre class="console"><code> --addr stringArray IPaddress:Port for the GUI server (default auto-chosen localhost port)--api-addr stringArray IPaddress:Port for the RC API server (default auto-chosen localhost port)--enable-metrics Enable OpenMetrics/Prometheus compatible endpoint at /metrics@@ -18727,8 +20249,7 @@ control.</strong> The default bind is loopbackreason to change it.</li><li><strong>Do not use <code>--rc-no-auth</code> on a non-loopbackbind.</strong> It disables authentication on the endpoints that access-remotes — see <a-href="#--rc-no-auth"><code>--rc-no-auth</code></a>.</li>+remotes — see <a href="#rc-no-auth"><code>--rc-no-auth</code></a>.</li><li><strong>Use authentication and TLS</strong>(<code>--rc-user</code>/<code>--rc-pass</code> or<code>--rc-htpasswd</code>, plus@@ -18820,9 +20341,18 @@ rclone.</p><p>Default Off.</p><h3 id="rc-allow-origin">--rc-allow-origin</h3><p>Set the allowed Access-Control-Allow-Origin for rc requests.</p>-<p>Can be used with --rc-web-gui if the rclone is running on different-IP than the web-gui.</p>-<p>Default is IP address on which rc is running.</p>+<p>This enables cross-origin (CORS) requests from browsers on the origin+given, for example+<code>--rc-allow-origin https://app.example.com</code>. Browser CORS+preflight OPTIONS requests are answered by rclone without authentication+as browsers never send credentials with them. All other requests,+including other OPTIONS requests, must be authenticated when+authentication is enabled.</p>+<p>The <code>rclone gui</code> command sets this automatically to the+origin the GUI is served from if it is not set explicitly.</p>+<p>Setting this to <code>*</code> allows any origin which is a security+risk on an authenticated server.</p>+<p>Default is off (no CORS).</p><h3 id="rc-web-fetch-url">--rc-web-fetch-url</h3><p>Set the URL to fetch the rclone-web-gui files from.</p><p>Default <a@@ -18854,6 +20384,8 @@ creating a remote as is <code>sync/copy</code>.</p><p>If this is set then no authorisation will be required on the serverto use these methods. The alternative is to use <code>--rc-user</code>and <code>--rc-pass</code> and use these credentials in the request.</p>+<p>Without authorisation or this flag, POST request bodies are limited+to 64 KiB.</p><p>Default Off.</p><h3 id="rc-baseurl">--rc-baseurl</h3><p>Prefix for URLs.</p>@@ -18902,6 +20434,25 @@ simplifies the command line.</p><pre class="console"><code>rclone rc operations/list fs=/tmp remote=test opt='{"showHash": true}'</code></pre><p>Rather than</p><pre class="console"><code>rclone rc operations/list --json '{"fs": "/tmp", "remote": "test", "opt": {"showHash": true}}'</code></pre>+<h3 id="paths-on-windows">Paths on Windows</h3>+<p>Prefer forward slashes (<code>/</code>) in Windows paths passed to+the rc API, for example <code>C:/path1/path2</code>. If you use+backslashes (<code>\</code>) in a JSON string, escape each one as+<code>\\</code>. These JSON objects contain equivalent Windows+paths:</p>+<div class="sourceCode" id="cb823"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb823-1"><a href="#cb823-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span><span class="dt">"remote"</span><span class="fu">:</span> <span class="st">"C:/path1/path2"</span><span class="fu">}</span></span></code></pre></div>+<div class="sourceCode" id="cb824"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb824-1"><a href="#cb824-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span><span class="dt">"remote"</span><span class="fu">:</span> <span class="st">"C:</span><span class="ch">\\</span><span class="st">path1</span><span class="ch">\\</span><span class="st">path2"</span><span class="fu">}</span></span></code></pre></div>+<p>This applies to JSON request bodies and to JSON-encoded object+parameters, such as <code>parameters</code> in+<code>config/create</code>, even when that JSON is sent in a query+string or form data. An unescaped backslash can cause a JSON parsing+error or be interpreted as an escape sequence, such as <code>\t</code>+for a tab.</p>+<p>String parameters sent directly in a query string or form data,+without JSON encoding, do not need JSON escaping. Use your HTTP client's+URL encoding for query strings and form data.</p><h2 id="special-parameters">Special parameters</h2><p>The rc interface supports some special parameters which apply to<strong>all</strong> commands. These start with <code>_</code> to show@@ -19006,11 +20557,11 @@ href="#option-blocks">the options blocks section</a> for more info).</p><p>For example, if you wished to run a sync with the<code>--checksum</code> parameter, you would pass this parameter in yourJSON blob.</p>-<div class="sourceCode" id="cb802"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb802-1"><a href="#cb802-1" aria-hidden="true" tabindex="-1"></a><span class="er">"_config":</span><span class="fu">{</span><span class="dt">"CheckSum"</span><span class="fu">:</span> <span class="kw">true</span><span class="fu">}</span></span></code></pre></div>+<div class="sourceCode" id="cb830"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb830-1"><a href="#cb830-1" aria-hidden="true" tabindex="-1"></a><span class="er">"_config":</span><span class="fu">{</span><span class="dt">"CheckSum"</span><span class="fu">:</span> <span class="kw">true</span><span class="fu">}</span></span></code></pre></div><p>Or pass it flat at the top level:</p>-<div class="sourceCode" id="cb803"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb803-1"><a href="#cb803-1" aria-hidden="true" tabindex="-1"></a><span class="er">"checksum":</span> <span class="er">true</span></span></code></pre></div>+<div class="sourceCode" id="cb831"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb831-1"><a href="#cb831-1" aria-hidden="true" tabindex="-1"></a><span class="er">"checksum":</span> <span class="er">true</span></span></code></pre></div><p>If using <code>rclone rc</code> this could be passed as</p><pre class="console"><code>rclone rc sync/sync ... _config='{"CheckSum": true}'</code></pre><p>Or simply flat:</p>@@ -19024,16 +20575,22 @@ which were set with command line flags or environment variables.</p>see <a href="#data-types">data types</a> for more info. Here is anexample setting the equivalent of <code>--buffer-size</code> in stringor integer format.</p>-<div class="sourceCode" id="cb806"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb806-1"><a href="#cb806-1" aria-hidden="true" tabindex="-1"></a><span class="er">"_config":</span><span class="fu">{</span><span class="dt">"BufferSize"</span><span class="fu">:</span> <span class="st">"42M"</span><span class="fu">}</span></span>-<span id="cb806-2"><a href="#cb806-2" aria-hidden="true" tabindex="-1"></a><span class="er">"_config":</span><span class="fu">{</span><span class="dt">"BufferSize"</span><span class="fu">:</span> <span class="dv">44040192</span><span class="fu">}</span></span></code></pre></div>+<div class="sourceCode" id="cb834"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb834-1"><a href="#cb834-1" aria-hidden="true" tabindex="-1"></a><span class="er">"_config":</span><span class="fu">{</span><span class="dt">"BufferSize"</span><span class="fu">:</span> <span class="st">"42M"</span><span class="fu">}</span></span>+<span id="cb834-2"><a href="#cb834-2" aria-hidden="true" tabindex="-1"></a><span class="er">"_config":</span><span class="fu">{</span><span class="dt">"BufferSize"</span><span class="fu">:</span> <span class="dv">44040192</span><span class="fu">}</span></span></code></pre></div><p>Or flat:</p>-<div class="sourceCode" id="cb807"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb807-1"><a href="#cb807-1" aria-hidden="true" tabindex="-1"></a><span class="er">"buffer_size":</span> <span class="er">"42M"</span></span>-<span id="cb807-2"><a href="#cb807-2" aria-hidden="true" tabindex="-1"></a><span class="er">"buffer_size":</span> <span class="er">44040192</span></span></code></pre></div>+<div class="sourceCode" id="cb835"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb835-1"><a href="#cb835-1" aria-hidden="true" tabindex="-1"></a><span class="er">"buffer_size":</span> <span class="er">"42M"</span></span>+<span id="cb835-2"><a href="#cb835-2" aria-hidden="true" tabindex="-1"></a><span class="er">"buffer_size":</span> <span class="er">44040192</span></span></code></pre></div><p>If you wish to check the <code>_config</code> assignment has workedproperly then calling <code>options/local</code> will show what thevalue got set to.</p>+<p>Setting <code>LogLevel</code> sets the log level for the logs made by+the rc call only, for example to debug a single call.</p>+<div class="sourceCode" id="cb836"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb836-1"><a href="#cb836-1" aria-hidden="true" tabindex="-1"></a><span class="er">"_config":</span><span class="fu">{</span><span class="dt">"LogLevel"</span><span class="fu">:</span> <span class="st">"DEBUG"</span><span class="fu">}</span></span></code></pre></div>+<p>Note that not all of rclone's logs support this yet. In particular+the backends log at the global log level.</p><h3 id="setting-filter-flags-with-_filter">Setting filter flags with_filter</h3><p>If you wish to set filters for the duration of an rc call only then@@ -19052,11 +20609,11 @@ href="#option-blocks">the options blocks section</a> for more info).</p><p>For example, if you wished to run a sync with these flags</p><pre class="text"><code>--max-size 1M --max-age 42s --include "a" --include "b"</code></pre><p>you would pass this parameter in your JSON blob.</p>-<div class="sourceCode" id="cb811"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb811-1"><a href="#cb811-1" aria-hidden="true" tabindex="-1"></a><span class="er">"_filter":</span><span class="fu">{</span><span class="dt">"MaxSize"</span><span class="fu">:</span><span class="st">"1M"</span><span class="fu">,</span> <span class="dt">"IncludeRule"</span><span class="fu">:</span><span class="ot">[</span><span class="st">"a"</span><span class="ot">,</span><span class="st">"b"</span><span class="ot">]</span><span class="fu">,</span> <span class="dt">"MaxAge"</span><span class="fu">:</span><span class="st">"42s"</span><span class="fu">}</span></span></code></pre></div>+<div class="sourceCode" id="cb840"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb840-1"><a href="#cb840-1" aria-hidden="true" tabindex="-1"></a><span class="er">"_filter":</span><span class="fu">{</span><span class="dt">"MaxSize"</span><span class="fu">:</span><span class="st">"1M"</span><span class="fu">,</span> <span class="dt">"IncludeRule"</span><span class="fu">:</span><span class="ot">[</span><span class="st">"a"</span><span class="ot">,</span><span class="st">"b"</span><span class="ot">]</span><span class="fu">,</span> <span class="dt">"MaxAge"</span><span class="fu">:</span><span class="st">"42s"</span><span class="fu">}</span></span></code></pre></div><p>Or pass them flat at the top level:</p>-<div class="sourceCode" id="cb812"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb812-1"><a href="#cb812-1" aria-hidden="true" tabindex="-1"></a><span class="er">"max_size":"1M",</span> <span class="er">"include":</span><span class="ot">[</span><span class="st">"a"</span><span class="ot">,</span><span class="st">"b"</span><span class="ot">]</span><span class="er">,</span> <span class="er">"max_age":"42s"</span></span></code></pre></div>+<div class="sourceCode" id="cb841"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb841-1"><a href="#cb841-1" aria-hidden="true" tabindex="-1"></a><span class="er">"max_size":"1M",</span> <span class="er">"include":</span><span class="ot">[</span><span class="st">"a"</span><span class="ot">,</span><span class="st">"b"</span><span class="ot">]</span><span class="er">,</span> <span class="er">"max_age":"42s"</span></span></code></pre></div><p>If using <code>rclone rc</code> this could be passed as</p><pre class="console"><code>rclone rc ... _filter='{"MaxSize":"1M", "IncludeRule":["a","b"], "MaxAge":"42s"}'</code></pre><p>Or simply flat:</p>@@ -19070,12 +20627,12 @@ which were set with command line flags or environment variables.</p>see <a href="#data-types">data types</a> for more info. Here is anexample setting the equivalent of <code>--buffer-size</code> in stringor integer format.</p>-<div class="sourceCode" id="cb815"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb815-1"><a href="#cb815-1" aria-hidden="true" tabindex="-1"></a><span class="er">"_filter":</span><span class="fu">{</span><span class="dt">"MinSize"</span><span class="fu">:</span> <span class="st">"42M"</span><span class="fu">}</span></span>-<span id="cb815-2"><a href="#cb815-2" aria-hidden="true" tabindex="-1"></a><span class="er">"_filter":</span><span class="fu">{</span><span class="dt">"MinSize"</span><span class="fu">:</span> <span class="dv">44040192</span><span class="fu">}</span></span></code></pre></div>+<div class="sourceCode" id="cb844"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb844-1"><a href="#cb844-1" aria-hidden="true" tabindex="-1"></a><span class="er">"_filter":</span><span class="fu">{</span><span class="dt">"MinSize"</span><span class="fu">:</span> <span class="st">"42M"</span><span class="fu">}</span></span>+<span id="cb844-2"><a href="#cb844-2" aria-hidden="true" tabindex="-1"></a><span class="er">"_filter":</span><span class="fu">{</span><span class="dt">"MinSize"</span><span class="fu">:</span> <span class="dv">44040192</span><span class="fu">}</span></span></code></pre></div><p>Or flat:</p>-<div class="sourceCode" id="cb816"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb816-1"><a href="#cb816-1" aria-hidden="true" tabindex="-1"></a><span class="er">"min_size":</span> <span class="er">"42M"</span></span></code></pre></div>+<div class="sourceCode" id="cb845"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb845-1"><a href="#cb845-1" aria-hidden="true" tabindex="-1"></a><span class="er">"min_size":</span> <span class="er">"42M"</span></span></code></pre></div><p>If you wish to check the <code>_filter</code> assignment has workedproperly then calling <code>options/local</code> will show what thevalue got set to.</p>@@ -19255,36 +20812,36 @@ allowed unless Required or Default is set)</td><p>An example of this might be the <code>--log-level</code> flag. Notethat the <code>Name</code> of the option becomes the command line flagwith <code>_</code> replaced with <code>-</code>.</p>-<div class="sourceCode" id="cb818"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb818-1"><a href="#cb818-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>-<span id="cb818-2"><a href="#cb818-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Advanced"</span><span class="fu">:</span> <span class="kw">false</span><span class="fu">,</span></span>-<span id="cb818-3"><a href="#cb818-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Default"</span><span class="fu">:</span> <span class="dv">5</span><span class="fu">,</span></span>-<span id="cb818-4"><a href="#cb818-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"DefaultStr"</span><span class="fu">:</span> <span class="st">"NOTICE"</span><span class="fu">,</span></span>-<span id="cb818-5"><a href="#cb818-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Examples"</span><span class="fu">:</span> <span class="ot">[</span></span>-<span id="cb818-6"><a href="#cb818-6" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>-<span id="cb818-7"><a href="#cb818-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Help"</span><span class="fu">:</span> <span class="st">""</span><span class="fu">,</span></span>-<span id="cb818-8"><a href="#cb818-8" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Value"</span><span class="fu">:</span> <span class="st">"EMERGENCY"</span></span>-<span id="cb818-9"><a href="#cb818-9" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span><span class="ot">,</span></span>-<span id="cb818-10"><a href="#cb818-10" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>-<span id="cb818-11"><a href="#cb818-11" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Help"</span><span class="fu">:</span> <span class="st">""</span><span class="fu">,</span></span>-<span id="cb818-12"><a href="#cb818-12" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Value"</span><span class="fu">:</span> <span class="st">"ALERT"</span></span>-<span id="cb818-13"><a href="#cb818-13" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span><span class="ot">,</span></span>-<span id="cb818-14"><a href="#cb818-14" aria-hidden="true" tabindex="-1"></a> <span class="er">...</span></span>-<span id="cb818-15"><a href="#cb818-15" aria-hidden="true" tabindex="-1"></a> <span class="ot">]</span><span class="fu">,</span></span>-<span id="cb818-16"><a href="#cb818-16" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Exclusive"</span><span class="fu">:</span> <span class="kw">true</span><span class="fu">,</span></span>-<span id="cb818-17"><a href="#cb818-17" aria-hidden="true" tabindex="-1"></a> <span class="dt">"FieldName"</span><span class="fu">:</span> <span class="st">"LogLevel"</span><span class="fu">,</span></span>-<span id="cb818-18"><a href="#cb818-18" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Groups"</span><span class="fu">:</span> <span class="st">"Logging"</span><span class="fu">,</span></span>-<span id="cb818-19"><a href="#cb818-19" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Help"</span><span class="fu">:</span> <span class="st">"Log level DEBUG|INFO|NOTICE|ERROR"</span><span class="fu">,</span></span>-<span id="cb818-20"><a href="#cb818-20" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Hide"</span><span class="fu">:</span> <span class="dv">0</span><span class="fu">,</span></span>-<span id="cb818-21"><a href="#cb818-21" aria-hidden="true" tabindex="-1"></a> <span class="dt">"IsPassword"</span><span class="fu">:</span> <span class="kw">false</span><span class="fu">,</span></span>-<span id="cb818-22"><a href="#cb818-22" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Name"</span><span class="fu">:</span> <span class="st">"log_level"</span><span class="fu">,</span></span>-<span id="cb818-23"><a href="#cb818-23" aria-hidden="true" tabindex="-1"></a> <span class="dt">"NoPrefix"</span><span class="fu">:</span> <span class="kw">true</span><span class="fu">,</span></span>-<span id="cb818-24"><a href="#cb818-24" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Required"</span><span class="fu">:</span> <span class="kw">true</span><span class="fu">,</span></span>-<span id="cb818-25"><a href="#cb818-25" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Sensitive"</span><span class="fu">:</span> <span class="kw">false</span><span class="fu">,</span></span>-<span id="cb818-26"><a href="#cb818-26" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Type"</span><span class="fu">:</span> <span class="st">"LogLevel"</span><span class="fu">,</span></span>-<span id="cb818-27"><a href="#cb818-27" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Value"</span><span class="fu">:</span> <span class="kw">null</span><span class="fu">,</span></span>-<span id="cb818-28"><a href="#cb818-28" aria-hidden="true" tabindex="-1"></a> <span class="dt">"ValueStr"</span><span class="fu">:</span> <span class="st">"NOTICE"</span></span>-<span id="cb818-29"><a href="#cb818-29" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span><span class="er">,</span></span></code></pre></div>+<div class="sourceCode" id="cb847"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb847-1"><a href="#cb847-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb847-2"><a href="#cb847-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Advanced"</span><span class="fu">:</span> <span class="kw">false</span><span class="fu">,</span></span>+<span id="cb847-3"><a href="#cb847-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Default"</span><span class="fu">:</span> <span class="dv">5</span><span class="fu">,</span></span>+<span id="cb847-4"><a href="#cb847-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"DefaultStr"</span><span class="fu">:</span> <span class="st">"NOTICE"</span><span class="fu">,</span></span>+<span id="cb847-5"><a href="#cb847-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Examples"</span><span class="fu">:</span> <span class="ot">[</span></span>+<span id="cb847-6"><a href="#cb847-6" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>+<span id="cb847-7"><a href="#cb847-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Help"</span><span class="fu">:</span> <span class="st">""</span><span class="fu">,</span></span>+<span id="cb847-8"><a href="#cb847-8" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Value"</span><span class="fu">:</span> <span class="st">"EMERGENCY"</span></span>+<span id="cb847-9"><a href="#cb847-9" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span><span class="ot">,</span></span>+<span id="cb847-10"><a href="#cb847-10" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>+<span id="cb847-11"><a href="#cb847-11" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Help"</span><span class="fu">:</span> <span class="st">""</span><span class="fu">,</span></span>+<span id="cb847-12"><a href="#cb847-12" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Value"</span><span class="fu">:</span> <span class="st">"ALERT"</span></span>+<span id="cb847-13"><a href="#cb847-13" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span><span class="ot">,</span></span>+<span id="cb847-14"><a href="#cb847-14" aria-hidden="true" tabindex="-1"></a> <span class="er">...</span></span>+<span id="cb847-15"><a href="#cb847-15" aria-hidden="true" tabindex="-1"></a> <span class="ot">]</span><span class="fu">,</span></span>+<span id="cb847-16"><a href="#cb847-16" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Exclusive"</span><span class="fu">:</span> <span class="kw">true</span><span class="fu">,</span></span>+<span id="cb847-17"><a href="#cb847-17" aria-hidden="true" tabindex="-1"></a> <span class="dt">"FieldName"</span><span class="fu">:</span> <span class="st">"LogLevel"</span><span class="fu">,</span></span>+<span id="cb847-18"><a href="#cb847-18" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Groups"</span><span class="fu">:</span> <span class="st">"Logging"</span><span class="fu">,</span></span>+<span id="cb847-19"><a href="#cb847-19" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Help"</span><span class="fu">:</span> <span class="st">"Log level DEBUG|INFO|NOTICE|ERROR"</span><span class="fu">,</span></span>+<span id="cb847-20"><a href="#cb847-20" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Hide"</span><span class="fu">:</span> <span class="dv">0</span><span class="fu">,</span></span>+<span id="cb847-21"><a href="#cb847-21" aria-hidden="true" tabindex="-1"></a> <span class="dt">"IsPassword"</span><span class="fu">:</span> <span class="kw">false</span><span class="fu">,</span></span>+<span id="cb847-22"><a href="#cb847-22" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Name"</span><span class="fu">:</span> <span class="st">"log_level"</span><span class="fu">,</span></span>+<span id="cb847-23"><a href="#cb847-23" aria-hidden="true" tabindex="-1"></a> <span class="dt">"NoPrefix"</span><span class="fu">:</span> <span class="kw">true</span><span class="fu">,</span></span>+<span id="cb847-24"><a href="#cb847-24" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Required"</span><span class="fu">:</span> <span class="kw">true</span><span class="fu">,</span></span>+<span id="cb847-25"><a href="#cb847-25" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Sensitive"</span><span class="fu">:</span> <span class="kw">false</span><span class="fu">,</span></span>+<span id="cb847-26"><a href="#cb847-26" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Type"</span><span class="fu">:</span> <span class="st">"LogLevel"</span><span class="fu">,</span></span>+<span id="cb847-27"><a href="#cb847-27" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Value"</span><span class="fu">:</span> <span class="kw">null</span><span class="fu">,</span></span>+<span id="cb847-28"><a href="#cb847-28" aria-hidden="true" tabindex="-1"></a> <span class="dt">"ValueStr"</span><span class="fu">:</span> <span class="st">"NOTICE"</span></span>+<span id="cb847-29"><a href="#cb847-29" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span><span class="er">,</span></span></code></pre></div><p>Note that the <code>Help</code> may be multiple lines separated by<code>\n</code>. The first line will always be a short sentence and thisis the sentence shown when running <code>rclone help flags</code>.</p>@@ -19310,25 +20867,25 @@ set. If the <code>local</code> backend is desired then <code>type</code>should be set to <code>local</code>. If <code>_root</code> isn'tspecified then it defaults to the root of the remote.</p><p>For example this JSON is equivalent to <code>remote:/tmp</code></p>-<div class="sourceCode" id="cb819"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb819-1"><a href="#cb819-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>-<span id="cb819-2"><a href="#cb819-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"_name"</span><span class="fu">:</span> <span class="st">"remote"</span><span class="fu">,</span></span>-<span id="cb819-3"><a href="#cb819-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"_root"</span><span class="fu">:</span> <span class="st">"/tmp"</span></span>-<span id="cb819-4"><a href="#cb819-4" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<div class="sourceCode" id="cb848"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb848-1"><a href="#cb848-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb848-2"><a href="#cb848-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"_name"</span><span class="fu">:</span> <span class="st">"remote"</span><span class="fu">,</span></span>+<span id="cb848-3"><a href="#cb848-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"_root"</span><span class="fu">:</span> <span class="st">"/tmp"</span></span>+<span id="cb848-4"><a href="#cb848-4" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div><p>And this is equivalent to<code>:sftp,host='example.com':/tmp</code></p>-<div class="sourceCode" id="cb820"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb820-1"><a href="#cb820-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>-<span id="cb820-2"><a href="#cb820-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"type"</span><span class="fu">:</span> <span class="st">"sftp"</span><span class="fu">,</span></span>-<span id="cb820-3"><a href="#cb820-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"host"</span><span class="fu">:</span> <span class="st">"example.com"</span><span class="fu">,</span></span>-<span id="cb820-4"><a href="#cb820-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"_root"</span><span class="fu">:</span> <span class="st">"/tmp"</span></span>-<span id="cb820-5"><a href="#cb820-5" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<div class="sourceCode" id="cb849"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb849-1"><a href="#cb849-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb849-2"><a href="#cb849-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"type"</span><span class="fu">:</span> <span class="st">"sftp"</span><span class="fu">,</span></span>+<span id="cb849-3"><a href="#cb849-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"host"</span><span class="fu">:</span> <span class="st">"example.com"</span><span class="fu">,</span></span>+<span id="cb849-4"><a href="#cb849-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"_root"</span><span class="fu">:</span> <span class="st">"/tmp"</span></span>+<span id="cb849-5"><a href="#cb849-5" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div><p>And this is equivalent to <code>/tmp/dir</code></p>-<div class="sourceCode" id="cb821"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb821-1"><a href="#cb821-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>-<span id="cb821-2"><a href="#cb821-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"type"</span><span class="fu">:</span> <span class="st">"local"</span><span class="fu">,</span></span>-<span id="cb821-3"><a href="#cb821-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"_root"</span><span class="fu">:</span> <span class="st">"/tmp/dir"</span></span>-<span id="cb821-4"><a href="#cb821-4" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<div class="sourceCode" id="cb850"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb850-1"><a href="#cb850-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb850-2"><a href="#cb850-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"type"</span><span class="fu">:</span> <span class="st">"local"</span><span class="fu">,</span></span>+<span id="cb850-3"><a href="#cb850-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"_root"</span><span class="fu">:</span> <span class="st">"/tmp/dir"</span></span>+<span id="cb850-4"><a href="#cb850-4" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div><h2 id="supported-commands">Supported commands</h2><!-- autogenerated start "- run make rcdocs - don't edit here" --><h3 id="backend-command">backend/command: Runs a backend command.</h3>@@ -19740,6 +21297,7 @@ returned.</p>"totalTransfers": total number of transfers in the group,"transferTime" : total time spent on running jobs,"transfers": number of transferred files,+ "updatedDirs": number of directories updated (modtime or metadata set),"transferring": an array of currently active file transfers:[{@@ -19928,12 +21486,12 @@ concurrently.</h3><li>inputs - an list of inputs to the commands with an extra<code>_path</code> parameter</li></ul>-<div class="sourceCode" id="cb841"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb841-1"><a href="#cb841-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>-<span id="cb841-2"><a href="#cb841-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"_path"</span><span class="fu">:</span> <span class="st">"rc/path"</span><span class="fu">,</span></span>-<span id="cb841-3"><a href="#cb841-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"param1"</span><span class="fu">:</span> <span class="st">"parameter for the path as documented"</span><span class="fu">,</span></span>-<span id="cb841-4"><a href="#cb841-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"param2"</span><span class="fu">:</span> <span class="st">"parameter for the path as documented, etc"</span><span class="fu">,</span></span>-<span id="cb841-5"><a href="#cb841-5" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<div class="sourceCode" id="cb870"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb870-1"><a href="#cb870-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb870-2"><a href="#cb870-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"_path"</span><span class="fu">:</span> <span class="st">"rc/path"</span><span class="fu">,</span></span>+<span id="cb870-3"><a href="#cb870-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"param1"</span><span class="fu">:</span> <span class="st">"parameter for the path as documented"</span><span class="fu">,</span></span>+<span id="cb870-4"><a href="#cb870-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"param2"</span><span class="fu">:</span> <span class="st">"parameter for the path as documented, etc"</span><span class="fu">,</span></span>+<span id="cb870-5"><a href="#cb870-5" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div><p>The inputs may use <code>_async</code>, <code>_group</code>,<code>_config</code> and <code>_filter</code> as normal when using therc.</p>@@ -19943,37 +21501,37 @@ rc.</p>each in inputs.</li></ul><p>For example:</p>-<div class="sourceCode" id="cb842"><pre-class="sourceCode sh"><code class="sourceCode bash"><span id="cb842-1"><a href="#cb842-1" aria-hidden="true" tabindex="-1"></a><span class="ex">rclone</span> rc job/batch <span class="at">--json</span> <span class="st">'{</span></span>-<span id="cb842-2"><a href="#cb842-2" aria-hidden="true" tabindex="-1"></a><span class="st"> "inputs": [</span></span>-<span id="cb842-3"><a href="#cb842-3" aria-hidden="true" tabindex="-1"></a><span class="st"> {</span></span>-<span id="cb842-4"><a href="#cb842-4" aria-hidden="true" tabindex="-1"></a><span class="st"> "_path": "rc/noop",</span></span>-<span id="cb842-5"><a href="#cb842-5" aria-hidden="true" tabindex="-1"></a><span class="st"> "parameter": "OK"</span></span>-<span id="cb842-6"><a href="#cb842-6" aria-hidden="true" tabindex="-1"></a><span class="st"> },</span></span>-<span id="cb842-7"><a href="#cb842-7" aria-hidden="true" tabindex="-1"></a><span class="st"> {</span></span>-<span id="cb842-8"><a href="#cb842-8" aria-hidden="true" tabindex="-1"></a><span class="st"> "_path": "rc/error",</span></span>-<span id="cb842-9"><a href="#cb842-9" aria-hidden="true" tabindex="-1"></a><span class="st"> "parameter": "BAD"</span></span>-<span id="cb842-10"><a href="#cb842-10" aria-hidden="true" tabindex="-1"></a><span class="st"> }</span></span>-<span id="cb842-11"><a href="#cb842-11" aria-hidden="true" tabindex="-1"></a><span class="st"> ]</span></span>-<span id="cb842-12"><a href="#cb842-12" aria-hidden="true" tabindex="-1"></a><span class="st">}</span></span>-<span id="cb842-13"><a href="#cb842-13" aria-hidden="true" tabindex="-1"></a><span class="st">'</span></span></code></pre></div>+<div class="sourceCode" id="cb871"><pre+class="sourceCode sh"><code class="sourceCode bash"><span id="cb871-1"><a href="#cb871-1" aria-hidden="true" tabindex="-1"></a><span class="ex">rclone</span> rc job/batch <span class="at">--json</span> <span class="st">'{</span></span>+<span id="cb871-2"><a href="#cb871-2" aria-hidden="true" tabindex="-1"></a><span class="st"> "inputs": [</span></span>+<span id="cb871-3"><a href="#cb871-3" aria-hidden="true" tabindex="-1"></a><span class="st"> {</span></span>+<span id="cb871-4"><a href="#cb871-4" aria-hidden="true" tabindex="-1"></a><span class="st"> "_path": "rc/noop",</span></span>+<span id="cb871-5"><a href="#cb871-5" aria-hidden="true" tabindex="-1"></a><span class="st"> "parameter": "OK"</span></span>+<span id="cb871-6"><a href="#cb871-6" aria-hidden="true" tabindex="-1"></a><span class="st"> },</span></span>+<span id="cb871-7"><a href="#cb871-7" aria-hidden="true" tabindex="-1"></a><span class="st"> {</span></span>+<span id="cb871-8"><a href="#cb871-8" aria-hidden="true" tabindex="-1"></a><span class="st"> "_path": "rc/error",</span></span>+<span id="cb871-9"><a href="#cb871-9" aria-hidden="true" tabindex="-1"></a><span class="st"> "parameter": "BAD"</span></span>+<span id="cb871-10"><a href="#cb871-10" aria-hidden="true" tabindex="-1"></a><span class="st"> }</span></span>+<span id="cb871-11"><a href="#cb871-11" aria-hidden="true" tabindex="-1"></a><span class="st"> ]</span></span>+<span id="cb871-12"><a href="#cb871-12" aria-hidden="true" tabindex="-1"></a><span class="st">}</span></span>+<span id="cb871-13"><a href="#cb871-13" aria-hidden="true" tabindex="-1"></a><span class="st">'</span></span></code></pre></div><p>Gives the result:</p>-<div class="sourceCode" id="cb843"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb843-1"><a href="#cb843-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>-<span id="cb843-2"><a href="#cb843-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"results"</span><span class="fu">:</span> <span class="ot">[</span></span>-<span id="cb843-3"><a href="#cb843-3" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>-<span id="cb843-4"><a href="#cb843-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"parameter"</span><span class="fu">:</span> <span class="st">"OK"</span></span>-<span id="cb843-5"><a href="#cb843-5" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span><span class="ot">,</span></span>-<span id="cb843-6"><a href="#cb843-6" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>-<span id="cb843-7"><a href="#cb843-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"error"</span><span class="fu">:</span> <span class="st">"arbitrary error on input map[parameter:BAD]"</span><span class="fu">,</span></span>-<span id="cb843-8"><a href="#cb843-8" aria-hidden="true" tabindex="-1"></a> <span class="dt">"input"</span><span class="fu">:</span> <span class="fu">{</span></span>-<span id="cb843-9"><a href="#cb843-9" aria-hidden="true" tabindex="-1"></a> <span class="dt">"parameter"</span><span class="fu">:</span> <span class="st">"BAD"</span></span>-<span id="cb843-10"><a href="#cb843-10" aria-hidden="true" tabindex="-1"></a> <span class="fu">},</span></span>-<span id="cb843-11"><a href="#cb843-11" aria-hidden="true" tabindex="-1"></a> <span class="dt">"path"</span><span class="fu">:</span> <span class="st">"rc/error"</span><span class="fu">,</span></span>-<span id="cb843-12"><a href="#cb843-12" aria-hidden="true" tabindex="-1"></a> <span class="dt">"status"</span><span class="fu">:</span> <span class="dv">500</span></span>-<span id="cb843-13"><a href="#cb843-13" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>-<span id="cb843-14"><a href="#cb843-14" aria-hidden="true" tabindex="-1"></a> <span class="ot">]</span></span>-<span id="cb843-15"><a href="#cb843-15" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<div class="sourceCode" id="cb872"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb872-1"><a href="#cb872-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb872-2"><a href="#cb872-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"results"</span><span class="fu">:</span> <span class="ot">[</span></span>+<span id="cb872-3"><a href="#cb872-3" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>+<span id="cb872-4"><a href="#cb872-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"parameter"</span><span class="fu">:</span> <span class="st">"OK"</span></span>+<span id="cb872-5"><a href="#cb872-5" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span><span class="ot">,</span></span>+<span id="cb872-6"><a href="#cb872-6" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>+<span id="cb872-7"><a href="#cb872-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"error"</span><span class="fu">:</span> <span class="st">"arbitrary error on input map[parameter:BAD]"</span><span class="fu">,</span></span>+<span id="cb872-8"><a href="#cb872-8" aria-hidden="true" tabindex="-1"></a> <span class="dt">"input"</span><span class="fu">:</span> <span class="fu">{</span></span>+<span id="cb872-9"><a href="#cb872-9" aria-hidden="true" tabindex="-1"></a> <span class="dt">"parameter"</span><span class="fu">:</span> <span class="st">"BAD"</span></span>+<span id="cb872-10"><a href="#cb872-10" aria-hidden="true" tabindex="-1"></a> <span class="fu">},</span></span>+<span id="cb872-11"><a href="#cb872-11" aria-hidden="true" tabindex="-1"></a> <span class="dt">"path"</span><span class="fu">:</span> <span class="st">"rc/error"</span><span class="fu">,</span></span>+<span id="cb872-12"><a href="#cb872-12" aria-hidden="true" tabindex="-1"></a> <span class="dt">"status"</span><span class="fu">:</span> <span class="dv">500</span></span>+<span id="cb872-13"><a href="#cb872-13" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>+<span id="cb872-14"><a href="#cb872-14" aria-hidden="true" tabindex="-1"></a> <span class="ot">]</span></span>+<span id="cb872-15"><a href="#cb872-15" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div><h3 id="job-list">job/list: Lists the IDs of the running jobs</h3><p>Parameters: None.</p><p>Results:</p>@@ -19985,7 +21543,6 @@ restart)</li><li>runningIds - array of integer job ids that are running</li><li>finishedIds - array of integer job ids that are finished</li></ul>-<p><strong>Authentication is not required for this call.</strong></p><h3 id="job-status">job/status: Reads the status of the job ID</h3><p>Parameters:</p><ul>@@ -20009,7 +21566,6 @@ with id uniquely identifies a job</li>synchronously</li><li>progress - output of the progress related to the underlying job</li></ul>-<p><strong>Authentication is not required for this call.</strong></p><h3 id="job-stop">job/stop: Stop the running job</h3><p>Parameters:</p><ul>@@ -20327,6 +21883,30 @@ the remote</h3><p>This command does not have a command line equivalent so use thisinstead:</p><pre><code>rclone rc --loopback operations/fsinfo fs=remote:</code></pre>+<h3 id="operations-getfile">operations/getfile: Get a single file into a+JSON response.</h3>+<p>This takes the following parameters:</p>+<ul>+<li>fs - a remote name string e.g. "drive:path/to/file" or+"drive:path/to/dir"</li>+<li>remote - (optional) a path within that remote e.g. "file.txt"</li>+<li>offset - (optional) start reading at offset N (or from end if+negative) (default 0)</li>+<li>count - (optional) only read N bytes (default -1)</li>+<li>head - (optional) only read the first N bytes (default 0)</li>+<li>tail - (optional) only read the last N bytes (default 0)</li>+<li>maxSize - (optional) maximum size limit in bytes (can only lower the+built-in maximum) (default 2097152 / 2MiB)</li>+<li>base64 - (optional) if true, returns the file contents+base64-encoded (default false)</li>+</ul>+<p>Returns:</p>+<ul>+<li>result - contents of the file as string or base64-encoded+string</li>+</ul>+<p>For streaming file content over HTTP with Range support, use the+--rc-serve flag instead.</p><h3 id="operations-hashsum">operations/hashsum: Produces a hashsum filefor all the objects in the path.</h3><p>Produces a hash file for all the objects in the path using the hash@@ -20398,6 +21978,35 @@ data.</p><p>See the <ahref="https://rclone.org/commands/rclone_hashsum/">hashsum</a> commandfor more information on the above.</p>+<h3 id="operations-index">operations/index: Write static directory+listings into a remote</h3>+<p>This takes the following parameters:</p>+<ul>+<li>fs - a remote name string e.g. "drive:"</li>+<li>opt - a dictionary of options to control the indexing (optional)+<ul>+<li>output - list of listings to write in each directory as+"NAME=FORMAT" (default ["index.html=html"])</li>+<li>rules - which directories get listings, e.g. {"ExcludeRule":+["/private/**"]}</li>+<li>maxDepth - only make listings this many directories deep (default -1+for no limit)</li>+<li>linkIndex - link to "dir/NAME" rather than "dir/"</li>+<li>dirTime - how to work out directory times: "newest", "dir" or+"none"</li>+<li>noModTime - don't show modification times</li>+<li>rewrite - write every listing even if it is unchanged</li>+<li>changed - list of changed paths for a partial run, directories+ending in "/"</li>+<li>changedFrom - list of files of changed paths, one per line</li>+<li>changedCombined - list of files in the sync --combined report+format</li>+<li>changedMaxDirs - do a full run if a partial one would list more+directories than this</li>+</ul></li>+</ul>+<p>See the <a href="https://rclone.org/commands/rclone_index/">index</a>+command for more information on the above.</p><h3 id="operations-list">operations/list: List the given remote and pathin JSON format</h3><p>This takes the following parameters:</p>@@ -20621,76 +22230,6 @@ rclone rc options/set --json '{"main": {"LogLevel": 8}}&<pre><code>rclone rc options/set --json '{"main": {"LogLevel": "INFO"}}'</code></pre><p>And this sets NOTICE level logs (normal without -v)</p><pre><code>rclone rc options/set --json '{"main": {"LogLevel": "NOTICE"}}'</code></pre>-<h3 id="pluginsctl-addPlugin">pluginsctl/addPlugin: Add a plugin using-url</h3>-<p>Used for adding a plugin to the webgui.</p>-<p>This takes the following parameters:</p>-<ul>-<li>url - http url of the github repo where the plugin is hosted-(http://github.com/rclone/rclone-webui-react).</li>-</ul>-<p>Example:</p>-<p>rclone rc pluginsctl/addPlugin</p>-<h3 id="pluginsctl-getPluginsForType">pluginsctl/getPluginsForType: Get-plugins with type criteria</h3>-<p>This shows all possible plugins by a mime type.</p>-<p>This takes the following parameters:</p>-<ul>-<li>type - supported mime type by a loaded plugin e.g. (video/mp4,-audio/mp3).</li>-<li>pluginType - filter plugins based on their type e.g. (DASHBOARD,-FILE_HANDLER, TERMINAL).</li>-</ul>-<p>Returns:</p>-<ul>-<li>loadedPlugins - list of current production plugins.</li>-<li>testPlugins - list of temporarily loaded development plugins,-usually running on a different server.</li>-</ul>-<p>Example:</p>-<p>rclone rc pluginsctl/getPluginsForType type=video/mp4</p>-<h3 id="pluginsctl-listPlugins">pluginsctl/listPlugins: Get the list of-currently loaded plugins</h3>-<p>This allows you to get the currently enabled plugins and their-details.</p>-<p>This takes no parameters and returns:</p>-<ul>-<li>loadedPlugins - list of current production plugins.</li>-<li>testPlugins - list of temporarily loaded development plugins,-usually running on a different server.</li>-</ul>-<p>E.g.</p>-<p>rclone rc pluginsctl/listPlugins</p>-<h3 id="pluginsctl-listTestPlugins">pluginsctl/listTestPlugins: Show-currently loaded test plugins</h3>-<p>Allows listing of test plugins with the rclone.test set to true in-package.json of the plugin.</p>-<p>This takes no parameters and returns:</p>-<ul>-<li>loadedTestPlugins - list of currently available test plugins.</li>-</ul>-<p>E.g.</p>-<pre><code>rclone rc pluginsctl/listTestPlugins</code></pre>-<h3 id="pluginsctl-removePlugin">pluginsctl/removePlugin: Remove a-loaded plugin</h3>-<p>This allows you to remove a plugin using it's name.</p>-<p>This takes parameters:</p>-<ul>-<li>name - name of the plugin in the format-<code>author</code>/<code>plugin_name</code>.</li>-</ul>-<p>E.g.</p>-<p>rclone rc pluginsctl/removePlugin name=rclone/video-plugin</p>-<h3 id="pluginsctl-removeTestPlugin">pluginsctl/removeTestPlugin: Remove-a test plugin</h3>-<p>This allows you to remove a plugin using it's name.</p>-<p>This takes the following parameters:</p>-<ul>-<li>name - name of the plugin in the format-<code>author</code>/<code>plugin_name</code>.</li>-</ul>-<p>Example:</p>-<pre><code>rclone rc pluginsctl/removeTestPlugin name=rclone/rclone-webui-react</code></pre><h3 id="rc-error">rc/error: This returns an error</h3><p>This returns an error with the input as part of its error string.Useful for testing error handling.</p>@@ -20731,25 +22270,25 @@ Useful for testing error handling.</p><p>Eg</p><pre><code>rclone rc serve/list</code></pre><p>Returns</p>-<div class="sourceCode" id="cb860"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb860-1"><a href="#cb860-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>-<span id="cb860-2"><a href="#cb860-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"list"</span><span class="fu">:</span> <span class="ot">[</span></span>-<span id="cb860-3"><a href="#cb860-3" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>-<span id="cb860-4"><a href="#cb860-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"addr"</span><span class="fu">:</span> <span class="st">"[::]:4321"</span><span class="fu">,</span></span>-<span id="cb860-5"><a href="#cb860-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"id"</span><span class="fu">:</span> <span class="st">"nfs-ffc2a4e5"</span><span class="fu">,</span></span>-<span id="cb860-6"><a href="#cb860-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">"params"</span><span class="fu">:</span> <span class="fu">{</span></span>-<span id="cb860-7"><a href="#cb860-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"fs"</span><span class="fu">:</span> <span class="st">"remote:"</span><span class="fu">,</span></span>-<span id="cb860-8"><a href="#cb860-8" aria-hidden="true" tabindex="-1"></a> <span class="dt">"opt"</span><span class="fu">:</span> <span class="fu">{</span></span>-<span id="cb860-9"><a href="#cb860-9" aria-hidden="true" tabindex="-1"></a> <span class="dt">"ListenAddr"</span><span class="fu">:</span> <span class="st">":4321"</span></span>-<span id="cb860-10"><a href="#cb860-10" aria-hidden="true" tabindex="-1"></a> <span class="fu">},</span></span>-<span id="cb860-11"><a href="#cb860-11" aria-hidden="true" tabindex="-1"></a> <span class="dt">"type"</span><span class="fu">:</span> <span class="st">"nfs"</span><span class="fu">,</span></span>-<span id="cb860-12"><a href="#cb860-12" aria-hidden="true" tabindex="-1"></a> <span class="dt">"vfsOpt"</span><span class="fu">:</span> <span class="fu">{</span></span>-<span id="cb860-13"><a href="#cb860-13" aria-hidden="true" tabindex="-1"></a> <span class="dt">"CacheMode"</span><span class="fu">:</span> <span class="st">"full"</span></span>-<span id="cb860-14"><a href="#cb860-14" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>-<span id="cb860-15"><a href="#cb860-15" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>-<span id="cb860-16"><a href="#cb860-16" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>-<span id="cb860-17"><a href="#cb860-17" aria-hidden="true" tabindex="-1"></a> <span class="ot">]</span></span>-<span id="cb860-18"><a href="#cb860-18" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<div class="sourceCode" id="cb887"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb887-1"><a href="#cb887-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb887-2"><a href="#cb887-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"list"</span><span class="fu">:</span> <span class="ot">[</span></span>+<span id="cb887-3"><a href="#cb887-3" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>+<span id="cb887-4"><a href="#cb887-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"addr"</span><span class="fu">:</span> <span class="st">"[::]:4321"</span><span class="fu">,</span></span>+<span id="cb887-5"><a href="#cb887-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"id"</span><span class="fu">:</span> <span class="st">"nfs-ffc2a4e5"</span><span class="fu">,</span></span>+<span id="cb887-6"><a href="#cb887-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">"params"</span><span class="fu">:</span> <span class="fu">{</span></span>+<span id="cb887-7"><a href="#cb887-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"fs"</span><span class="fu">:</span> <span class="st">"remote:"</span><span class="fu">,</span></span>+<span id="cb887-8"><a href="#cb887-8" aria-hidden="true" tabindex="-1"></a> <span class="dt">"opt"</span><span class="fu">:</span> <span class="fu">{</span></span>+<span id="cb887-9"><a href="#cb887-9" aria-hidden="true" tabindex="-1"></a> <span class="dt">"ListenAddr"</span><span class="fu">:</span> <span class="st">":4321"</span></span>+<span id="cb887-10"><a href="#cb887-10" aria-hidden="true" tabindex="-1"></a> <span class="fu">},</span></span>+<span id="cb887-11"><a href="#cb887-11" aria-hidden="true" tabindex="-1"></a> <span class="dt">"type"</span><span class="fu">:</span> <span class="st">"nfs"</span><span class="fu">,</span></span>+<span id="cb887-12"><a href="#cb887-12" aria-hidden="true" tabindex="-1"></a> <span class="dt">"vfsOpt"</span><span class="fu">:</span> <span class="fu">{</span></span>+<span id="cb887-13"><a href="#cb887-13" aria-hidden="true" tabindex="-1"></a> <span class="dt">"CacheMode"</span><span class="fu">:</span> <span class="st">"full"</span></span>+<span id="cb887-14"><a href="#cb887-14" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>+<span id="cb887-15"><a href="#cb887-15" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>+<span id="cb887-16"><a href="#cb887-16" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>+<span id="cb887-17"><a href="#cb887-17" aria-hidden="true" tabindex="-1"></a> <span class="ot">]</span></span>+<span id="cb887-18"><a href="#cb887-18" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div><h3 id="serve-start">serve/start: Create a new server</h3><p>Create a new server with the specified parameters.</p><p>This takes the following parameters:</p>@@ -20782,11 +22321,11 @@ above.</p>rclone rc serve/start --json '{"type":"nfs","fs":"remote:","addr":":1234","vfs_cache_mode":"full"}'rclone rc serve/start type=webdav fs=remote: vfsOpt='{"CacheMode": 2}' proxyOpt='{"AuthProxy": "http://127.0.0.1:8080"}'</code></pre><p>This will give the reply</p>-<div class="sourceCode" id="cb862"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb862-1"><a href="#cb862-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>-<span id="cb862-2"><a href="#cb862-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"addr"</span><span class="fu">:</span> <span class="st">"[::]:4321"</span><span class="fu">,</span> <span class="er">//</span> <span class="er">Address</span> <span class="er">the</span> <span class="er">server</span> <span class="er">was</span> <span class="er">started</span> <span class="er">on</span></span>-<span id="cb862-3"><a href="#cb862-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"id"</span><span class="fu">:</span> <span class="st">"nfs-ecfc6852"</span> <span class="er">//</span> <span class="er">Unique</span> <span class="er">identifier</span> <span class="er">for</span> <span class="er">the</span> <span class="er">server</span> <span class="er">instance</span></span>-<span id="cb862-4"><a href="#cb862-4" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<div class="sourceCode" id="cb889"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb889-1"><a href="#cb889-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb889-2"><a href="#cb889-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"addr"</span><span class="fu">:</span> <span class="st">"[::]:4321"</span><span class="fu">,</span> <span class="er">//</span> <span class="er">Address</span> <span class="er">the</span> <span class="er">server</span> <span class="er">was</span> <span class="er">started</span> <span class="er">on</span></span>+<span id="cb889-3"><a href="#cb889-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"id"</span><span class="fu">:</span> <span class="st">"nfs-ecfc6852"</span> <span class="er">//</span> <span class="er">Unique</span> <span class="er">identifier</span> <span class="er">for</span> <span class="er">the</span> <span class="er">server</span> <span class="er">instance</span></span>+<span id="cb889-4"><a href="#cb889-4" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div><p>Or an error if it failed to start.</p><p>Stop the server with <code>serve/stop</code> and list the runningservers with <code>serve/list</code>.</p>@@ -20815,14 +22354,14 @@ be passed to serve/start as the serveType parameter.</p><p>Eg</p><pre><code>rclone rc serve/types</code></pre><p>Returns</p>-<div class="sourceCode" id="cb866"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb866-1"><a href="#cb866-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>-<span id="cb866-2"><a href="#cb866-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"types"</span><span class="fu">:</span> <span class="ot">[</span></span>-<span id="cb866-3"><a href="#cb866-3" aria-hidden="true" tabindex="-1"></a> <span class="st">"http"</span><span class="ot">,</span></span>-<span id="cb866-4"><a href="#cb866-4" aria-hidden="true" tabindex="-1"></a> <span class="st">"sftp"</span><span class="ot">,</span></span>-<span id="cb866-5"><a href="#cb866-5" aria-hidden="true" tabindex="-1"></a> <span class="st">"nfs"</span></span>-<span id="cb866-6"><a href="#cb866-6" aria-hidden="true" tabindex="-1"></a> <span class="ot">]</span></span>-<span id="cb866-7"><a href="#cb866-7" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<div class="sourceCode" id="cb893"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb893-1"><a href="#cb893-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb893-2"><a href="#cb893-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"types"</span><span class="fu">:</span> <span class="ot">[</span></span>+<span id="cb893-3"><a href="#cb893-3" aria-hidden="true" tabindex="-1"></a> <span class="st">"http"</span><span class="ot">,</span></span>+<span id="cb893-4"><a href="#cb893-4" aria-hidden="true" tabindex="-1"></a> <span class="st">"sftp"</span><span class="ot">,</span></span>+<span id="cb893-5"><a href="#cb893-5" aria-hidden="true" tabindex="-1"></a> <span class="st">"nfs"</span></span>+<span id="cb893-6"><a href="#cb893-6" aria-hidden="true" tabindex="-1"></a> <span class="ot">]</span></span>+<span id="cb893-7"><a href="#cb893-7" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div><h3 id="sync-bisync">sync/bisync: Perform bidirectional synchronizationbetween two paths.</h3><!--- Docs generated by help.go - use go generate to rebuild - DO NOT EDIT --->@@ -20833,6 +22372,9 @@ between two paths.</h3><li>path2 (required) - (string) a remote directory string e.g.<code>drive:path2</code></li><li>dryRun - (bool) dry-run mode</li>+<li>maxDelete - (int) Safety check on maximum percentage of deleted+files allowed. If exceeded, the bisync run will abort. (default:+50%)</li><li>backupDir1 - (string) --backup-dir for Path1. Must be anon-overlapping path on the same remote.<br /></li>@@ -20879,6 +22421,9 @@ Consider using with --verbose<br />(add --ignore- checksum to additionally skip post-copy checksumchecks)<br /></li>+<li>maxDeleteRenamesAware - (bool) Exclude tracked renames from the+--max-delete safety check (requires --track-renames)<br />+</li><li>maxLock - (Duration) Consider lock files older than this to beexpired (default: 0 (never expire)) (minimum: 2m)<br /></li>@@ -20922,7 +22467,35 @@ destination remote</h3>destination</li><li>createEmptySrcDirs - create empty src directories on destination ifset</li>+<li>combined - make a combined report of changes (default false)</li>+<li>missingOnSrc - report all files missing from the source (default+false)</li>+<li>missingOnDst - report all files missing from the destination+(default false)</li>+<li>match - report all matching files (default false)</li>+<li>differ - report all non-matching files (default false)</li>+<li>error - report all files with errors (hashing or reading) (default+false)</li>+<li>destAfter - report all files that exist on the destination post-copy+(default false)</li></ul>+<p>Returns:</p>+<ul>+<li>combined - array of strings of combined report of changes</li>+<li>missingOnSrc - array of strings of all files missing from the+source</li>+<li>missingOnDst - array of strings of all files missing from the+destination</li>+<li>match - array of strings of all matching files</li>+<li>differ - array of strings of all non-matching files</li>+<li>error - array of strings of all files with errors (hashing or+reading)</li>+<li>destAfter - array of strings of all files that exist on the+destination post-copy</li>+</ul>+<p>Each report is only returned if its parameter is set to true. If the+operation fails the reports are only available if it was run with+<code>_async</code>, as part of the job output.</p><p>See the <a href="https://rclone.org/commands/rclone_copy/">copy</a>command for more information on the above.</p><h3 id="sync-move">sync/move: move a directory from source remote to@@ -20935,7 +22508,35 @@ destination</li><li>createEmptySrcDirs - create empty src directories on destination ifset</li><li>deleteEmptySrcDirs - delete empty src directories if set</li>+<li>combined - make a combined report of changes (default false)</li>+<li>missingOnSrc - report all files missing from the source (default+false)</li>+<li>missingOnDst - report all files missing from the destination+(default false)</li>+<li>match - report all matching files (default false)</li>+<li>differ - report all non-matching files (default false)</li>+<li>error - report all files with errors (hashing or reading) (default+false)</li>+<li>destAfter - report all files that exist on the destination post-move+(default false)</li>+</ul>+<p>Returns:</p>+<ul>+<li>combined - array of strings of combined report of changes</li>+<li>missingOnSrc - array of strings of all files missing from the+source</li>+<li>missingOnDst - array of strings of all files missing from the+destination</li>+<li>match - array of strings of all matching files</li>+<li>differ - array of strings of all non-matching files</li>+<li>error - array of strings of all files with errors (hashing or+reading)</li>+<li>destAfter - array of strings of all files that exist on the+destination post-move</li></ul>+<p>Each report is only returned if its parameter is set to true. If the+operation fails the reports are only available if it was run with+<code>_async</code>, as part of the job output.</p><p>See the <a href="https://rclone.org/commands/rclone_move/">move</a>command for more information on the above.</p><h3 id="sync-sync">sync/sync: sync a directory from source remote to@@ -20947,7 +22548,35 @@ destination remote</h3>destination</li><li>createEmptySrcDirs - create empty src directories on destination ifset</li>+<li>combined - make a combined report of changes (default false)</li>+<li>missingOnSrc - report all files missing from the source (default+false)</li>+<li>missingOnDst - report all files missing from the destination+(default false)</li>+<li>match - report all matching files (default false)</li>+<li>differ - report all non-matching files (default false)</li>+<li>error - report all files with errors (hashing or reading) (default+false)</li>+<li>destAfter - report all files that exist on the destination post-sync+(default false)</li>+</ul>+<p>Returns:</p>+<ul>+<li>combined - array of strings of combined report of changes</li>+<li>missingOnSrc - array of strings of all files missing from the+source</li>+<li>missingOnDst - array of strings of all files missing from the+destination</li>+<li>match - array of strings of all matching files</li>+<li>differ - array of strings of all non-matching files</li>+<li>error - array of strings of all files with errors (hashing or+reading)</li>+<li>destAfter - array of strings of all files that exist on the+destination post-sync</li></ul>+<p>Each report is only returned if its parameter is set to true. If the+operation fails the reports are only available if it was run with+<code>_async</code>, as part of the job output.</p><p>See the <a href="https://rclone.org/commands/rclone_sync/">sync</a>command for more information on the above.</p><h3 id="vfs-forget">vfs/forget: Forget files or directories in the@@ -21119,16 +22748,16 @@ formatted to be reasonably human-readable.</p><p>If an error occurs then there will be an HTTP error status (e.g. 500)and the body of the response will contain a JSON encoded error object,e.g.</p>-<div class="sourceCode" id="cb874"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb874-1"><a href="#cb874-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>-<span id="cb874-2"><a href="#cb874-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"error"</span><span class="fu">:</span> <span class="st">"Expecting string value for key </span><span class="ch">\"</span><span class="st">remote</span><span class="ch">\"</span><span class="st"> (was float64)"</span><span class="fu">,</span></span>-<span id="cb874-3"><a href="#cb874-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"input"</span><span class="fu">:</span> <span class="fu">{</span></span>-<span id="cb874-4"><a href="#cb874-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"fs"</span><span class="fu">:</span> <span class="st">"/tmp"</span><span class="fu">,</span></span>-<span id="cb874-5"><a href="#cb874-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"remote"</span><span class="fu">:</span> <span class="dv">3</span></span>-<span id="cb874-6"><a href="#cb874-6" aria-hidden="true" tabindex="-1"></a> <span class="fu">},</span></span>-<span id="cb874-7"><a href="#cb874-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"status"</span><span class="fu">:</span> <span class="dv">400</span><span class="fu">,</span></span>-<span id="cb874-8"><a href="#cb874-8" aria-hidden="true" tabindex="-1"></a> <span class="dt">"path"</span><span class="fu">:</span> <span class="st">"operations/rmdir"</span></span>-<span id="cb874-9"><a href="#cb874-9" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<div class="sourceCode" id="cb901"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb901-1"><a href="#cb901-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb901-2"><a href="#cb901-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"error"</span><span class="fu">:</span> <span class="st">"Expecting string value for key </span><span class="ch">\"</span><span class="st">remote</span><span class="ch">\"</span><span class="st"> (was float64)"</span><span class="fu">,</span></span>+<span id="cb901-3"><a href="#cb901-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"input"</span><span class="fu">:</span> <span class="fu">{</span></span>+<span id="cb901-4"><a href="#cb901-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"fs"</span><span class="fu">:</span> <span class="st">"/tmp"</span><span class="fu">,</span></span>+<span id="cb901-5"><a href="#cb901-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"remote"</span><span class="fu">:</span> <span class="dv">3</span></span>+<span id="cb901-6"><a href="#cb901-6" aria-hidden="true" tabindex="-1"></a> <span class="fu">},</span></span>+<span id="cb901-7"><a href="#cb901-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"status"</span><span class="fu">:</span> <span class="dv">400</span><span class="fu">,</span></span>+<span id="cb901-8"><a href="#cb901-8" aria-hidden="true" tabindex="-1"></a> <span class="dt">"path"</span><span class="fu">:</span> <span class="st">"operations/rmdir"</span></span>+<span id="cb901-9"><a href="#cb901-9" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div><p>The keys in the error response are:</p><ul><li>error - error string</li>@@ -21137,28 +22766,31 @@ class="sourceCode json"><code class="sourceCode json"><span id="cb874-1"><a href<li>path - the path of the call</li></ul><h3 id="cors">CORS</h3>-<p>The sever implements basic CORS support and allows all origins for-that. The response to a preflight OPTIONS request will echo the-requested "Access-Control-Request-Headers" back.</p>+<p>The server implements basic CORS support when+<code>--rc-allow-origin</code> is set. Browser CORS preflight OPTIONS+requests are then answered by rclone without authentication with the+<code>Access-Control-*</code> headers. Without+<code>--rc-allow-origin</code> no CORS headers are sent and a preflight+request is authenticated like any other request.</p><h3 id="using-post-with-url-parameters-only">Using POST with URLparameters only</h3><pre class="console"><code>curl -X POST 'http://localhost:5572/rc/noop?potato=1&sausage=2'</code></pre><p>Response</p>-<div class="sourceCode" id="cb876"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb876-1"><a href="#cb876-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>-<span id="cb876-2"><a href="#cb876-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"potato"</span><span class="fu">:</span> <span class="st">"1"</span><span class="fu">,</span></span>-<span id="cb876-3"><a href="#cb876-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"sausage"</span><span class="fu">:</span> <span class="st">"2"</span></span>-<span id="cb876-4"><a href="#cb876-4" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<div class="sourceCode" id="cb903"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb903-1"><a href="#cb903-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb903-2"><a href="#cb903-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"potato"</span><span class="fu">:</span> <span class="st">"1"</span><span class="fu">,</span></span>+<span id="cb903-3"><a href="#cb903-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"sausage"</span><span class="fu">:</span> <span class="st">"2"</span></span>+<span id="cb903-4"><a href="#cb903-4" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div><p>Here is what an error response looks like:</p><pre class="console"><code>curl -X POST 'http://localhost:5572/rc/error?potato=1&sausage=2'</code></pre>-<div class="sourceCode" id="cb878"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb878-1"><a href="#cb878-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>-<span id="cb878-2"><a href="#cb878-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"error"</span><span class="fu">:</span> <span class="st">"arbitrary error on input map[potato:1 sausage:2]"</span><span class="fu">,</span></span>-<span id="cb878-3"><a href="#cb878-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"input"</span><span class="fu">:</span> <span class="fu">{</span></span>-<span id="cb878-4"><a href="#cb878-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"potato"</span><span class="fu">:</span> <span class="st">"1"</span><span class="fu">,</span></span>-<span id="cb878-5"><a href="#cb878-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"sausage"</span><span class="fu">:</span> <span class="st">"2"</span></span>-<span id="cb878-6"><a href="#cb878-6" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>-<span id="cb878-7"><a href="#cb878-7" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<div class="sourceCode" id="cb905"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb905-1"><a href="#cb905-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb905-2"><a href="#cb905-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"error"</span><span class="fu">:</span> <span class="st">"arbitrary error on input map[potato:1 sausage:2]"</span><span class="fu">,</span></span>+<span id="cb905-3"><a href="#cb905-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"input"</span><span class="fu">:</span> <span class="fu">{</span></span>+<span id="cb905-4"><a href="#cb905-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"potato"</span><span class="fu">:</span> <span class="st">"1"</span><span class="fu">,</span></span>+<span id="cb905-5"><a href="#cb905-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"sausage"</span><span class="fu">:</span> <span class="st">"2"</span></span>+<span id="cb905-6"><a href="#cb905-6" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>+<span id="cb905-7"><a href="#cb905-7" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div><p>Note that curl doesn't return errors to the shell unless you use the<code>-f</code> option</p><pre class="console"><code>$ curl -f -X POST 'http://localhost:5572/rc/error?potato=1&sausage=2'@@ -21168,38 +22800,38 @@ $ echo $?<h3 id="using-post-with-a-form">Using POST with a form</h3><pre class="console"><code>curl --data "potato=1" --data "sausage=2" http://localhost:5572/rc/noop</code></pre><p>Response</p>-<div class="sourceCode" id="cb881"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb881-1"><a href="#cb881-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>-<span id="cb881-2"><a href="#cb881-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"potato"</span><span class="fu">:</span> <span class="st">"1"</span><span class="fu">,</span></span>-<span id="cb881-3"><a href="#cb881-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"sausage"</span><span class="fu">:</span> <span class="st">"2"</span></span>-<span id="cb881-4"><a href="#cb881-4" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<div class="sourceCode" id="cb908"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb908-1"><a href="#cb908-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb908-2"><a href="#cb908-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"potato"</span><span class="fu">:</span> <span class="st">"1"</span><span class="fu">,</span></span>+<span id="cb908-3"><a href="#cb908-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"sausage"</span><span class="fu">:</span> <span class="st">"2"</span></span>+<span id="cb908-4"><a href="#cb908-4" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div><p>Note that you can combine these with URL parameters too with the POSTparameters taking precedence.</p><pre class="console"><code>curl --data "potato=1" --data "sausage=2" "http://localhost:5572/rc/noop?rutabaga=3&sausage=4"</code></pre><p>Response</p>-<div class="sourceCode" id="cb883"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb883-1"><a href="#cb883-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>-<span id="cb883-2"><a href="#cb883-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"potato"</span><span class="fu">:</span> <span class="st">"1"</span><span class="fu">,</span></span>-<span id="cb883-3"><a href="#cb883-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"rutabaga"</span><span class="fu">:</span> <span class="st">"3"</span><span class="fu">,</span></span>-<span id="cb883-4"><a href="#cb883-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"sausage"</span><span class="fu">:</span> <span class="st">"4"</span></span>-<span id="cb883-5"><a href="#cb883-5" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<div class="sourceCode" id="cb910"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb910-1"><a href="#cb910-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb910-2"><a href="#cb910-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"potato"</span><span class="fu">:</span> <span class="st">"1"</span><span class="fu">,</span></span>+<span id="cb910-3"><a href="#cb910-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"rutabaga"</span><span class="fu">:</span> <span class="st">"3"</span><span class="fu">,</span></span>+<span id="cb910-4"><a href="#cb910-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"sausage"</span><span class="fu">:</span> <span class="st">"4"</span></span>+<span id="cb910-5"><a href="#cb910-5" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div><h3 id="using-post-with-a-json-blob">Using POST with a JSON blob</h3><pre class="console"><code>curl -H "Content-Type: application/json" -X POST -d '{"potato":2,"sausage":1}' http://localhost:5572/rc/noop</code></pre><p>response</p>-<div class="sourceCode" id="cb885"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb885-1"><a href="#cb885-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>-<span id="cb885-2"><a href="#cb885-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"password"</span><span class="fu">:</span> <span class="st">"xyz"</span><span class="fu">,</span></span>-<span id="cb885-3"><a href="#cb885-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"username"</span><span class="fu">:</span> <span class="st">"xyz"</span></span>-<span id="cb885-4"><a href="#cb885-4" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<div class="sourceCode" id="cb912"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb912-1"><a href="#cb912-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb912-2"><a href="#cb912-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"password"</span><span class="fu">:</span> <span class="st">"xyz"</span><span class="fu">,</span></span>+<span id="cb912-3"><a href="#cb912-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"username"</span><span class="fu">:</span> <span class="st">"xyz"</span></span>+<span id="cb912-4"><a href="#cb912-4" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div><p>This can be combined with URL parameters too if required. The JSONblob takes precedence.</p><pre class="console"><code>curl -H "Content-Type: application/json" -X POST -d '{"potato":2,"sausage":1}' 'http://localhost:5572/rc/noop?rutabaga=3&potato=4'</code></pre>-<div class="sourceCode" id="cb887"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb887-1"><a href="#cb887-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>-<span id="cb887-2"><a href="#cb887-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"potato"</span><span class="fu">:</span> <span class="dv">2</span><span class="fu">,</span></span>-<span id="cb887-3"><a href="#cb887-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"rutabaga"</span><span class="fu">:</span> <span class="st">"3"</span><span class="fu">,</span></span>-<span id="cb887-4"><a href="#cb887-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"sausage"</span><span class="fu">:</span> <span class="dv">1</span></span>-<span id="cb887-5"><a href="#cb887-5" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<div class="sourceCode" id="cb914"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb914-1"><a href="#cb914-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb914-2"><a href="#cb914-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"potato"</span><span class="fu">:</span> <span class="dv">2</span><span class="fu">,</span></span>+<span id="cb914-3"><a href="#cb914-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"rutabaga"</span><span class="fu">:</span> <span class="st">"3"</span><span class="fu">,</span></span>+<span id="cb914-4"><a href="#cb914-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"sausage"</span><span class="fu">:</span> <span class="dv">1</span></span>+<span id="cb914-5"><a href="#cb914-5" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div><h2 id="debugging-rclone-with-pprof">Debugging rclone with pprof</h2><p>If you use the <code>--rc</code> flag this will also enable the useof the go profiling tools on the same port.</p>@@ -21207,7 +22839,7 @@ of the go profiling tools on the same port.</p>rest of the rc: they are only served if authentication has been set up(<code>--rc-user</code>/<code>--rc-pass</code> or<code>--rc-htpasswd</code>) or the <a-href="#--rc-no-auth"><code>--rc-no-auth</code></a> flag is in use. For+href="#rc-no-auth"><code>--rc-no-auth</code></a> flag is in use. Fordebugging on the default localhost port the easiest thing is to use<code>--rc --rc-no-auth</code> (but see the warning about using<code>--rc-no-auth</code> on a non-loopback bind).</p>@@ -22272,7 +23904,8 @@ split into groups.</p><pre><code> --max-backlog int Maximum number of objects in sync or check backlog (default 10000)</code></pre><h2 id="networking">Networking</h2><p>Flags for general networking and HTTP stuff.</p>-<pre><code> --bind string Local address to bind to for outgoing connections, IPv4, IPv6 or name+<pre><code> --allow-https-downgrade Follow redirects from HTTPS to HTTP (insecure)+ --bind string Local address to bind to for outgoing connections, IPv4, IPv6 or name--bwlimit BwTimetable Bandwidth limit in KiB/s, or use suffix B|K|M|G|T|P or a full timetable--bwlimit-file BwTimetable Bandwidth limit per file in KiB/s, or use suffix B|K|M|G|T|P or a full timetable--ca-cert stringArray CA certificate used to verify servers@@ -22295,8 +23928,8 @@ split into groups.</p>--tpslimit float Limit HTTP transactions per second to this--tpslimit-burst int Max burst of transactions for --tpslimit (default 1)--use-cookies Enable session cookiejar- --user-agent string Set the user-agent to a specified string (default "rclone/v1.75.0")</code></pre>-<h2 id="performance">Performance</h2>+ --user-agent string Set the user-agent to a specified string (default "rclone/v1.76.0")</code></pre>+<h2 id="performance-1">Performance</h2><p>Flags helpful for increasing performance.</p><pre><code> --buffer-size SizeSuffix In memory buffer size when reading files for each --transfer (default 16Mi)--checkers int Number of checkers to run in parallel (default 8)@@ -22427,12 +24060,7 @@ split into groups.</p>--rc-server-write-timeout Duration Timeout for server writing data (default 1h0m0s)--rc-template string User-specified template--rc-user string User name for authentication- --rc-user-from-header string User name from a defined HTTP header- --rc-web-fetch-url string URL to fetch the releases for webgui (default "https://api.github.com/repos/rclone/rclone-webui-react/releases/latest")- --rc-web-gui Launch WebGUI on localhost- --rc-web-gui-force-update Force update to latest version of web gui- --rc-web-gui-no-open-browser Don't open the browser automatically- --rc-web-gui-update Check and update to latest version of web gui</code></pre>+ --rc-user-from-header string User name from a defined HTTP header</code></pre><h2 id="metrics-1">Metrics</h2><p>Flags to control the Metrics HTTP endpoint..</p><pre><code> --metrics-addr stringArray IPaddress:Port or :Port to bind server to@@ -22482,8 +24110,10 @@ split into groups.</p>--azureblob-encoding Encoding The encoding for the backend (default Slash,BackSlash,Del,Ctl,RightPeriod,InvalidUtf8)--azureblob-endpoint string Endpoint for the service--azureblob-env-auth Read credentials from runtime (environment variables, CLI or MSI)+ --azureblob-hns Tristate Set if the storage account has a hierarchical namespace (default unset)--azureblob-key string Storage Account Shared Key--azureblob-list-chunk int Size of blob list (default 5000)+ --azureblob-list-parallelism int Number of parallel shards to list a directory with--azureblob-msi-client-id string Object ID of the user-assigned MSI to use, if any--azureblob-msi-mi-res-id string Azure resource ID of the user-assigned MSI to use, if any--azureblob-msi-object-id string Object ID of the user-assigned MSI to use, if any@@ -22496,6 +24126,7 @@ split into groups.</p>--azureblob-tenant string ID of the service principal's tenant. Also called its directory ID--azureblob-upload-concurrency int Concurrency for multipart uploads (default 16)--azureblob-upload-cutoff string Cutoff for switching to chunked upload (<= 256 MiB) (deprecated)+ --azureblob-use-arrow-list Use the Apache Arrow listing format--azureblob-use-az Use Azure CLI tool az for authentication--azureblob-use-copy-blob Whether to use the Copy Blob API when copying to the same storage account (default true)--azureblob-use-emulator Uses local storage emulator if provided as 'true'@@ -22628,6 +24259,12 @@ split into groups.</p>--doi-doi string The DOI or the doi.org URL--doi-doi-resolver-api-url string The URL of the DOI resolver API to use--doi-provider string DOI provider+ --dosya-api-key string Your API Key, get it from https://dosya.dev/profile under "API keys"+ --dosya-api-url string API base URL (default "https://api.dosya.dev")+ --dosya-description string Description of the remote+ --dosya-encoding Encoding The encoding for the backend (default Slash,LtGt,DoubleQuote,SingleQuote,BackSlash,Del,Ctl,LeftSpace,RightSpace,InvalidUtf8,Dot)+ --dosya-upload-concurrency int Concurrency for multipart uploads (default 4)+ --dosya-workspace-id string Your workspace ID--drime-access-token string API Access token--drime-chunk-size SizeSuffix Chunk size to use for uploading (default 5Mi)--drime-description string Description of the remote@@ -22850,6 +24487,8 @@ split into groups.</p>--http-headers CommaSepList Set HTTP headers for all transactions--http-no-escape Do not escape URL metacharacters in path names--http-no-head Don't use HEAD requests+ --http-no-head-size SizeSuffix Set the reported file size for --http-no-head (default off)+ --http-no-head-time Time Set a default modification time for --http-no-head (default off)--http-no-slash Set this if the site doesn't end directories with /--http-url string URL of HTTP host to connect to--huaweidrive-auth-url string Auth server URL@@ -22909,6 +24548,17 @@ split into groups.</p>--jottacloud-token-url string Token server url--jottacloud-trashed-only Only show files that are in the trash--jottacloud-upload-resume-limit SizeSuffix Files bigger than this can be resumed if the upload fail's (default 10Mi)+ --kdrive-auth-url string Auth server URL+ --kdrive-chunk-size SizeSuffix Files above this size are uploaded in chunks of this size (default 20Mi)+ --kdrive-client-id string OAuth Client Id+ --kdrive-description string Description of the remote+ --kdrive-drive-id string ID of the kDrive to use+ --kdrive-encoding Encoding The encoding for the backend (default Slash,Del,Ctl,LeftSpace,RightSpace,InvalidUtf8,Dot)+ --kdrive-endpoint string The API endpoint to use+ --kdrive-root-folder-id string The folder to use as root+ --kdrive-token string OAuth Access Token as a JSON blob+ --kdrive-token-url string Token server url+ --kdrive-upload-concurrency int Concurrency for chunked uploads (default 4)--koofr-description string Description of the remote--koofr-encoding Encoding The encoding for the backend (default Slash,BackSlash,Del,Ctl,InvalidUtf8,Dot)--koofr-endpoint string The Koofr API endpoint to use@@ -22994,6 +24644,7 @@ split into groups.</p>--onedrive-root-folder-id string ID of the root folder--onedrive-server-side-across-configs Deprecated: use --server-side-across-configs instead--onedrive-tenant string ID of the service principal's tenant. Also called its directory ID+ --onedrive-tenant-api-version string The SharePoint API version to use with tenant_url (default "v2.0")--onedrive-tenant-url string The tenant URL for non-admin OneDrive access--onedrive-token string OAuth Access Token as a JSON blob--onedrive-token-url string Token server url@@ -23008,7 +24659,7 @@ split into groups.</p>--oos-decompress If set this will decompress gzip encoded objects--oos-description string Description of the remote--oos-disable-checksum Don't store MD5 checksum with object metadata- --oos-encoding Encoding The encoding for the backend (default Slash,InvalidUtf8,Dot)+ --oos-encoding Encoding The encoding for the backend (default Slash,CrLf,InvalidUtf8,Dot)--oos-endpoint string Endpoint for Object storage API--oos-leave-parts-on-error If true avoid calling abort upload on a failure, leaving all successfully uploaded parts for manual recovery--oos-max-upload-parts int Maximum number of parts in a multipart upload (default 10000)@@ -23203,7 +24854,7 @@ split into groups.</p>--sftp-disable-concurrent-reads If set don't use concurrent reads--sftp-disable-concurrent-writes If set don't use concurrent writes--sftp-disable-hashcheck Disable the execution of SSH commands to determine if remote file hashing is available- --sftp-encoding Encoding The encoding for the backend (default Slash,Del,Ctl,Dot)+ --sftp-encoding Encoding The encoding for the backend (default Slash,Dot)--sftp-hashes CommaSepList Comma separated list of supported checksum types--sftp-host string SSH host to connect to--sftp-host-key-algorithms SpaceSepList Space separated list of host key algorithms, ordered by preference@@ -23218,6 +24869,7 @@ split into groups.</p>--sftp-known-hosts-file string Optional path to known_hosts file--sftp-macs SpaceSepList Space separated list of MACs (message authentication code) algorithms, ordered by preference--sftp-md5sum-command string The command used to read MD5 hashes+ --sftp-multithread-upload Set this to use multi-thread (multi-connection) uploads--sftp-pass string SSH password, leave blank to use ssh-agent (obscured)--sftp-path-override string Override path used by SSH shell commands--sftp-pin-host-key Pin the server host key on first connection (Trust On First Use)@@ -23280,7 +24932,8 @@ split into groups.</p>--smb-port int SMB port number (default 445)--smb-spn string Service principal name--smb-use-kerberos Use Kerberos authentication- --smb-user string SMB username (default "$USER")+ --smb-user string SMB username+ --smb-workstation string Workstation name sent in NTLM authentication--storj-access-grant string Access grant--storj-api-key string API key--storj-description string Description of the remote@@ -23355,6 +25008,7 @@ split into groups.</p>--webdav-url string URL of http host to connect to--webdav-user string User name--webdav-vendor string Name of the WebDAV site/service/software you are using+ --yandex-app-folder Use the application folder as the root--yandex-auth-url string Auth server URL--yandex-client-credentials Use client credentials OAuth flow--yandex-client-id string OAuth Client Id@@ -23473,30 +25127,30 @@ to the Swarm cluster and save as<em>every</em> node. By default this location is accessible only to theroot user so you will need appropriate privileges. The resulting configwill look like this:</p>-<div class="sourceCode" id="cb917"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb917-1"><a href="#cb917-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[gdrive]</span></span>-<span id="cb917-2"><a href="#cb917-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> drive</span></span>-<span id="cb917-3"><a href="#cb917-3" aria-hidden="true" tabindex="-1"></a><span class="dt">scope </span><span class="ot">=</span><span class="st"> drive</span></span>-<span id="cb917-4"><a href="#cb917-4" aria-hidden="true" tabindex="-1"></a><span class="dt">drive_id </span><span class="ot">=</span><span class="st"> 1234567...</span></span>-<span id="cb917-5"><a href="#cb917-5" aria-hidden="true" tabindex="-1"></a><span class="dt">root_folder_id </span><span class="ot">=</span><span class="st"> 0Abcd...</span></span>-<span id="cb917-6"><a href="#cb917-6" aria-hidden="true" tabindex="-1"></a><span class="dt">token </span><span class="ot">=</span><span class="st"> {"access_token":...}</span></span></code></pre></div>+<div class="sourceCode" id="cb944"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb944-1"><a href="#cb944-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[gdrive]</span></span>+<span id="cb944-2"><a href="#cb944-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> drive</span></span>+<span id="cb944-3"><a href="#cb944-3" aria-hidden="true" tabindex="-1"></a><span class="dt">scope </span><span class="ot">=</span><span class="st"> drive</span></span>+<span id="cb944-4"><a href="#cb944-4" aria-hidden="true" tabindex="-1"></a><span class="dt">drive_id </span><span class="ot">=</span><span class="st"> 1234567...</span></span>+<span id="cb944-5"><a href="#cb944-5" aria-hidden="true" tabindex="-1"></a><span class="dt">root_folder_id </span><span class="ot">=</span><span class="st"> 0Abcd...</span></span>+<span id="cb944-6"><a href="#cb944-6" aria-hidden="true" tabindex="-1"></a><span class="dt">token </span><span class="ot">=</span><span class="st"> {"access_token":...}</span></span></code></pre></div><p>Now create the file named <code>example.yml</code> with a swarm stackdescription like this:</p>-<div class="sourceCode" id="cb918"><pre-class="sourceCode yaml"><code class="sourceCode yaml"><span id="cb918-1"><a href="#cb918-1" aria-hidden="true" tabindex="-1"></a><span class="fu">version</span><span class="kw">:</span><span class="at"> </span><span class="st">'3'</span></span>-<span id="cb918-2"><a href="#cb918-2" aria-hidden="true" tabindex="-1"></a><span class="fu">services</span><span class="kw">:</span></span>-<span id="cb918-3"><a href="#cb918-3" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">heimdall</span><span class="kw">:</span></span>-<span id="cb918-4"><a href="#cb918-4" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">image</span><span class="kw">:</span><span class="at"> linuxserver/heimdall:latest</span></span>-<span id="cb918-5"><a href="#cb918-5" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">ports</span><span class="kw">:</span><span class="at"> </span><span class="kw">[</span><span class="dv">8080</span><span class="at">:</span><span class="dv">80</span><span class="kw">]</span></span>-<span id="cb918-6"><a href="#cb918-6" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">volumes</span><span class="kw">:</span><span class="at"> </span><span class="kw">[</span><span class="at">configdata:/config</span><span class="kw">]</span></span>-<span id="cb918-7"><a href="#cb918-7" aria-hidden="true" tabindex="-1"></a><span class="fu">volumes</span><span class="kw">:</span></span>-<span id="cb918-8"><a href="#cb918-8" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">configdata</span><span class="kw">:</span></span>-<span id="cb918-9"><a href="#cb918-9" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">driver</span><span class="kw">:</span><span class="at"> rclone</span></span>-<span id="cb918-10"><a href="#cb918-10" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">driver_opts</span><span class="kw">:</span></span>-<span id="cb918-11"><a href="#cb918-11" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">remote</span><span class="kw">:</span><span class="at"> </span><span class="st">'gdrive:heimdall'</span></span>-<span id="cb918-12"><a href="#cb918-12" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">allow_other</span><span class="kw">:</span><span class="at"> </span><span class="st">'true'</span></span>-<span id="cb918-13"><a href="#cb918-13" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">vfs_cache_mode</span><span class="kw">:</span><span class="at"> full</span></span>-<span id="cb918-14"><a href="#cb918-14" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">poll_interval</span><span class="kw">:</span><span class="at"> </span><span class="dv">0</span></span></code></pre></div>+<div class="sourceCode" id="cb945"><pre+class="sourceCode yaml"><code class="sourceCode yaml"><span id="cb945-1"><a href="#cb945-1" aria-hidden="true" tabindex="-1"></a><span class="fu">version</span><span class="kw">:</span><span class="at"> </span><span class="st">'3'</span></span>+<span id="cb945-2"><a href="#cb945-2" aria-hidden="true" tabindex="-1"></a><span class="fu">services</span><span class="kw">:</span></span>+<span id="cb945-3"><a href="#cb945-3" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">heimdall</span><span class="kw">:</span></span>+<span id="cb945-4"><a href="#cb945-4" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">image</span><span class="kw">:</span><span class="at"> linuxserver/heimdall:latest</span></span>+<span id="cb945-5"><a href="#cb945-5" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">ports</span><span class="kw">:</span><span class="at"> </span><span class="kw">[</span><span class="dv">8080</span><span class="at">:</span><span class="dv">80</span><span class="kw">]</span></span>+<span id="cb945-6"><a href="#cb945-6" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">volumes</span><span class="kw">:</span><span class="at"> </span><span class="kw">[</span><span class="at">configdata:/config</span><span class="kw">]</span></span>+<span id="cb945-7"><a href="#cb945-7" aria-hidden="true" tabindex="-1"></a><span class="fu">volumes</span><span class="kw">:</span></span>+<span id="cb945-8"><a href="#cb945-8" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">configdata</span><span class="kw">:</span></span>+<span id="cb945-9"><a href="#cb945-9" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">driver</span><span class="kw">:</span><span class="at"> rclone</span></span>+<span id="cb945-10"><a href="#cb945-10" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">driver_opts</span><span class="kw">:</span></span>+<span id="cb945-11"><a href="#cb945-11" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">remote</span><span class="kw">:</span><span class="at"> </span><span class="st">'gdrive:heimdall'</span></span>+<span id="cb945-12"><a href="#cb945-12" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">allow_other</span><span class="kw">:</span><span class="at"> </span><span class="st">'true'</span></span>+<span id="cb945-13"><a href="#cb945-13" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">vfs_cache_mode</span><span class="kw">:</span><span class="at"> full</span></span>+<span id="cb945-14"><a href="#cb945-14" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">poll_interval</span><span class="kw">:</span><span class="at"> </span><span class="dv">0</span></span></code></pre></div><p>and run the stack:</p><pre class="console"><code>docker stack deploy example -c ./example.yml</code></pre><p>After a few seconds docker will spread the parsed stack description@@ -23620,16 +25274,16 @@ volume and have at least two elements, the self-explanatory<code>driver: rclone</code> value and the <code>driver_opts:</code>structure playing the same role as <code>-o key=val</code> CLIflags:</p>-<div class="sourceCode" id="cb927"><pre-class="sourceCode yaml"><code class="sourceCode yaml"><span id="cb927-1"><a href="#cb927-1" aria-hidden="true" tabindex="-1"></a><span class="fu">volumes</span><span class="kw">:</span></span>-<span id="cb927-2"><a href="#cb927-2" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">volume_name_1</span><span class="kw">:</span></span>-<span id="cb927-3"><a href="#cb927-3" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">driver</span><span class="kw">:</span><span class="at"> rclone</span></span>-<span id="cb927-4"><a href="#cb927-4" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">driver_opts</span><span class="kw">:</span></span>-<span id="cb927-5"><a href="#cb927-5" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">remote</span><span class="kw">:</span><span class="at"> </span><span class="st">'gdrive:'</span></span>-<span id="cb927-6"><a href="#cb927-6" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">allow_other</span><span class="kw">:</span><span class="at"> </span><span class="st">'true'</span></span>-<span id="cb927-7"><a href="#cb927-7" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">vfs_cache_mode</span><span class="kw">:</span><span class="at"> full</span></span>-<span id="cb927-8"><a href="#cb927-8" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">token</span><span class="kw">:</span><span class="at"> </span><span class="st">'{"type": "borrower", "expires": "2021-12-31"}'</span></span>-<span id="cb927-9"><a href="#cb927-9" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">poll_interval</span><span class="kw">:</span><span class="at"> </span><span class="dv">0</span></span></code></pre></div>+<div class="sourceCode" id="cb954"><pre+class="sourceCode yaml"><code class="sourceCode yaml"><span id="cb954-1"><a href="#cb954-1" aria-hidden="true" tabindex="-1"></a><span class="fu">volumes</span><span class="kw">:</span></span>+<span id="cb954-2"><a href="#cb954-2" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">volume_name_1</span><span class="kw">:</span></span>+<span id="cb954-3"><a href="#cb954-3" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">driver</span><span class="kw">:</span><span class="at"> rclone</span></span>+<span id="cb954-4"><a href="#cb954-4" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">driver_opts</span><span class="kw">:</span></span>+<span id="cb954-5"><a href="#cb954-5" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">remote</span><span class="kw">:</span><span class="at"> </span><span class="st">'gdrive:'</span></span>+<span id="cb954-6"><a href="#cb954-6" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">allow_other</span><span class="kw">:</span><span class="at"> </span><span class="st">'true'</span></span>+<span id="cb954-7"><a href="#cb954-7" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">vfs_cache_mode</span><span class="kw">:</span><span class="at"> full</span></span>+<span id="cb954-8"><a href="#cb954-8" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">token</span><span class="kw">:</span><span class="at"> </span><span class="st">'{"type": "borrower", "expires": "2021-12-31"}'</span></span>+<span id="cb954-9"><a href="#cb954-9" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">poll_interval</span><span class="kw">:</span><span class="at"> </span><span class="dv">0</span></span></code></pre></div><p>Notice a few important details:</p><ul><li>YAML prefers <code>_</code> in option names instead of@@ -23781,16 +25435,16 @@ docker plugin inspect rclone</code></pre>to inform the docker daemon that a volume is (un-)available. As aworkaround you can setup a healthcheck to verify that the mount isresponding, for example:</p>-<div class="sourceCode" id="cb931"><pre-class="sourceCode yaml"><code class="sourceCode yaml"><span id="cb931-1"><a href="#cb931-1" aria-hidden="true" tabindex="-1"></a><span class="fu">services</span><span class="kw">:</span></span>-<span id="cb931-2"><a href="#cb931-2" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">my_service</span><span class="kw">:</span></span>-<span id="cb931-3"><a href="#cb931-3" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">image</span><span class="kw">:</span><span class="at"> my_image</span></span>-<span id="cb931-4"><a href="#cb931-4" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">healthcheck</span><span class="kw">:</span></span>-<span id="cb931-5"><a href="#cb931-5" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">test</span><span class="kw">:</span><span class="at"> ls /path/to/rclone/mount || exit 1</span></span>-<span id="cb931-6"><a href="#cb931-6" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">interval</span><span class="kw">:</span><span class="at"> 1m</span></span>-<span id="cb931-7"><a href="#cb931-7" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">timeout</span><span class="kw">:</span><span class="at"> 15s</span></span>-<span id="cb931-8"><a href="#cb931-8" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">retries</span><span class="kw">:</span><span class="at"> </span><span class="dv">3</span></span>-<span id="cb931-9"><a href="#cb931-9" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">start_period</span><span class="kw">:</span><span class="at"> 15s</span></span></code></pre></div>+<div class="sourceCode" id="cb958"><pre+class="sourceCode yaml"><code class="sourceCode yaml"><span id="cb958-1"><a href="#cb958-1" aria-hidden="true" tabindex="-1"></a><span class="fu">services</span><span class="kw">:</span></span>+<span id="cb958-2"><a href="#cb958-2" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">my_service</span><span class="kw">:</span></span>+<span id="cb958-3"><a href="#cb958-3" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">image</span><span class="kw">:</span><span class="at"> my_image</span></span>+<span id="cb958-4"><a href="#cb958-4" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">healthcheck</span><span class="kw">:</span></span>+<span id="cb958-5"><a href="#cb958-5" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">test</span><span class="kw">:</span><span class="at"> ls /path/to/rclone/mount || exit 1</span></span>+<span id="cb958-6"><a href="#cb958-6" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">interval</span><span class="kw">:</span><span class="at"> 1m</span></span>+<span id="cb958-7"><a href="#cb958-7" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">timeout</span><span class="kw">:</span><span class="at"> 15s</span></span>+<span id="cb958-8"><a href="#cb958-8" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">retries</span><span class="kw">:</span><span class="at"> </span><span class="dv">3</span></span>+<span id="cb958-9"><a href="#cb958-9" aria-hidden="true" tabindex="-1"></a><span class="at"> </span><span class="fu">start_period</span><span class="kw">:</span><span class="at"> 15s</span></span></code></pre></div><h2 id="running-plugin-under-systemd">Running Plugin under Systemd</h2><p>In most cases you should prefer managed mode. Moreover, MacOS andWindows do not support native Docker plugins. Please use managed mode on@@ -23984,6 +25638,7 @@ Optional Flags:--force Bypass --max-delete safety check and run the sync. Consider using with --verbose-h, --help help for bisync--ignore-listing-checksum Do not use checksums for listings (add --ignore-checksum to additionally skip post-copy checksum checks)+ --max-delete-renames-aware Exclude tracked renames from the --max-delete safety check (requires --track-renames)--max-lock Duration Consider lock files older than this to be expired (default: 0 (never expire)) (minimum: 2m) (default 0s)--no-cleanup Retain working files (useful for troubleshooting and testing).--no-slow-hash Ignore listing checksums only on backends where they are slow@@ -24049,7 +25704,7 @@ href="https://github.com/rclone/rclone/issues/5681#issuecomment-938761815">NOTentirely symmetrical</a>, and more symmetrical options can be specifiedwith the <a href="#resync-mode"><code>--resync-mode</code></a> flag.)Carefully evaluate deltas using <a-href="https://rclone.org/flags/#non-backend-flags">--dry-run</a>.</p>+href="https://rclone.org/docs/#n-dry-run">--dry-run</a>.</p><p>For a resync run, one of the paths may be empty (no files in the pathtree). The resync run should result in files on both paths, else anormal non-resync run will fail.</p>@@ -24205,7 +25860,7 @@ not important, just the names and locations. If you have symbolic linksin your sync tree it is recommended to place <code>RCLONE_TEST</code>files in the linked-to directory tree to protect against bisync assuminga bunch of deleted files if the linked-to tree should not be accessible.-See also the <a href="--check-filename">--check-filename</a> flag.</p>+See also the <a href="#check-filename">--check-filename</a> flag.</p><h3 id="check-filename">--check-filename</h3><p>Name of the file(s) used in access health validation. The default<code>--check-filename</code> is <code>RCLONE_TEST</code>. One or more@@ -24380,6 +26035,25 @@ different delete percentage limit, e.g. <code>--max-delete 75</code>check.</p><p>Also see the <a href="#all-files-changed">all files changed</a>check.</p>+<h3 id="max-delete-renames-aware">--max-delete-renames-aware</h3>+<p>When <code>--track-renames</code> is in use,+<code>--max-delete-renames-aware</code> can exclude files that are+guaranteed to match as tracked renames from the deletion count. If the+initial deletion count exceeds <code>--max-delete</code>, bisync+performs a rename matching preflight before starting either sync+direction. The run still aborts if the remaining unmatched deletions+exceed the limit.</p>+<p>The preflight uses the selected <a+href="https://rclone.org/docs/#track-renames-strategy"><code>--track-renames-strategy</code></a>+and the attributes stored in the normal bisync listings. A hash strategy+forces both listings to use a common hash and is incompatible with+<code>--ignore-listing-checksum</code>. The option only exempts matches+when the destination supports server-side moves or copies and the+selected strategy is supported by both paths. It requires+<code>--track-renames</code> and is disabled by default because hashing+may add listing time. <code>--force</code> bypasses both the safety+check and this preflight. The preflight is also disabled during+<code>--resync</code>, when rename tracking is unavailable.</p><h3 id="filters-file">--filters-file</h3><p>By using rclone filter features you can exclude file types ordirectory sub-trees from the sync. See the <a href="#filtering">bisync@@ -24591,7 +26265,8 @@ the run. This is normal, and any differences will be detected and syncedon the next run.</p><p>For a robust integrity check of the current state of the remotes (asopposed to just their listing snapshots), consider using <a-href="commands/rclone_check/"><code>check</code></a> (or <a+href="https://rclone.org/commands/rclone_check/"><code>check</code></a>+(or <ahref="https://rclone.org/commands/rclone_cryptcheck/"><code>cryptcheck</code></a>,if at least one path is a <code>crypt</code> remote) instead of<code>--check-sync</code>, keeping in mind that differences are expected@@ -24935,10 +26610,10 @@ resulted in unnecessary renames and duplicates. Now, when bisync comesto a file that it wants to rename (because it is new/changed on bothsides), it first checks whether the Path1 and Path2 versions arecurrently <em>identical</em> (using the same underlying function as <a-href="commands/rclone_check/"><code>check</code></a>.) If bisync-concludes that the files are identical, it will skip them and move on.-Otherwise, it will create renamed duplicates, as before. This behavior-also <a+href="https://rclone.org/commands/rclone_check/"><code>check</code></a>.)+If bisync concludes that the files are identical, it will skip them and+move on. Otherwise, it will create renamed duplicates, as before. This+behavior also <ahref="https://forum.rclone.org/t/bisync-bugs-and-feature-requests/37636#:~:text=Renamed%20directories">improvesthe experience of renaming directories</a>, as a <code>--resync</code>is no longer required, so long as the same change has been made on both@@ -24953,7 +26628,7 @@ timestamp files wins). Alternately, a <code>--resync</code> may be used(Path1 versions will be pushed to Path2). Consider the situationcarefully and perhaps use <code>--dry-run</code> before you commit tothe changes.</p>-<h3 id="modification-times">Modification times</h3>+<h3 id="modification-times-1">Modification times</h3><p>By default, bisync compares files by modification time and size. Ifyou or your application should change the content of a file withoutchanging the modification time and size, then bisync will <em>not</em>@@ -25061,22 +26736,35 @@ href="https://pub.rclone.org/integration-tests/current/huaweidrive-cmd.bisync-Te<li><ahref="https://pub.rclone.org/integration-tests/current/huaweidrive-cmd.bisync-TestHuaweiDrive-1.txt"><code>TestBisyncRemoteLocal/extended_filenames</code></a></li><li><a-href="https://pub.rclone.org/integration-tests/current/huaweidrive-cmd.bisync-TestHuaweiDrive-1.txt"><code>TestBisyncRemoteLocal/normalization</code></a></li>-<li><ahref="https://pub.rclone.org/integration-tests/current/huaweidrive-cmd.bisync-TestHuaweiDrive-1.txt"><code>TestBisyncLocalRemote/ext_paths</code></a></li><li><ahref="https://pub.rclone.org/integration-tests/current/huaweidrive-cmd.bisync-TestHuaweiDrive-1.txt"><code>TestBisyncLocalRemote/extended_filenames</code></a></li>-<li><a href="https://pub.rclone.org/integration-tests/current/">4+<li><a+href="https://pub.rclone.org/integration-tests/current/huaweidrive-cmd.bisync-TestHuaweiDrive-1.txt"><code>TestBisyncLocalRemote/normalization</code></a></li>+<li><a href="https://pub.rclone.org/integration-tests/current/">3more</a></li></ul></li><li><code>TestPcloud</code> (<code>pcloud</code>)<ul><li><a-href="https://pub.rclone.org/integration-tests/current/pcloud-cmd.bisync-TestPcloud-1.txt"><code>TestBisyncRemoteRemote/check_access</code></a></li>+href="https://pub.rclone.org/integration-tests/current/pcloud-cmd.bisync-TestPcloud-1.txt"><code>TestBisyncRemoteRemote/extended_filenames</code></a></li>+</ul></li>+<li><code>TestYandex</code> (<code>yandex</code>)+<ul>+<li><a+href="https://pub.rclone.org/integration-tests/current/yandex-cmd.bisync-TestYandex-1.txt"><code>TestBisyncRemoteLocal/all_changed</code></a></li>+<li><a+href="https://pub.rclone.org/integration-tests/current/yandex-cmd.bisync-TestYandex-1.txt"><code>TestBisyncRemoteLocal/backupdir</code></a></li>+<li><a+href="https://pub.rclone.org/integration-tests/current/yandex-cmd.bisync-TestYandex-1.txt"><code>TestBisyncRemoteLocal/basic</code></a></li>+<li><a+href="https://pub.rclone.org/integration-tests/current/yandex-cmd.bisync-TestYandex-1.txt"><code>TestBisyncRemoteLocal/changes</code></a></li><li><a-href="https://pub.rclone.org/integration-tests/current/pcloud-cmd.bisync-TestPcloud-1.txt"><code>TestBisyncRemoteRemote/rmdirs</code></a></li>+href="https://pub.rclone.org/integration-tests/current/yandex-cmd.bisync-TestYandex-1.txt"><code>TestBisyncRemoteLocal/check_access</code></a></li>+<li><a href="https://pub.rclone.org/integration-tests/current/">94+more</a></li></ul></li>-<li>Updated: 2026-07-31-010017+<li>Updated: 2026-11-06-010024<!--- end list_failures - DO NOT EDIT THIS SECTION - use make commanddocs ---></li></ul><p>The following backends either have not been tested recently or have@@ -25096,6 +26784,10 @@ known issues that are deemed unfixable for the time being:</p><li><code>TestFilesCom</code> (<code>filescom</code>)</li><li><code>TestImageKit</code> (<code>imagekit</code>)</li><li><code>TestJottacloud</code> (<code>jottacloud</code>)</li>+<li><code>TestKdrive</code> (<code>kdrive</code>)+<ul>+<li><code>TestBisyncRemoteRemote/normalization</code></li>+</ul></li><li><code>TestLinkbox</code> (<code>linkbox</code>)</li><li><code>TestMailru</code> (<code>mailru</code>)</li><li><code>TestMega</code> (<code>mega</code>)</li>@@ -25110,8 +26802,9 @@ known issues that are deemed unfixable for the time being:</p><li><code>TestS3GCS</code> (<code>s3</code>)<ul><li><code>TestBisyncRemoteRemote/extended_filenames</code></li>+<li><code>TestBisyncRemoteLocal/extended_filenames</code></li>+<li><code>TestBisyncLocalRemote/extended_filenames</code></li></ul></li>-<li><code>TestS3Rclone</code> (<code>s3</code>)</li><li><code>TestSFTPRsyncNet</code> (<code>sftp</code>)</li><li><code>TestStorj</code> (<code>storj</code>)</li><li><code>TestWebdavNextcloud</code> (<code>webdav</code>)</li>@@ -25199,11 +26892,12 @@ directory is to rename it to the same name on both sides. (As of<code>rclone v1.64</code>, a <code>--resync</code> is no longer requiredafter doing so, as bisync will automatically detect that Path1 and Path2are in agreement.)</p>-<p>Note that although the flag --track-renames ensures that-renamed/moved files won't be deleted and uploaded again, they are still-counted as deleted files for purposes of the --max-delete flag (as this-check happens before the rename detection operation). See <a-href="https://github.com/rclone/rclone/issues/8685">this issue</a>.</p>+<p>By default, renamed or moved files are still counted as deleted files+for purposes of <code>--max-delete</code>, because the safety check+happens before rename detection. Use+<code>--max-delete-renames-aware</code> together with+<code>--track-renames</code> to run a rename preflight and exclude+guaranteed matches from this count.</p><h3 id="fast-list-used-by-default"><code>--fast-list</code> used bydefault</h3><p>Unlike most other rclone commands, bisync uses <a@@ -25640,19 +27334,19 @@ versions I manually run the following command:<li>The Dropbox client then syncs the changes with Dropbox.</li></ul><h3 id="rclone-conf-snippet">rclone.conf snippet</h3>-<div class="sourceCode" id="cb964"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb964-1"><a href="#cb964-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[Dropbox]</span></span>-<span id="cb964-2"><a href="#cb964-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> dropbox</span></span>-<span id="cb964-3"><a href="#cb964-3" aria-hidden="true" tabindex="-1"></a><span class="dt">...</span></span>-<span id="cb964-4"><a href="#cb964-4" aria-hidden="true" tabindex="-1"></a></span>-<span id="cb964-5"><a href="#cb964-5" aria-hidden="true" tabindex="-1"></a><span class="kw">[Dropcrypt]</span></span>-<span id="cb964-6"><a href="#cb964-6" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> crypt</span></span>-<span id="cb964-7"><a href="#cb964-7" aria-hidden="true" tabindex="-1"></a><span class="dt">remote </span><span class="ot">=</span><span class="st"> /path/to/DBoxroot/crypt # on the Linux server</span></span>-<span id="cb964-8"><a href="#cb964-8" aria-hidden="true" tabindex="-1"></a><span class="dt">remote </span><span class="ot">=</span><span class="st"> C:\Users\MyLogin\Dropbox\crypt # on the Windows notebook</span></span>-<span id="cb964-9"><a href="#cb964-9" aria-hidden="true" tabindex="-1"></a><span class="dt">filename_encryption </span><span class="ot">=</span><span class="st"> standard</span></span>-<span id="cb964-10"><a href="#cb964-10" aria-hidden="true" tabindex="-1"></a><span class="dt">directory_name_encryption </span><span class="ot">=</span><span class="st"> </span><span class="kw">true</span></span>-<span id="cb964-11"><a href="#cb964-11" aria-hidden="true" tabindex="-1"></a><span class="dt">password </span><span class="ot">=</span><span class="st"> ...</span></span>-<span id="cb964-12"><a href="#cb964-12" aria-hidden="true" tabindex="-1"></a><span class="dt">...</span></span></code></pre></div>+<div class="sourceCode" id="cb991"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb991-1"><a href="#cb991-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[Dropbox]</span></span>+<span id="cb991-2"><a href="#cb991-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> dropbox</span></span>+<span id="cb991-3"><a href="#cb991-3" aria-hidden="true" tabindex="-1"></a><span class="dt">...</span></span>+<span id="cb991-4"><a href="#cb991-4" aria-hidden="true" tabindex="-1"></a></span>+<span id="cb991-5"><a href="#cb991-5" aria-hidden="true" tabindex="-1"></a><span class="kw">[Dropcrypt]</span></span>+<span id="cb991-6"><a href="#cb991-6" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> crypt</span></span>+<span id="cb991-7"><a href="#cb991-7" aria-hidden="true" tabindex="-1"></a><span class="dt">remote </span><span class="ot">=</span><span class="st"> /path/to/DBoxroot/crypt # on the Linux server</span></span>+<span id="cb991-8"><a href="#cb991-8" aria-hidden="true" tabindex="-1"></a><span class="dt">remote </span><span class="ot">=</span><span class="st"> C:\Users\MyLogin\Dropbox\crypt # on the Windows notebook</span></span>+<span id="cb991-9"><a href="#cb991-9" aria-hidden="true" tabindex="-1"></a><span class="dt">filename_encryption </span><span class="ot">=</span><span class="st"> standard</span></span>+<span id="cb991-10"><a href="#cb991-10" aria-hidden="true" tabindex="-1"></a><span class="dt">directory_name_encryption </span><span class="ot">=</span><span class="st"> </span><span class="kw">true</span></span>+<span id="cb991-11"><a href="#cb991-11" aria-hidden="true" tabindex="-1"></a><span class="dt">password </span><span class="ot">=</span><span class="st"> ...</span></span>+<span id="cb991-12"><a href="#cb991-12" aria-hidden="true" tabindex="-1"></a><span class="dt">...</span></span></code></pre></div><h2 id="testing">Testing</h2><p>You should read this section only if you are developing for rclone.You need to have rclone source code locally to work with bisync@@ -25990,146 +27684,11 @@ article</a>.</p>href="http://www.cis.upenn.edu/%7Ebcpierce/papers/index.shtml#File%20Synchronization">BenjaminPierce</a> about <em>Unison</em> and synchronization in general.</p><h2 id="changelog">Changelog</h2>-<h3 id="v1.74.2"><code>v1.74.2</code></h3>-<ul>-<li>Fixed an issue causing <code>--conflict-loser pathname</code> to-produce unexpected behavior if using a non-default-<code>--conflict-resolve</code> value.</li>-</ul>-<h3 id="v1.74"><code>v1.74</code></h3>-<ul>-<li>Added several missing <code>rc</code> parameters.</li>-<li>Optional <code>rc</code> parameters are now truly optional.</li>-<li><code>rc</code> output now provides more structured-information.</li>-</ul>-<h3 id="v1.71"><code>v1.71</code></h3>-<ul>-<li><p><code>bisync</code> is now officially released from-beta.</p></li>-<li><p>Fixed an issue causing hash type to be set incorrectly if-<code>path2</code> had slow hashes and <code>--no-slow-hash</code> or-<code>--slow-hash-sync-only</code> was in use.</p></li>-<li><p>Fixed an issue causing errors when running concurrent bisync runs-through the <code>rc</code>.</p></li>-</ul>-<h3 id="v1.69.1"><code>v1.69.1</code></h3>-<ul>-<li>Fixed an issue causing listings to not capture concurrent-modifications under certain conditions</li>-</ul>-<h3 id="v1.68"><code>v1.68</code></h3>-<ul>-<li>Fixed an issue affecting backends that round modtimes to a lower-precision.</li>-</ul>-<h3 id="v1.67"><code>v1.67</code></h3>-<ul>-<li>Added integration tests against all backends.</li>-</ul>-<h3 id="v1.66"><code>v1.66</code></h3>-<ul>-<li><p>Copies and deletes are now handled in one operation instead of-two</p></li>-<li><p><code>--track-renames</code> and <code>--backup-dir</code> are-now supported</p></li>-<li><p>Partial uploads known issue on-<code>local</code>/<code>ftp</code>/<code>sftp</code> has been resolved-(unless using <code>--inplace</code>)</p></li>-<li><p>Final listings are now generated from sync results, to avoid-needing to re-list</p></li>-<li><p>Bisync is now much more resilient to changes that happen during a-bisync run, and far less prone to critical errors / undetected-changes</p></li>-<li><p>Bisync is now capable of rolling a file listing back in cases of-uncertainty, essentially marking the file as needing to be rechecked-next time.</p></li>-<li><p>A few basic terminal colors are now supported, controllable with-<a href="https://rclone.org/docs/#color"><code>--color</code></a>-(<code>AUTO</code>|<code>NEVER</code>|<code>ALWAYS</code>)</p></li>-<li><p>Initial listing snapshots of Path1 and Path2 are now generated-concurrently, using the same "march" infrastructure as-<code>check</code> and <code>sync</code>, for performance improvements-and less <a-href="https://forum.rclone.org/t/bisync-bugs-and-feature-requests/37636#:~:text=4.%20Listings%20should%20alternate%20between%20paths%20to%20minimize%20errors">risk-of error</a>.</p></li>-<li><p>Fixed handling of unicode normalization and case insensitivity,-support for <a-href="https://rclone.org/docs/#fix-case"><code>--fix-case</code></a>, <a-href="/docs/#ignore-case-sync"><code>--ignore-case-sync</code></a>, <a-href="https://rclone.org/docs/#no-unicode-normalization"><code>--no-unicode-normalization</code></a></p></li>-<li><p><code>--resync</code> is now much more efficient (especially for-users of <code>--create-empty-src-dirs</code>)</p></li>-<li><p>Google Docs (and other files of unknown size) are now supported-(with the same options as in <code>sync</code>)</p></li>-<li><p>Equality checks before a sync conflict rename now fall back to-<code>cryptcheck</code> (when possible) or <code>--download</code>,-instead of of <code>--size-only</code>, when <code>check</code> is not-available.</p></li>-<li><p>Bisync no longer fails to find the correct listing file when-configs are overridden with backend-specific flags.</p></li>-<li><p>Bisync now fully supports comparing based on any combination of-size, modtime, and checksum, lifting the prior restriction on backends-without modtime support.</p></li>-<li><p>Bisync now supports a "Graceful Shutdown" mode to cleanly cancel-a run early without requiring <code>--resync</code>.</p></li>-<li><p>New <code>--recover</code> flag allows robust recovery in the-event of interruptions, without requiring-<code>--resync</code>.</p></li>-<li><p>A new <code>--max-lock</code> setting allows lock files to-automatically renew and expire, for better automatic recovery when a run-is interrupted.</p></li>-<li><p>Bisync now supports auto-resolving sync conflicts and customizing-rename behavior with new <a-href="#conflict-resolve"><code>--conflict-resolve</code></a>, <a-href="#conflict-loser"><code>--conflict-loser</code></a>, and <a-href="#conflict-suffix"><code>--conflict-suffix</code></a>-flags.</p></li>-<li><p>A new <a href="#resync-mode"><code>--resync-mode</code></a> flag-allows more control over which version of a file gets kept during a-<code>--resync</code>.</p></li>-<li><p>Bisync now supports <a-href="https://rclone.org/docs/#retries-int"><code>--retries</code></a>-and <a href="/docs/#retries-sleep-time"><code>--retries-sleep</code></a>-(when <a href="#resilient"><code>--resilient</code></a> is-set.)</p></li>-</ul>-<h3 id="v1.64"><code>v1.64</code></h3>-<ul>-<li><p>Fixed an <a-href="https://forum.rclone.org/t/bisync-bugs-and-feature-requests/37636#:~:text=1.%20Dry%20runs%20are%20not%20completely%20dry">issue</a>-causing dry runs to inadvertently commit filter changes</p></li>-<li><p>Fixed an <a-href="https://forum.rclone.org/t/bisync-bugs-and-feature-requests/37636#:~:text=2.%20%2D%2Dresync%20deletes%20data%2C%20contrary%20to%20docs">issue</a>-causing <code>--resync</code> to erroneously delete empty folders and-duplicate files unique to Path2</p></li>-<li><p><code>--check-access</code> is now enforced during-<code>--resync</code>, preventing data loss in <a-href="https://forum.rclone.org/t/bisync-bugs-and-feature-requests/37636#:~:text=%2D%2Dcheck%2Daccess%20doesn%27t%20always%20fail%20when%20it%20should">certain-user error scenarios</a></p></li>-<li><p>Fixed an <a-href="https://forum.rclone.org/t/bisync-bugs-and-feature-requests/37636#:~:text=5.%20Bisync%20reads%20files%20in%20excluded%20directories%20during%20delete%20operations">issue</a>-causing bisync to consider more files than necessary due to overbroad-filters during delete operations</p></li>-<li><p><a-href="https://forum.rclone.org/t/bisync-bugs-and-feature-requests/37636#:~:text=1.%20Identical%20files%20should%20be%20left%20alone%2C%20even%20if%20new/newer/changed%20on%20both%20sides">Improved-detection of false positive change conflicts</a> (identical files are-now left alone instead of renamed)</p></li>-<li><p>Added <a-href="https://forum.rclone.org/t/bisync-bugs-and-feature-requests/37636#:~:text=3.%20Bisync%20should%20create/delete%20empty%20directories%20as%20sync%20does%2C%20when%20%2D%2Dcreate%2Dempty%2Dsrc%2Ddirs%20is%20passed">support-for <code>--create-empty-src-dirs</code></a></p></li>-<li><p>Added experimental <code>--resilient</code> mode to allow <a-href="https://forum.rclone.org/t/bisync-bugs-and-feature-requests/37636#:~:text=2.%20Bisync%20should%20be%20more%20resilient%20to%20self%2Dcorrectable%20errors">recovery-from self-correctable errors</a></p></li>-<li><p>Added <a-href="https://forum.rclone.org/t/bisync-bugs-and-feature-requests/37636#:~:text=6.%20%2D%2Dignore%2Dchecksum%20should%20be%20split%20into%20two%20flags%20for%20separate%20purposes">new-<code>--ignore-listing-checksum</code> flag</a> to distinguish from-<code>--ignore-checksum</code></p></li>-<li><p><a-href="https://forum.rclone.org/t/bisync-bugs-and-feature-requests/37636#:~:text=6.%20Deletes%20take%20several%20times%20longer%20than%20copies">Performance-improvements</a> for large remotes</p></li>-<li><p>Documentation and testing improvements</p></li>-</ul>+<p>Changes to bisync are listed in the <a+href="https://rclone.org/changelog/">rclone changelog</a>. The changelog+this page kept until <code>v1.76</code> is <a+href="https://github.com/rclone/rclone/blob/0891ce30606c726990c98d70812d31b555782791/docs/content/bisync.md?plain=1#L1934-L2061">archived+on GitHub</a>.</p><h1 id="release-signing">Release signing</h1><p>The hashes of the binary artefacts of the rclone release are signedwith a public PGP/GPG key. This can be verified manually as described@@ -26232,14 +27791,13 @@ desired.</p><h3 id="verify-the-hashes">Verify the hashes</h3><p>Now that we know the signatures on the hashes are OK we can verifythe binaries match the hashes, completing the verification.</p>-<pre class="console"><code>$ sha256sum -c SHA256SUMS 2>&1 | grep OK+<p>Extract the signed data with:</p>+<pre class="console"><code>$ gpg --output SHA256SUMS.signed --decrypt SHA256SUMS</code></pre>+<p>Then check the binaries:</p>+<pre class="console"><code>$ sha256sum -c --ignore-missing SHA256SUMS.signedrclone-v1.63.1-windows-amd64.zip: OK</code></pre><p>Or do the check with rclone</p>-<pre class="console"><code>$ rclone hashsum sha256 -C SHA256SUMS rclone-v1.63.1-windows-amd64.zip-2023/09/11 10:53:58 NOTICE: SHA256SUMS: improperly formatted checksum line 0-2023/09/11 10:53:58 NOTICE: SHA256SUMS: improperly formatted checksum line 1-2023/09/11 10:53:58 NOTICE: SHA256SUMS: improperly formatted checksum line 49-2023/09/11 10:53:58 NOTICE: SHA256SUMS: 4 warning(s) suppressed...+<pre class="console"><code>$ rclone hashsum sha256 -C SHA256SUMS.signed rclone-v1.63.1-windows-amd64.zip= rclone-v1.63.1-windows-amd64.zip2023/09/11 10:53:58 NOTICE: Local file system at /tmp/check: 0 differences found2023/09/11 10:53:58 NOTICE: Local file system at /tmp/check: 1 matching files</code></pre>@@ -26597,8 +28155,10 @@ existing remote).</p><li>Alibaba Cloud (Aliyun) Object Storage System (OSS)</li><li>Ceph</li><li>China Mobile Ecloud Elastic Object Storage (EOS)</li>+<li>Clever Cloud Cellar</li><li>Cloudflare R2</li><li>Arvan Cloud Object Storage (AOS)</li>+<li>Autonomys Auto Drive</li><li>Bizfly Cloud Simple Storage</li><li>Cubbit DS3</li><li>DigitalOcean Spaces</li>@@ -26613,6 +28173,7 @@ existing remote).</p><li>IBM COS S3</li><li>IDrive e2</li><li>Impossible Cloud</li>+<li>Infomaniak Swiss Backup</li><li>Intercolo Object Storage</li><li>IONOS Cloud</li><li>Leviia Object Storage</li>@@ -26621,6 +28182,7 @@ existing remote).</p><li>Magalu Object Storage</li><li>MEGA S4 Object Storage</li><li>Minio</li>+<li>Neon Object Storage</li><li>Outscale</li><li>OVHcloud</li><li>Petabox</li>@@ -26629,6 +28191,7 @@ existing remote).</p><li>Rabata Cloud Storage</li><li>RackCorp Object Storage</li><li>Rclone Serve S3</li>+<li>RelAix.Storage</li><li>Scaleway</li><li>Scality (RING / ARTESCA)</li><li>Seagate Lyve Cloud</li>@@ -26863,7 +28426,7 @@ d) Delete this remotey/e/d></code></pre><h3 id="modification-times-and-hashes-1">Modification times andhashes</h3>-<h4 id="modification-times-1">Modification times</h4>+<h4 id="modification-times-2">Modification times</h4><p>The modified time is stored as metadata on the object as<code>X-Amz-Meta-Mtime</code> as floating point since the epoch,accurate to 1 ns.</p>@@ -27345,24 +28908,24 @@ An external ID is provided for additional security as required by therole's trust policy</p><p>The target role's trust policy in the destination account must allowthe source account or user to assume it. Example trust policy:</p>-<div class="sourceCode" id="cb1000"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb1000-1"><a href="#cb1000-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>-<span id="cb1000-2"><a href="#cb1000-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Version"</span><span class="fu">:</span> <span class="st">"2012-10-17"</span><span class="fu">,</span></span>-<span id="cb1000-3"><a href="#cb1000-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Statement"</span><span class="fu">:</span> <span class="ot">[</span></span>-<span id="cb1000-4"><a href="#cb1000-4" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>-<span id="cb1000-5"><a href="#cb1000-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Effect"</span><span class="fu">:</span> <span class="st">"Allow"</span><span class="fu">,</span></span>-<span id="cb1000-6"><a href="#cb1000-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Principal"</span><span class="fu">:</span> <span class="fu">{</span></span>-<span id="cb1000-7"><a href="#cb1000-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"AWS"</span><span class="fu">:</span> <span class="st">"arn:aws:iam::SOURCE-ACCOUNT-ID:root"</span></span>-<span id="cb1000-8"><a href="#cb1000-8" aria-hidden="true" tabindex="-1"></a> <span class="fu">},</span></span>-<span id="cb1000-9"><a href="#cb1000-9" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Action"</span><span class="fu">:</span> <span class="st">"sts:AssumeRole"</span><span class="fu">,</span></span>-<span id="cb1000-10"><a href="#cb1000-10" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Condition"</span><span class="fu">:</span> <span class="fu">{</span></span>-<span id="cb1000-11"><a href="#cb1000-11" aria-hidden="true" tabindex="-1"></a> <span class="dt">"StringEquals"</span><span class="fu">:</span> <span class="fu">{</span></span>-<span id="cb1000-12"><a href="#cb1000-12" aria-hidden="true" tabindex="-1"></a> <span class="dt">"sts:ExternalID"</span><span class="fu">:</span> <span class="st">"unique-role-external-id-12345"</span></span>-<span id="cb1000-13"><a href="#cb1000-13" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>-<span id="cb1000-14"><a href="#cb1000-14" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>-<span id="cb1000-15"><a href="#cb1000-15" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>-<span id="cb1000-16"><a href="#cb1000-16" aria-hidden="true" tabindex="-1"></a> <span class="ot">]</span></span>-<span id="cb1000-17"><a href="#cb1000-17" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<div class="sourceCode" id="cb1028"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb1028-1"><a href="#cb1028-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb1028-2"><a href="#cb1028-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Version"</span><span class="fu">:</span> <span class="st">"2012-10-17"</span><span class="fu">,</span></span>+<span id="cb1028-3"><a href="#cb1028-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Statement"</span><span class="fu">:</span> <span class="ot">[</span></span>+<span id="cb1028-4"><a href="#cb1028-4" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>+<span id="cb1028-5"><a href="#cb1028-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Effect"</span><span class="fu">:</span> <span class="st">"Allow"</span><span class="fu">,</span></span>+<span id="cb1028-6"><a href="#cb1028-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Principal"</span><span class="fu">:</span> <span class="fu">{</span></span>+<span id="cb1028-7"><a href="#cb1028-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"AWS"</span><span class="fu">:</span> <span class="st">"arn:aws:iam::SOURCE-ACCOUNT-ID:root"</span></span>+<span id="cb1028-8"><a href="#cb1028-8" aria-hidden="true" tabindex="-1"></a> <span class="fu">},</span></span>+<span id="cb1028-9"><a href="#cb1028-9" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Action"</span><span class="fu">:</span> <span class="st">"sts:AssumeRole"</span><span class="fu">,</span></span>+<span id="cb1028-10"><a href="#cb1028-10" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Condition"</span><span class="fu">:</span> <span class="fu">{</span></span>+<span id="cb1028-11"><a href="#cb1028-11" aria-hidden="true" tabindex="-1"></a> <span class="dt">"StringEquals"</span><span class="fu">:</span> <span class="fu">{</span></span>+<span id="cb1028-12"><a href="#cb1028-12" aria-hidden="true" tabindex="-1"></a> <span class="dt">"sts:ExternalID"</span><span class="fu">:</span> <span class="st">"unique-role-external-id-12345"</span></span>+<span id="cb1028-13"><a href="#cb1028-13" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>+<span id="cb1028-14"><a href="#cb1028-14" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>+<span id="cb1028-15"><a href="#cb1028-15" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>+<span id="cb1028-16"><a href="#cb1028-16" aria-hidden="true" tabindex="-1"></a> <span class="ot">]</span></span>+<span id="cb1028-17"><a href="#cb1028-17" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div><h3 id="s3-permissions">S3 Permissions</h3><p>When using the <code>sync</code> subcommand of <code>rclone</code>the following minimum permissions are required to be available on the@@ -27379,34 +28942,34 @@ href="#s3-no-check-bucket">s3-no-check-bucket</a>)</li><p>When using the <code>lsd</code> subcommand, the<code>ListAllMyBuckets</code> permission is required.</p><p>Example policy:</p>-<div class="sourceCode" id="cb1001"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb1001-1"><a href="#cb1001-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>-<span id="cb1001-2"><a href="#cb1001-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Version"</span><span class="fu">:</span> <span class="st">"2012-10-17"</span><span class="fu">,</span></span>-<span id="cb1001-3"><a href="#cb1001-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Statement"</span><span class="fu">:</span> <span class="ot">[</span></span>-<span id="cb1001-4"><a href="#cb1001-4" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>-<span id="cb1001-5"><a href="#cb1001-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Effect"</span><span class="fu">:</span> <span class="st">"Allow"</span><span class="fu">,</span></span>-<span id="cb1001-6"><a href="#cb1001-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Principal"</span><span class="fu">:</span> <span class="fu">{</span></span>-<span id="cb1001-7"><a href="#cb1001-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"AWS"</span><span class="fu">:</span> <span class="st">"arn:aws:iam::USER_SID:user/USER_NAME"</span></span>-<span id="cb1001-8"><a href="#cb1001-8" aria-hidden="true" tabindex="-1"></a> <span class="fu">},</span></span>-<span id="cb1001-9"><a href="#cb1001-9" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Action"</span><span class="fu">:</span> <span class="ot">[</span></span>-<span id="cb1001-10"><a href="#cb1001-10" aria-hidden="true" tabindex="-1"></a> <span class="st">"s3:ListBucket"</span><span class="ot">,</span></span>-<span id="cb1001-11"><a href="#cb1001-11" aria-hidden="true" tabindex="-1"></a> <span class="st">"s3:DeleteObject"</span><span class="ot">,</span></span>-<span id="cb1001-12"><a href="#cb1001-12" aria-hidden="true" tabindex="-1"></a> <span class="st">"s3:GetObject"</span><span class="ot">,</span></span>-<span id="cb1001-13"><a href="#cb1001-13" aria-hidden="true" tabindex="-1"></a> <span class="st">"s3:PutObject"</span><span class="ot">,</span></span>-<span id="cb1001-14"><a href="#cb1001-14" aria-hidden="true" tabindex="-1"></a> <span class="st">"s3:PutObjectAcl"</span></span>-<span id="cb1001-15"><a href="#cb1001-15" aria-hidden="true" tabindex="-1"></a> <span class="ot">]</span><span class="fu">,</span></span>-<span id="cb1001-16"><a href="#cb1001-16" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Resource"</span><span class="fu">:</span> <span class="ot">[</span></span>-<span id="cb1001-17"><a href="#cb1001-17" aria-hidden="true" tabindex="-1"></a> <span class="st">"arn:aws:s3:::BUCKET_NAME/*"</span><span class="ot">,</span></span>-<span id="cb1001-18"><a href="#cb1001-18" aria-hidden="true" tabindex="-1"></a> <span class="st">"arn:aws:s3:::BUCKET_NAME"</span></span>-<span id="cb1001-19"><a href="#cb1001-19" aria-hidden="true" tabindex="-1"></a> <span class="ot">]</span></span>-<span id="cb1001-20"><a href="#cb1001-20" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span><span class="ot">,</span></span>-<span id="cb1001-21"><a href="#cb1001-21" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>-<span id="cb1001-22"><a href="#cb1001-22" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Effect"</span><span class="fu">:</span> <span class="st">"Allow"</span><span class="fu">,</span></span>-<span id="cb1001-23"><a href="#cb1001-23" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Action"</span><span class="fu">:</span> <span class="st">"s3:ListAllMyBuckets"</span><span class="fu">,</span></span>-<span id="cb1001-24"><a href="#cb1001-24" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Resource"</span><span class="fu">:</span> <span class="st">"arn:aws:s3:::*"</span></span>-<span id="cb1001-25"><a href="#cb1001-25" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>-<span id="cb1001-26"><a href="#cb1001-26" aria-hidden="true" tabindex="-1"></a> <span class="ot">]</span></span>-<span id="cb1001-27"><a href="#cb1001-27" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<div class="sourceCode" id="cb1029"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb1029-1"><a href="#cb1029-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb1029-2"><a href="#cb1029-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Version"</span><span class="fu">:</span> <span class="st">"2012-10-17"</span><span class="fu">,</span></span>+<span id="cb1029-3"><a href="#cb1029-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Statement"</span><span class="fu">:</span> <span class="ot">[</span></span>+<span id="cb1029-4"><a href="#cb1029-4" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>+<span id="cb1029-5"><a href="#cb1029-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Effect"</span><span class="fu">:</span> <span class="st">"Allow"</span><span class="fu">,</span></span>+<span id="cb1029-6"><a href="#cb1029-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Principal"</span><span class="fu">:</span> <span class="fu">{</span></span>+<span id="cb1029-7"><a href="#cb1029-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"AWS"</span><span class="fu">:</span> <span class="st">"arn:aws:iam::USER_SID:user/USER_NAME"</span></span>+<span id="cb1029-8"><a href="#cb1029-8" aria-hidden="true" tabindex="-1"></a> <span class="fu">},</span></span>+<span id="cb1029-9"><a href="#cb1029-9" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Action"</span><span class="fu">:</span> <span class="ot">[</span></span>+<span id="cb1029-10"><a href="#cb1029-10" aria-hidden="true" tabindex="-1"></a> <span class="st">"s3:ListBucket"</span><span class="ot">,</span></span>+<span id="cb1029-11"><a href="#cb1029-11" aria-hidden="true" tabindex="-1"></a> <span class="st">"s3:DeleteObject"</span><span class="ot">,</span></span>+<span id="cb1029-12"><a href="#cb1029-12" aria-hidden="true" tabindex="-1"></a> <span class="st">"s3:GetObject"</span><span class="ot">,</span></span>+<span id="cb1029-13"><a href="#cb1029-13" aria-hidden="true" tabindex="-1"></a> <span class="st">"s3:PutObject"</span><span class="ot">,</span></span>+<span id="cb1029-14"><a href="#cb1029-14" aria-hidden="true" tabindex="-1"></a> <span class="st">"s3:PutObjectAcl"</span></span>+<span id="cb1029-15"><a href="#cb1029-15" aria-hidden="true" tabindex="-1"></a> <span class="ot">]</span><span class="fu">,</span></span>+<span id="cb1029-16"><a href="#cb1029-16" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Resource"</span><span class="fu">:</span> <span class="ot">[</span></span>+<span id="cb1029-17"><a href="#cb1029-17" aria-hidden="true" tabindex="-1"></a> <span class="st">"arn:aws:s3:::BUCKET_NAME/*"</span><span class="ot">,</span></span>+<span id="cb1029-18"><a href="#cb1029-18" aria-hidden="true" tabindex="-1"></a> <span class="st">"arn:aws:s3:::BUCKET_NAME"</span></span>+<span id="cb1029-19"><a href="#cb1029-19" aria-hidden="true" tabindex="-1"></a> <span class="ot">]</span></span>+<span id="cb1029-20"><a href="#cb1029-20" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span><span class="ot">,</span></span>+<span id="cb1029-21"><a href="#cb1029-21" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>+<span id="cb1029-22"><a href="#cb1029-22" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Effect"</span><span class="fu">:</span> <span class="st">"Allow"</span><span class="fu">,</span></span>+<span id="cb1029-23"><a href="#cb1029-23" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Action"</span><span class="fu">:</span> <span class="st">"s3:ListAllMyBuckets"</span><span class="fu">,</span></span>+<span id="cb1029-24"><a href="#cb1029-24" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Resource"</span><span class="fu">:</span> <span class="st">"arn:aws:s3:::*"</span></span>+<span id="cb1029-25"><a href="#cb1029-25" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>+<span id="cb1029-26"><a href="#cb1029-26" aria-hidden="true" tabindex="-1"></a> <span class="ot">]</span></span>+<span id="cb1029-27"><a href="#cb1029-27" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div><p>Notes on above:</p><ol type="1"><li>This is a policy that can be used when creating bucket. It assumes@@ -27436,6 +28999,8 @@ policy</a>. The bucket can still be synced or copied into normally, butif rclone tries to access data from the glacier storage class you willsee an error like below.</p><pre class="text"><code>2017/09/11 19:07:43 Failed to sync: failed to open source object: Object in GLACIER, restore first: path/to/file</code></pre>+<p>The message shows the object's actual storage class, for example+<code>DEEP_ARCHIVE</code>.</p><p>In this case you need to <ahref="http://docs.aws.amazon.com/AmazonS3/latest/user-guide/restore-archived-objects.html">restore</a>the object(s) in question before accessing object contents. The <a@@ -27504,14 +29069,15 @@ Lock headers during PUT)</li><!-- markdownlint-disable-line line-length --><h3 id="standard-options-2">Standard options</h3><p>Here are the Standard options specific to s3 (Amazon S3 Compliant-Storage Providers including AWS, Alibaba, ArvanCloud, BizflyCloud, Ceph,-ChinaMobile, Cloudflare, Cubbit, DigitalOcean, Dreamhost, Exaba, Fastly,-FileLu, FlashBlade, GCS, HCP, Hetzner, HuaweiOBS, IBMCOS, IDrive,-ImpossibleCloud, Intercolo, IONOS, Leviia, Liara, Linode, LyveCloud,-Magalu, Mega, Minio, Netease, Outscale, OVHcloud, Petabox, Qiniu,-Rabata, RackCorp, Rclone, Scaleway, Scality, SeaweedFS, Selectel,-Servercore, SpectraLogic, Storj, Synology, TencentCOS, US3, Wasabi,-Zadara, Zata, ZeroServices, Other).</p>+Storage Providers including AWS, Alibaba, ArvanCloud, Autonomys,+BizflyCloud, Cellar, Ceph, ChinaMobile, Cloudflare, Cubbit,+DigitalOcean, Dreamhost, Exaba, Fastly, FileLu, FlashBlade, GCS, HCP,+Hetzner, HuaweiOBS, IBMCOS, IDrive, ImpossibleCloud, Infomaniak,+Intercolo, IONOS, Leviia, Liara, Linode, LyveCloud, Magalu, Mega, Minio,+Neon, Netease, Outscale, OVHcloud, Petabox, Qiniu, Rabata, RackCorp,+Rclone, RelAix, Scaleway, Scality, SeaweedFS, Selectel, Servercore,+SpectraLogic, Storj, Synology, TencentCOS, US3, Wasabi, Zadara, Zata,+ZeroServices, Other).</p><h4 id="s3-provider">--s3-provider</h4><p>Choose your S3 provider.</p><p>Properties:</p>@@ -27534,10 +29100,18 @@ Zadara, Zata, ZeroServices, Other).</p><ul><li>Arvan Cloud Object Storage (AOS)</li></ul></li>+<li>"Autonomys"+<ul>+<li>Autonomys Auto Drive</li>+</ul></li><li>"BizflyCloud"<ul><li>Bizfly Cloud Simple Storage</li></ul></li>+<li>"Cellar"+<ul>+<li>Clever Cloud Cellar Object Storage</li>+</ul></li><li>"Ceph"<ul><li>Ceph Object Storage</li>@@ -27606,6 +29180,10 @@ Zadara, Zata, ZeroServices, Other).</p><ul><li>Impossible Cloud Object Storage</li></ul></li>+<li>"Infomaniak"+<ul>+<li>Infomaniak Swiss Backup</li>+</ul></li><li>"Intercolo"<ul><li>Intercolo Object Storage</li>@@ -27642,6 +29220,10 @@ Zadara, Zata, ZeroServices, Other).</p><ul><li>Minio Object Storage</li></ul></li>+<li>"Neon"+<ul>+<li>Neon Object Storage</li>+</ul></li><li>"Netease"<ul><li>Netease Object Storage (NOS)</li>@@ -27674,6 +29256,10 @@ Zadara, Zata, ZeroServices, Other).</p><ul><li>Rclone S3 Server</li></ul></li>+<li>"RelAix"+<ul>+<li>RelAix.Storage</li>+</ul></li><li>"Scaleway"<ul><li>Scaleway Object Storage</li>@@ -27787,7 +29373,7 @@ region.</p><li>Config: region</li><li>Env Var: RCLONE_S3_REGION</li><li>Provider:-AWS,BizflyCloud,Ceph,Cloudflare,Cubbit,DigitalOcean,Dreamhost,Exaba,Fastly,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IBMCOS,ImpossibleCloud,Intercolo,IONOS,Leviia,LyveCloud,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,Rabata,RackCorp,Scaleway,Scality,SeaweedFS,Selectel,Servercore,Synology,Wasabi,Zadara,Zata,ZeroServices,Other</li>+AWS,BizflyCloud,Ceph,Cloudflare,Cubbit,DigitalOcean,Dreamhost,Exaba,Fastly,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IBMCOS,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,LyveCloud,Minio,Neon,Netease,Outscale,OVHcloud,Petabox,Qiniu,Rabata,RackCorp,Scaleway,Scality,SeaweedFS,Selectel,Servercore,Synology,Wasabi,Zadara,Zata,ZeroServices,Other</li><li>Type: string</li><li>Required: false</li><li>Examples:@@ -27994,6 +29580,11 @@ centers for low latency.</li><li>EU Central</li><li>Provider: Fastly</li></ul></li>+<li>"eu-central-1"+<ul>+<li>EU Central 1</li>+<li>Provider: Fastly</li>+</ul></li><li>"eu-south-1"<ul><li>EU South 1</li>@@ -28079,11 +29670,21 @@ centers for low latency.</li><li>Nuremberg</li><li>Provider: Hetzner</li></ul></li>+<li>"af-north-1"+<ul>+<li>AF-Cairo</li>+<li>Provider: HuaweiOBS</li>+</ul></li><li>"af-south-1"<ul><li>AF-Johannesburg</li><li>Provider: HuaweiOBS</li></ul></li>+<li>"ap-southeast-1"+<ul>+<li>CN-Hong Kong</li>+<li>Provider: HuaweiOBS</li>+</ul></li><li>"ap-southeast-2"<ul><li>AP-Bangkok</li>@@ -28094,9 +29695,14 @@ centers for low latency.</li><li>AP-Singapore</li><li>Provider: HuaweiOBS</li></ul></li>-<li>"cn-east-3"+<li>"ap-southeast-4"<ul>-<li>CN East-Shanghai1</li>+<li>AP-Jakarta</li>+<li>Provider: HuaweiOBS</li>+</ul></li>+<li>"ap-southeast-5"+<ul>+<li>AP-Manila</li><li>Provider: HuaweiOBS</li></ul></li><li>"cn-east-2"@@ -28104,34 +29710,94 @@ centers for low latency.</li><li>CN East-Shanghai2</li><li>Provider: HuaweiOBS</li></ul></li>+<li>"cn-east-3"+<ul>+<li>CN East-Shanghai1</li>+<li>Provider: HuaweiOBS</li>+</ul></li>+<li>"cn-east-4"+<ul>+<li>CN East-Qingdao</li>+<li>Provider: HuaweiOBS</li>+</ul></li>+<li>"cn-east-5"+<ul>+<li>CN East-Suzhou</li>+<li>Provider: HuaweiOBS</li>+</ul></li><li>"cn-north-1"<ul><li>CN North-Beijing1</li><li>Provider: HuaweiOBS</li></ul></li>+<li>"cn-north-2"+<ul>+<li>CN North-Beijing2</li>+<li>Provider: HuaweiOBS</li>+</ul></li><li>"cn-north-4"<ul><li>CN North-Beijing4</li><li>Provider: HuaweiOBS</li></ul></li>+<li>"cn-north-9"+<ul>+<li>CN North-Ulanqab</li>+<li>Provider: HuaweiOBS</li>+</ul></li>+<li>"cn-north-11"+<ul>+<li>CN North-Hebei</li>+<li>Provider: HuaweiOBS</li>+</ul></li>+<li>"cn-north-12"+<ul>+<li>CN North-Hohhot</li>+<li>Provider: HuaweiOBS</li>+</ul></li><li>"cn-south-1"<ul><li>CN South-Guangzhou</li><li>Provider: HuaweiOBS</li></ul></li>-<li>"ap-southeast-1"+<li>"cn-south-2"<ul>-<li>CN-Hong Kong</li>+<li>CN South-Shenzhen</li><li>Provider: HuaweiOBS</li></ul></li>-<li>"sa-argentina-1"+<li>"cn-south-4"<ul>-<li>LA-Buenos Aires1</li>+<li>CN South-Guangzhou-InvitationOnly</li><li>Provider: HuaweiOBS</li></ul></li>-<li>"sa-peru-1"+<li>"cn-southwest-2"<ul>-<li>LA-Lima1</li>+<li>CN Southwest-Guiyang1</li>+<li>Provider: HuaweiOBS</li>+</ul></li>+<li>"cn-southwest-3"+<ul>+<li>CN Southwest-Chongqing</li>+<li>Provider: HuaweiOBS</li>+</ul></li>+<li>"eu-west-101"+<ul>+<li>EU-Dublin</li>+<li>Provider: HuaweiOBS</li>+</ul></li>+<li>"la-north-2"+<ul>+<li>LA-Mexico City2</li>+<li>Provider: HuaweiOBS</li>+</ul></li>+<li>"la-south-2"+<ul>+<li>LA-Santiago</li>+<li>Provider: HuaweiOBS</li>+</ul></li>+<li>"me-east-1"+<ul>+<li>ME-Riyadh</li><li>Provider: HuaweiOBS</li></ul></li><li>"na-mexico-1"@@ -28139,9 +29805,14 @@ centers for low latency.</li><li>LA-Mexico City1</li><li>Provider: HuaweiOBS</li></ul></li>-<li>"sa-chile-1"+<li>"ru-northwest-2"<ul>-<li>LA-Santiago2</li>+<li>RU-Moscow2</li>+<li>Provider: HuaweiOBS</li>+</ul></li>+<li>"sa-argentina-1"+<ul>+<li>LA-Buenos Aires1</li><li>Provider: HuaweiOBS</li></ul></li><li>"sa-brazil-1"@@ -28149,9 +29820,19 @@ centers for low latency.</li><li>LA-Sao Paulo1</li><li>Provider: HuaweiOBS</li></ul></li>-<li>"ru-northwest-2"+<li>"sa-chile-1"<ul>-<li>RU-Moscow2</li>+<li>LA-Santiago2</li>+<li>Provider: HuaweiOBS</li>+</ul></li>+<li>"sa-peru-1"+<ul>+<li>LA-Lima1</li>+<li>Provider: HuaweiOBS</li>+</ul></li>+<li>"tr-west-1"+<ul>+<li>TR-Istanbul</li><li>Provider: HuaweiOBS</li></ul></li><li>"eu-central-2"@@ -28189,6 +29870,12 @@ centers for low latency.</li><li>New York, USA</li><li>Provider: ImpossibleCloud</li></ul></li>+<li>"us-east-1"+<ul>+<li>Default region for compatibility.</li>+<li>All data is stored in Switzerland.</li>+<li>Provider: Infomaniak</li>+</ul></li><li>"de-fra"<ul><li>Frankfurt, Germany</li>@@ -28224,6 +29911,26 @@ centers for low latency.</li><li>Lenexa, USA</li><li>Provider: IONOS</li></ul></li>+<li>"us-east-1"+<ul>+<li>US East (N. Virginia)</li>+<li>Provider: Neon,Petabox,Rabata</li>+</ul></li>+<li>"us-east-2"+<ul>+<li>US East (Ohio)</li>+<li>Provider: Neon</li>+</ul></li>+<li>"eu-central-1"+<ul>+<li>Europe (Frankfurt)</li>+<li>Provider: Neon,Petabox</li>+</ul></li>+<li>"ap-southeast-1"+<ul>+<li>Asia Pacific (Singapore)</li>+<li>Provider: Neon,Petabox</li>+</ul></li><li>"eu-west-2"<ul><li>Paris, France</li>@@ -28319,21 +30026,6 @@ centers for low latency.</li><li>Roubaix, France (Cold Archive)</li><li>Provider: OVHcloud</li></ul></li>-<li>"us-east-1"-<ul>-<li>US East (N. Virginia)</li>-<li>Provider: Petabox,Rabata</li>-</ul></li>-<li>"eu-central-1"-<ul>-<li>Europe (Frankfurt)</li>-<li>Provider: Petabox</li>-</ul></li>-<li>"ap-southeast-1"-<ul>-<li>Asia Pacific (Singapore)</li>-<li>Provider: Petabox</li>-</ul></li><li>"me-south-1"<ul><li>Middle East (Bahrain)</li>@@ -28507,6 +30199,11 @@ centers for low latency.</li><li>Warsaw, Poland</li><li>Provider: Scaleway</li></ul></li>+<li>"it-mil"+<ul>+<li>Milan, Italy</li>+<li>Provider: Scaleway</li>+</ul></li><li>"ru-1"<ul><li>St. Petersburg</li>@@ -28608,7 +30305,7 @@ centers for low latency.</li><li>Config: endpoint</li><li>Env Var: RCLONE_S3_ENDPOINT</li><li>Provider:-AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,ChinaMobile,Cloudflare,Cubbit,DigitalOcean,Dreamhost,Exaba,Fastly,FileLu,FlashBlade,GCS,HCP,Hetzner,HuaweiOBS,IBMCOS,ImpossibleCloud,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Mega,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,Rabata,RackCorp,Rclone,Scaleway,Scality,SeaweedFS,Selectel,Servercore,SpectraLogic,Storj,Synology,TencentCOS,US3,Wasabi,Zadara,Zata,ZeroServices,Other</li>+AWS,Alibaba,ArvanCloud,Autonomys,BizflyCloud,Cellar,Ceph,ChinaMobile,Cloudflare,Cubbit,DigitalOcean,Dreamhost,Exaba,Fastly,FileLu,FlashBlade,GCS,HCP,Hetzner,HuaweiOBS,IBMCOS,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Mega,Minio,Neon,Netease,Outscale,OVHcloud,Petabox,Qiniu,Rabata,RackCorp,Rclone,RelAix,Scaleway,Scality,SeaweedFS,Selectel,Servercore,SpectraLogic,Storj,Synology,TencentCOS,US3,Wasabi,Zadara,Zata,ZeroServices,Other</li><li>Type: string</li><li>Required: false</li><li>Examples:@@ -28749,6 +30446,11 @@ AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,ChinaMobile,Cloudflare,Cubbit,DigitalOce<li>Tabriz Iran (Shahriar)</li><li>Provider: ArvanCloud</li></ul></li>+<li>"s3.auto-drive.autonomys.xyz"+<ul>+<li>Auto Drive mainnet S3 endpoint</li>+<li>Provider: Autonomys</li>+</ul></li><li>"hn.ss.bfcplatform.vn"<ul><li>Hanoi endpoint</li>@@ -28759,6 +30461,11 @@ AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,ChinaMobile,Cloudflare,Cubbit,DigitalOce<li>Ho Chi Minh endpoint</li><li>Provider: BizflyCloud</li></ul></li>+<li>"cellar-c2.services.clever-cloud.com"+<ul>+<li>Paris, France</li>+<li>Provider: Cellar</li>+</ul></li><li>"eos-wuxi-1.cmecloud.cn"<ul><li>The default endpoint - a good choice if you are unsure.</li>@@ -28986,6 +30693,11 @@ tenant name. Do not select this directly</li><li>EU Central</li><li>Provider: Fastly</li></ul></li>+<li>"eu-central-1.object.fastlystorage.app"+<ul>+<li>EU Central 1</li>+<li>Provider: Fastly</li>+</ul></li><li>"eu-south-1.object.fastlystorage.app"<ul><li>EU South 1</li>@@ -29076,11 +30788,21 @@ tenant name. Do not select this directly</li><li>Nuremberg</li><li>Provider: Hetzner</li></ul></li>+<li>"obs.af-north-1.myhuaweicloud.com"+<ul>+<li>AF-Cairo</li>+<li>Provider: HuaweiOBS</li>+</ul></li><li>"obs.af-south-1.myhuaweicloud.com"<ul><li>AF-Johannesburg</li><li>Provider: HuaweiOBS</li></ul></li>+<li>"obs.ap-southeast-1.myhuaweicloud.com"+<ul>+<li>CN-Hong Kong</li>+<li>Provider: HuaweiOBS</li>+</ul></li><li>"obs.ap-southeast-2.myhuaweicloud.com"<ul><li>AP-Bangkok</li>@@ -29091,9 +30813,14 @@ tenant name. Do not select this directly</li><li>AP-Singapore</li><li>Provider: HuaweiOBS</li></ul></li>-<li>"obs.cn-east-3.myhuaweicloud.com"+<li>"obs.ap-southeast-4.myhuaweicloud.com"<ul>-<li>CN East-Shanghai1</li>+<li>AP-Jakarta</li>+<li>Provider: HuaweiOBS</li>+</ul></li>+<li>"obs.ap-southeast-5.myhuaweicloud.com"+<ul>+<li>AP-Manila</li><li>Provider: HuaweiOBS</li></ul></li><li>"obs.cn-east-2.myhuaweicloud.com"@@ -29101,34 +30828,94 @@ tenant name. Do not select this directly</li><li>CN East-Shanghai2</li><li>Provider: HuaweiOBS</li></ul></li>+<li>"obs.cn-east-3.myhuaweicloud.com"+<ul>+<li>CN East-Shanghai1</li>+<li>Provider: HuaweiOBS</li>+</ul></li>+<li>"obs.cn-east-4.myhuaweicloud.com"+<ul>+<li>CN East-Qingdao</li>+<li>Provider: HuaweiOBS</li>+</ul></li>+<li>"obs.cn-east-5.myhuaweicloud.com"+<ul>+<li>CN East-Suzhou</li>+<li>Provider: HuaweiOBS</li>+</ul></li><li>"obs.cn-north-1.myhuaweicloud.com"<ul><li>CN North-Beijing1</li><li>Provider: HuaweiOBS</li></ul></li>+<li>"obs.cn-north-2.myhuaweicloud.com"+<ul>+<li>CN North-Beijing2</li>+<li>Provider: HuaweiOBS</li>+</ul></li><li>"obs.cn-north-4.myhuaweicloud.com"<ul><li>CN North-Beijing4</li><li>Provider: HuaweiOBS</li></ul></li>+<li>"obs.cn-north-9.myhuaweicloud.com"+<ul>+<li>CN North-Ulanqab</li>+<li>Provider: HuaweiOBS</li>+</ul></li>+<li>"obs.cn-north-11.myhuaweicloud.com"+<ul>+<li>CN North-Hebei</li>+<li>Provider: HuaweiOBS</li>+</ul></li>+<li>"obs.cn-north-12.myhuaweicloud.com"+<ul>+<li>CN North-Hohhot</li>+<li>Provider: HuaweiOBS</li>+</ul></li><li>"obs.cn-south-1.myhuaweicloud.com"<ul><li>CN South-Guangzhou</li><li>Provider: HuaweiOBS</li></ul></li>-<li>"obs.ap-southeast-1.myhuaweicloud.com"+<li>"obs.cn-south-2.myhuaweicloud.com"<ul>-<li>CN-Hong Kong</li>+<li>CN South-Shenzhen</li><li>Provider: HuaweiOBS</li></ul></li>-<li>"obs.sa-argentina-1.myhuaweicloud.com"+<li>"obs.cn-south-4.myhuaweicloud.com"<ul>-<li>LA-Buenos Aires1</li>+<li>CN South-Guangzhou-InvitationOnly</li><li>Provider: HuaweiOBS</li></ul></li>-<li>"obs.sa-peru-1.myhuaweicloud.com"+<li>"obs.cn-southwest-2.myhuaweicloud.com"<ul>-<li>LA-Lima1</li>+<li>CN Southwest-Guiyang1</li>+<li>Provider: HuaweiOBS</li>+</ul></li>+<li>"obs.cn-southwest-3.myhuaweicloud.com"+<ul>+<li>CN Southwest-Chongqing</li>+<li>Provider: HuaweiOBS</li>+</ul></li>+<li>"obs.eu-west-101.myhuaweicloud.eu"+<ul>+<li>EU-Dublin</li>+<li>Provider: HuaweiOBS</li>+</ul></li>+<li>"obs.la-north-2.myhuaweicloud.com"+<ul>+<li>LA-Mexico City2</li>+<li>Provider: HuaweiOBS</li>+</ul></li>+<li>"obs.la-south-2.myhuaweicloud.com"+<ul>+<li>LA-Santiago</li>+<li>Provider: HuaweiOBS</li>+</ul></li>+<li>"obs.me-east-1.myhuaweicloud.com"+<ul>+<li>ME-Riyadh</li><li>Provider: HuaweiOBS</li></ul></li><li>"obs.na-mexico-1.myhuaweicloud.com"@@ -29136,9 +30923,14 @@ tenant name. Do not select this directly</li><li>LA-Mexico City1</li><li>Provider: HuaweiOBS</li></ul></li>-<li>"obs.sa-chile-1.myhuaweicloud.com"+<li>"obs.ru-northwest-2.myhuaweicloud.com"<ul>-<li>LA-Santiago2</li>+<li>RU-Moscow2</li>+<li>Provider: HuaweiOBS</li>+</ul></li>+<li>"obs.sa-argentina-1.myhuaweicloud.com"+<ul>+<li>LA-Buenos Aires1</li><li>Provider: HuaweiOBS</li></ul></li><li>"obs.sa-brazil-1.myhuaweicloud.com"@@ -29146,9 +30938,19 @@ tenant name. Do not select this directly</li><li>LA-Sao Paulo1</li><li>Provider: HuaweiOBS</li></ul></li>-<li>"obs.ru-northwest-2.myhuaweicloud.com"+<li>"obs.sa-chile-1.myhuaweicloud.com"<ul>-<li>RU-Moscow2</li>+<li>LA-Santiago2</li>+<li>Provider: HuaweiOBS</li>+</ul></li>+<li>"obs.sa-peru-1.myhuaweicloud.com"+<ul>+<li>LA-Lima1</li>+<li>Provider: HuaweiOBS</li>+</ul></li>+<li>"obs.tr-west-1.myhuaweicloud.com"+<ul>+<li>TR-Istanbul</li><li>Provider: HuaweiOBS</li></ul></li><li>"s3.us.cloud-object-storage.appdomain.cloud"@@ -29496,6 +31298,26 @@ tenant name. Do not select this directly</li><li>New York, USA</li><li>Provider: ImpossibleCloud</li></ul></li>+<li>"s3.swiss-backup01.infomaniak.com"+<ul>+<li>Swiss Backup Location 01</li>+<li>Provider: Infomaniak</li>+</ul></li>+<li>"s3.swiss-backup02.infomaniak.com"+<ul>+<li>Swiss Backup Location 02</li>+<li>Provider: Infomaniak</li>+</ul></li>+<li>"s3.swiss-backup03.infomaniak.com"+<ul>+<li>Swiss Backup Location 03</li>+<li>Provider: Infomaniak</li>+</ul></li>+<li>"s3.swiss-backup04.infomaniak.com"+<ul>+<li>Swiss Backup Location 04</li>+<li>Provider: Infomaniak</li>+</ul></li><li>"de-fra.i3storage.com"<ul><li>Frankfurt, Germany</li>@@ -29668,59 +31490,74 @@ tenant name. Do not select this directly</li><li>Fortaleza, CE (BR), br-ne1</li><li>Provider: Magalu</li></ul></li>-<li>"s3.eu-amsterdam.megas4.com"+<li>"s3.eu-luxembourg-1.megas4.com"+<ul>+<li>Mega S4 Luxembourg 1</li>+<li>Provider: Mega</li>+</ul></li>+<li>"s3.eu-luxembourg-2.megas4.com"+<ul>+<li>Mega S4 Luxembourg 2</li>+<li>Provider: Mega</li>+</ul></li>+<li>"s3.eu-amsterdam-1.megas4.com"+<ul>+<li>Mega S4 Amsterdam 1</li>+<li>Provider: Mega</li>+</ul></li>+<li>"s3.eu-amsterdam-2.megas4.com"<ul>-<li>Mega S4 Amsterdam</li>+<li>Mega S4 Amsterdam 2</li><li>Provider: Mega</li></ul></li>-<li>"s3.eu-luxembourg.megas4.com"+<li>"s3.eu-paris-1.megas4.com"<ul>-<li>Mega S4 Luxembourg</li>+<li>Mega S4 Paris 1</li><li>Provider: Mega</li></ul></li>-<li>"s3.eu-paris.megas4.com"+<li>"s3.eu-paris-2.megas4.com"<ul>-<li>Mega S4 Paris</li>+<li>Mega S4 Paris 2</li><li>Provider: Mega</li></ul></li>-<li>"s3.eu-barcelona.megas4.com"+<li>"s3.eu-barcelona-1.megas4.com"<ul>-<li>Mega S4 Barcelona</li>+<li>Mega S4 Barcelona 1</li><li>Provider: Mega</li></ul></li>-<li>"s3.ca-montreal.megas4.com"+<li>"s3.eu-barcelona-2.megas4.com"<ul>-<li>Mega S4 Montreal</li>+<li>Mega S4 Barcelona 2</li><li>Provider: Mega</li></ul></li>-<li>"s3.ca-vancouver.megas4.com"+<li>"s3.ca-montreal-1.megas4.com"<ul>-<li>Mega S4 Vancouver</li>+<li>Mega S4 Montreal 1</li><li>Provider: Mega</li></ul></li>-<li>"s3.ap-tokyo.megas4.com"+<li>"s3.ca-montreal-2.megas4.com"<ul>-<li>Mega S4 Tokyo</li>+<li>Mega S4 Montreal 2</li><li>Provider: Mega</li></ul></li>-<li>"s3.eu-central-1.s4.mega.io"+<li>"s3.ca-vancouver-1.megas4.com"<ul>-<li>Mega S4 eu-central-1 (Amsterdam, legacy)</li>+<li>Mega S4 Vancouver 1</li><li>Provider: Mega</li></ul></li>-<li>"s3.eu-central-2.s4.mega.io"+<li>"s3.ca-vancouver-2.megas4.com"<ul>-<li>Mega S4 eu-central-2 (Bettembourg, legacy)</li>+<li>Mega S4 Vancouver 2</li><li>Provider: Mega</li></ul></li>-<li>"s3.ca-central-1.s4.mega.io"+<li>"s3.ap-tokyo-1.megas4.com"<ul>-<li>Mega S4 ca-central-1 (Montreal, legacy)</li>+<li>Mega S4 Tokyo 1</li><li>Provider: Mega</li></ul></li>-<li>"s3.ca-west-1.s4.mega.io"+<li>"s3.ap-tokyo-2.megas4.com"<ul>-<li>Mega S4 ca-west-1 (Vancouver, legacy)</li>+<li>Mega S4 Tokyo 2</li><li>Provider: Mega</li></ul></li><li>"oos.eu-west-2.outscale.com"@@ -29998,6 +31835,16 @@ tenant name. Do not select this directly</li><li>Auckland (New Zealand) Endpoint</li><li>Provider: RackCorp</li></ul></li>+<li>"s3-gw1.relaix.net"+<ul>+<li>Aachen-1, Germany</li>+<li>Provider: RelAix</li>+</ul></li>+<li>"s3-gw2.relaix.net"+<ul>+<li>Aachen-2, Germany</li>+<li>Provider: RelAix</li>+</ul></li><li>"s3.nl-ams.scw.cloud"<ul><li>Amsterdam Endpoint</li>@@ -30013,6 +31860,11 @@ tenant name. Do not select this directly</li><li>Warsaw Endpoint</li><li>Provider: Scaleway</li></ul></li>+<li>"s3.it-mil.scw.cloud"+<ul>+<li>Milan Endpoint</li>+<li>Provider: Scaleway</li>+</ul></li><li>"localhost:8333"<ul><li>SeaweedFS S3 localhost</li>@@ -30398,7 +32250,7 @@ tenant name. Do not select this directly</li><li>Config: location_constraint</li><li>Env Var: RCLONE_S3_LOCATION_CONSTRAINT</li><li>Provider:-AWS,ArvanCloud,Ceph,ChinaMobile,DigitalOcean,Dreamhost,Exaba,GCS,HCP,Hetzner,IBMCOS,ImpossibleCloud,LyveCloud,Minio,Netease,Qiniu,Rabata,RackCorp,Scality,SeaweedFS,Synology,Wasabi,Zata,ZeroServices,Other</li>+AWS,ArvanCloud,Ceph,ChinaMobile,DigitalOcean,Dreamhost,Exaba,GCS,HCP,Hetzner,IBMCOS,ImpossibleCloud,Infomaniak,LyveCloud,Minio,Netease,Qiniu,Rabata,RackCorp,Scality,SeaweedFS,Synology,Wasabi,Zata,ZeroServices,Other</li><li>Type: string</li><li>Required: false</li><li>Examples:@@ -31016,7 +32868,7 @@ the default (private) will be used.</p><li>Config: acl</li><li>Env Var: RCLONE_S3_ACL</li><li>Provider:-AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IBMCOS,IDrive,ImpossibleCloud,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,Scaleway,Scality,SeaweedFS,TencentCOS,US3,Wasabi,Zata,ZeroServices,Other</li>+AWS,Alibaba,ArvanCloud,BizflyCloud,Cellar,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IBMCOS,IDrive,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,RelAix,Scaleway,Scality,SeaweedFS,TencentCOS,US3,Wasabi,Zata,ZeroServices,Other</li><li>Type: string</li><li>Required: false</li><li>Examples:@@ -31026,14 +32878,14 @@ AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhos<li>Owner gets FULL_CONTROL.</li><li>No one else has access rights (default).</li><li>Provider:-AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,Scaleway,Scality,SeaweedFS,US3,Wasabi,Zata,ZeroServices,Other</li>+AWS,Alibaba,ArvanCloud,BizflyCloud,Cellar,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,RelAix,Scaleway,Scality,SeaweedFS,US3,Wasabi,Zata,ZeroServices,Other</li></ul></li><li>"public-read"<ul><li>Owner gets FULL_CONTROL.</li><li>The AllUsers group gets READ access.</li><li>Provider:-AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,Scaleway,Scality,SeaweedFS,TencentCOS,US3,Wasabi,Zata,ZeroServices,Other</li>+AWS,Alibaba,ArvanCloud,BizflyCloud,Cellar,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,RelAix,Scaleway,Scality,SeaweedFS,TencentCOS,US3,Wasabi,Zata,ZeroServices,Other</li></ul></li><li>"public-read-write"<ul>@@ -31041,14 +32893,14 @@ AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhos<li>The AllUsers group gets READ and WRITE access.</li><li>Granting this on a bucket is generally not recommended.</li><li>Provider:-AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,Scaleway,Scality,SeaweedFS,TencentCOS,Wasabi,Zata,ZeroServices,Other</li>+AWS,Alibaba,ArvanCloud,BizflyCloud,Cellar,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,RelAix,Scaleway,Scality,SeaweedFS,TencentCOS,Wasabi,Zata,ZeroServices,Other</li></ul></li><li>"authenticated-read"<ul><li>Owner gets FULL_CONTROL.</li><li>The AuthenticatedUsers group gets READ access.</li><li>Provider:-AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,Scaleway,Scality,SeaweedFS,TencentCOS,Wasabi,Zata,ZeroServices,Other</li>+AWS,Alibaba,ArvanCloud,BizflyCloud,Cellar,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,RelAix,Scaleway,Scality,SeaweedFS,TencentCOS,Wasabi,Zata,ZeroServices,Other</li></ul></li><li>"bucket-owner-read"<ul>@@ -31057,7 +32909,7 @@ AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhos<li>If you specify this canned ACL when creating a bucket, Amazon S3ignores it.</li><li>Provider:-AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,Scaleway,Scality,SeaweedFS,TencentCOS,Wasabi,Zata,ZeroServices,Other</li>+AWS,Alibaba,ArvanCloud,BizflyCloud,Cellar,Ceph,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,RelAix,Scaleway,Scality,SeaweedFS,TencentCOS,Wasabi,Zata,ZeroServices,Other</li></ul></li><li>"bucket-owner-full-control"<ul>@@ -31066,7 +32918,7 @@ object.</li><li>If you specify this canned ACL when creating a bucket, Amazon S3ignores it.</li><li>Provider:-AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,Scaleway,Scality,SeaweedFS,TencentCOS,Wasabi,Zata,ZeroServices,Other</li>+AWS,Alibaba,ArvanCloud,BizflyCloud,Cellar,Ceph,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,RelAix,Scaleway,Scality,SeaweedFS,TencentCOS,Wasabi,Zata,ZeroServices,Other</li></ul></li><li>"private"<ul>@@ -31333,14 +33185,15 @@ id="s3-bucket-object-lock-enabled">--s3-bucket-object-lock-enabled</h4></ul><h3 id="advanced-options-2">Advanced options</h3><p>Here are the Advanced options specific to s3 (Amazon S3 Compliant-Storage Providers including AWS, Alibaba, ArvanCloud, BizflyCloud, Ceph,-ChinaMobile, Cloudflare, Cubbit, DigitalOcean, Dreamhost, Exaba, Fastly,-FileLu, FlashBlade, GCS, HCP, Hetzner, HuaweiOBS, IBMCOS, IDrive,-ImpossibleCloud, Intercolo, IONOS, Leviia, Liara, Linode, LyveCloud,-Magalu, Mega, Minio, Netease, Outscale, OVHcloud, Petabox, Qiniu,-Rabata, RackCorp, Rclone, Scaleway, Scality, SeaweedFS, Selectel,-Servercore, SpectraLogic, Storj, Synology, TencentCOS, US3, Wasabi,-Zadara, Zata, ZeroServices, Other).</p>+Storage Providers including AWS, Alibaba, ArvanCloud, Autonomys,+BizflyCloud, Cellar, Ceph, ChinaMobile, Cloudflare, Cubbit,+DigitalOcean, Dreamhost, Exaba, Fastly, FileLu, FlashBlade, GCS, HCP,+Hetzner, HuaweiOBS, IBMCOS, IDrive, ImpossibleCloud, Infomaniak,+Intercolo, IONOS, Leviia, Liara, Linode, LyveCloud, Magalu, Mega, Minio,+Neon, Netease, Outscale, OVHcloud, Petabox, Qiniu, Rabata, RackCorp,+Rclone, RelAix, Scaleway, Scality, SeaweedFS, Selectel, Servercore,+SpectraLogic, Storj, Synology, TencentCOS, US3, Wasabi, Zadara, Zata,+ZeroServices, Other).</p><h4 id="s3-bucket-acl">--s3-bucket-acl</h4><p>Canned ACL used when creating buckets.</p><p>For more info visit@@ -31354,7 +33207,7 @@ header is added and the default (private) will be used.</p><li>Config: bucket_acl</li><li>Env Var: RCLONE_S3_BUCKET_ACL</li><li>Provider:-AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IBMCOS,IDrive,ImpossibleCloud,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Mega,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,Scaleway,SeaweedFS,Servercore,TencentCOS,US3,Wasabi,Zata,ZeroServices,Other</li>+AWS,Alibaba,ArvanCloud,BizflyCloud,Cellar,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IBMCOS,IDrive,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Mega,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,RelAix,Scaleway,SeaweedFS,Servercore,TencentCOS,US3,Wasabi,Zata,ZeroServices,Other</li><li>Type: string</li><li>Required: false</li><li>Examples:@@ -31399,7 +33252,7 @@ storing this object in S3.</p><ul><li>Config: sse_customer_algorithm</li><li>Env Var: RCLONE_S3_SSE_CUSTOMER_ALGORITHM</li>-<li>Provider: AWS,Ceph,ChinaMobile,Minio</li>+<li>Provider: AWS,Cellar,Ceph,ChinaMobile,Minio</li><li>Type: string</li><li>Required: false</li><li>Examples:@@ -31417,12 +33270,12 @@ storing this object in S3.</p><h4 id="s3-sse-customer-key">--s3-sse-customer-key</h4><p>To use SSE-C you may provide the secret encryption key used toencrypt/decrypt your data.</p>-<p>Alternatively you can provide --sse-customer-key-base64.</p>+<p>Alternatively you can provide --s3-sse-customer-key-base64.</p><p>Properties:</p><ul><li>Config: sse_customer_key</li><li>Env Var: RCLONE_S3_SSE_CUSTOMER_KEY</li>-<li>Provider: AWS,Ceph,ChinaMobile,Minio</li>+<li>Provider: AWS,Cellar,Ceph,ChinaMobile,Minio</li><li>Type: string</li><li>Required: false</li><li>Examples:@@ -31436,12 +33289,12 @@ encrypt/decrypt your data.</p><h4 id="s3-sse-customer-key-base64">--s3-sse-customer-key-base64</h4><p>If using SSE-C you must provide the secret encryption key encoded inbase64 format to encrypt/decrypt your data.</p>-<p>Alternatively you can provide --sse-customer-key.</p>+<p>Alternatively you can provide --s3-sse-customer-key.</p><p>Properties:</p><ul><li>Config: sse_customer_key_base64</li><li>Env Var: RCLONE_S3_SSE_CUSTOMER_KEY_BASE64</li>-<li>Provider: AWS,Ceph,ChinaMobile,Minio</li>+<li>Provider: AWS,Cellar,Ceph,ChinaMobile,Minio</li><li>Type: string</li><li>Required: false</li><li>Examples:@@ -31461,7 +33314,7 @@ sse_customer_key provided.</p><ul><li>Config: sse_customer_key_md5</li><li>Env Var: RCLONE_S3_SSE_CUSTOMER_KEY_MD5</li>-<li>Provider: AWS,Ceph,ChinaMobile,Minio</li>+<li>Provider: AWS,Cellar,Ceph,ChinaMobile,Minio</li><li>Type: string</li><li>Required: false</li><li>Examples:@@ -32465,6 +34318,42 @@ href="https://rclone.org/commands/rclone_backend/">backend</a> commandfor more info on how to pass options and arguments.</p><p>These can be run on a running backend using the rc command <ahref="https://rclone.org/rc/#backend-command">backend/command</a>.</p>+<h3 id="link">link</h3>+<p>Generate a signed link with response header overrides.</p>+<pre class="console"><code>rclone backend link remote: [options] [<arguments>+]</code></pre>+<p>This command generates a signed download link for one file, with+optional overrides for the HTTP response headers. Pass the file path as+a separate argument, relative to the remote path.</p>+<p>Usage examples:</p>+<pre class="console"><code>rclone backend link s3:bucket path/to/file -o expire=1h+rclone backend link s3:bucket path/to/file -o response-expires="Thu, 01 Jan 1970 00:00:00 GMT"+rclone backend link s3:bucket path/to/file -o response-content-disposition='attachment; filename="download.txt"'</code></pre>+<p>The link expires after 7 days by default. Use+<code>-o expire=1h</code> to change its lifetime. Durations must be at+least 1 second; values above 7 days are limited to 7 days, as with+<code>rclone link</code>.</p>+<p>The <code>response-expires</code> option sets the HTTP Expires+response header, not the lifetime of the signed link. It must be an HTTP+date, such as <code>Thu, 01 Jan 1970 00:00:00 GMT</code>.</p>+<p>The overrides are included in the signature and must not be changed+in the returned URL. They do not modify the object's stored+metadata.</p>+<p>Options:</p>+<ul>+<li>"expire": How long the link will be valid (default 7d, maximum+7d).</li>+<li>"response-cache-control": Set the Cache-Control response+header.</li>+<li>"response-content-disposition": Set the Content-Disposition response+header.</li>+<li>"response-content-encoding": Set the Content-Encoding response+header.</li>+<li>"response-content-language": Set the Content-Language response+header.</li>+<li>"response-content-type": Set the Content-Type response header.</li>+<li>"response-expires": Set the Expires response header to an HTTP+date.</li>+</ul><h3 id="restore">restore</h3><p>Restore objects from GLACIER or INTELLIGENT-TIERING archive tier.</p><pre class="console"><code>rclone backend restore remote: [options] [<arguments>+]</code></pre>@@ -32484,17 +34373,17 @@ rclone backend restore s3:bucket/path/to/directory -o priority=PRIORITY</code></<p>It returns a list of status dictionaries with Remote and Status keys.The Status will be OK if it was successful or an error message ifnot.</p>-<div class="sourceCode" id="cb1014"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb1014-1"><a href="#cb1014-1" aria-hidden="true" tabindex="-1"></a><span class="ot">[</span></span>-<span id="cb1014-2"><a href="#cb1014-2" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>-<span id="cb1014-3"><a href="#cb1014-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Status"</span><span class="fu">:</span> <span class="st">"OK"</span><span class="fu">,</span></span>-<span id="cb1014-4"><a href="#cb1014-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Remote"</span><span class="fu">:</span> <span class="st">"test.txt"</span></span>-<span id="cb1014-5"><a href="#cb1014-5" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span><span class="ot">,</span></span>-<span id="cb1014-6"><a href="#cb1014-6" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>-<span id="cb1014-7"><a href="#cb1014-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Status"</span><span class="fu">:</span> <span class="st">"OK"</span><span class="fu">,</span></span>-<span id="cb1014-8"><a href="#cb1014-8" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Remote"</span><span class="fu">:</span> <span class="st">"test/file4.txt"</span></span>-<span id="cb1014-9"><a href="#cb1014-9" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>-<span id="cb1014-10"><a href="#cb1014-10" aria-hidden="true" tabindex="-1"></a><span class="ot">]</span></span></code></pre></div>+<div class="sourceCode" id="cb1044"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb1044-1"><a href="#cb1044-1" aria-hidden="true" tabindex="-1"></a><span class="ot">[</span></span>+<span id="cb1044-2"><a href="#cb1044-2" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>+<span id="cb1044-3"><a href="#cb1044-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Status"</span><span class="fu">:</span> <span class="st">"OK"</span><span class="fu">,</span></span>+<span id="cb1044-4"><a href="#cb1044-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Remote"</span><span class="fu">:</span> <span class="st">"test.txt"</span></span>+<span id="cb1044-5"><a href="#cb1044-5" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span><span class="ot">,</span></span>+<span id="cb1044-6"><a href="#cb1044-6" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>+<span id="cb1044-7"><a href="#cb1044-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Status"</span><span class="fu">:</span> <span class="st">"OK"</span><span class="fu">,</span></span>+<span id="cb1044-8"><a href="#cb1044-8" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Remote"</span><span class="fu">:</span> <span class="st">"test/file4.txt"</span></span>+<span id="cb1044-9"><a href="#cb1044-9" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>+<span id="cb1044-10"><a href="#cb1044-10" aria-hidden="true" tabindex="-1"></a><span class="ot">]</span></span></code></pre></div><p>Options:</p><ul><li>"description": The optional description for the job.</li>@@ -32516,36 +34405,36 @@ rclone backend restore-status s3:bucket/path/to/directoryrclone backend restore-status -o all s3:bucket/path/to/directory</code></pre><p>This command does not obey the filters.</p><p>It returns a list of status dictionaries:</p>-<div class="sourceCode" id="cb1017"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb1017-1"><a href="#cb1017-1" aria-hidden="true" tabindex="-1"></a><span class="ot">[</span></span>-<span id="cb1017-2"><a href="#cb1017-2" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>-<span id="cb1017-3"><a href="#cb1017-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Remote"</span><span class="fu">:</span> <span class="st">"file.txt"</span><span class="fu">,</span></span>-<span id="cb1017-4"><a href="#cb1017-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"VersionID"</span><span class="fu">:</span> <span class="kw">null</span><span class="fu">,</span></span>-<span id="cb1017-5"><a href="#cb1017-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"RestoreStatus"</span><span class="fu">:</span> <span class="fu">{</span></span>-<span id="cb1017-6"><a href="#cb1017-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">"IsRestoreInProgress"</span><span class="fu">:</span> <span class="kw">true</span><span class="fu">,</span></span>-<span id="cb1017-7"><a href="#cb1017-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"RestoreExpiryDate"</span><span class="fu">:</span> <span class="st">"2023-09-06T12:29:19+01:00"</span></span>-<span id="cb1017-8"><a href="#cb1017-8" aria-hidden="true" tabindex="-1"></a> <span class="fu">},</span></span>-<span id="cb1017-9"><a href="#cb1017-9" aria-hidden="true" tabindex="-1"></a> <span class="dt">"StorageClass"</span><span class="fu">:</span> <span class="st">"GLACIER"</span></span>-<span id="cb1017-10"><a href="#cb1017-10" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span><span class="ot">,</span></span>-<span id="cb1017-11"><a href="#cb1017-11" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>-<span id="cb1017-12"><a href="#cb1017-12" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Remote"</span><span class="fu">:</span> <span class="st">"test.pdf"</span><span class="fu">,</span></span>-<span id="cb1017-13"><a href="#cb1017-13" aria-hidden="true" tabindex="-1"></a> <span class="dt">"VersionID"</span><span class="fu">:</span> <span class="kw">null</span><span class="fu">,</span></span>-<span id="cb1017-14"><a href="#cb1017-14" aria-hidden="true" tabindex="-1"></a> <span class="dt">"RestoreStatus"</span><span class="fu">:</span> <span class="fu">{</span></span>-<span id="cb1017-15"><a href="#cb1017-15" aria-hidden="true" tabindex="-1"></a> <span class="dt">"IsRestoreInProgress"</span><span class="fu">:</span> <span class="kw">false</span><span class="fu">,</span></span>-<span id="cb1017-16"><a href="#cb1017-16" aria-hidden="true" tabindex="-1"></a> <span class="dt">"RestoreExpiryDate"</span><span class="fu">:</span> <span class="st">"2023-09-06T12:29:19+01:00"</span></span>-<span id="cb1017-17"><a href="#cb1017-17" aria-hidden="true" tabindex="-1"></a> <span class="fu">},</span></span>-<span id="cb1017-18"><a href="#cb1017-18" aria-hidden="true" tabindex="-1"></a> <span class="dt">"StorageClass"</span><span class="fu">:</span> <span class="st">"DEEP_ARCHIVE"</span></span>-<span id="cb1017-19"><a href="#cb1017-19" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span><span class="ot">,</span></span>-<span id="cb1017-20"><a href="#cb1017-20" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>-<span id="cb1017-21"><a href="#cb1017-21" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Remote"</span><span class="fu">:</span> <span class="st">"test.gz"</span><span class="fu">,</span></span>-<span id="cb1017-22"><a href="#cb1017-22" aria-hidden="true" tabindex="-1"></a> <span class="dt">"VersionID"</span><span class="fu">:</span> <span class="kw">null</span><span class="fu">,</span></span>-<span id="cb1017-23"><a href="#cb1017-23" aria-hidden="true" tabindex="-1"></a> <span class="dt">"RestoreStatus"</span><span class="fu">:</span> <span class="fu">{</span></span>-<span id="cb1017-24"><a href="#cb1017-24" aria-hidden="true" tabindex="-1"></a> <span class="dt">"IsRestoreInProgress"</span><span class="fu">:</span> <span class="kw">true</span><span class="fu">,</span></span>-<span id="cb1017-25"><a href="#cb1017-25" aria-hidden="true" tabindex="-1"></a> <span class="dt">"RestoreExpiryDate"</span><span class="fu">:</span> <span class="st">"null"</span></span>-<span id="cb1017-26"><a href="#cb1017-26" aria-hidden="true" tabindex="-1"></a> <span class="fu">},</span></span>-<span id="cb1017-27"><a href="#cb1017-27" aria-hidden="true" tabindex="-1"></a> <span class="dt">"StorageClass"</span><span class="fu">:</span> <span class="st">"INTELLIGENT_TIERING"</span></span>-<span id="cb1017-28"><a href="#cb1017-28" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>-<span id="cb1017-29"><a href="#cb1017-29" aria-hidden="true" tabindex="-1"></a><span class="ot">]</span></span></code></pre></div>+<div class="sourceCode" id="cb1047"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb1047-1"><a href="#cb1047-1" aria-hidden="true" tabindex="-1"></a><span class="ot">[</span></span>+<span id="cb1047-2"><a href="#cb1047-2" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>+<span id="cb1047-3"><a href="#cb1047-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Remote"</span><span class="fu">:</span> <span class="st">"file.txt"</span><span class="fu">,</span></span>+<span id="cb1047-4"><a href="#cb1047-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"VersionID"</span><span class="fu">:</span> <span class="kw">null</span><span class="fu">,</span></span>+<span id="cb1047-5"><a href="#cb1047-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"RestoreStatus"</span><span class="fu">:</span> <span class="fu">{</span></span>+<span id="cb1047-6"><a href="#cb1047-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">"IsRestoreInProgress"</span><span class="fu">:</span> <span class="kw">true</span><span class="fu">,</span></span>+<span id="cb1047-7"><a href="#cb1047-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"RestoreExpiryDate"</span><span class="fu">:</span> <span class="st">"2023-09-06T12:29:19+01:00"</span></span>+<span id="cb1047-8"><a href="#cb1047-8" aria-hidden="true" tabindex="-1"></a> <span class="fu">},</span></span>+<span id="cb1047-9"><a href="#cb1047-9" aria-hidden="true" tabindex="-1"></a> <span class="dt">"StorageClass"</span><span class="fu">:</span> <span class="st">"GLACIER"</span></span>+<span id="cb1047-10"><a href="#cb1047-10" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span><span class="ot">,</span></span>+<span id="cb1047-11"><a href="#cb1047-11" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>+<span id="cb1047-12"><a href="#cb1047-12" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Remote"</span><span class="fu">:</span> <span class="st">"test.pdf"</span><span class="fu">,</span></span>+<span id="cb1047-13"><a href="#cb1047-13" aria-hidden="true" tabindex="-1"></a> <span class="dt">"VersionID"</span><span class="fu">:</span> <span class="kw">null</span><span class="fu">,</span></span>+<span id="cb1047-14"><a href="#cb1047-14" aria-hidden="true" tabindex="-1"></a> <span class="dt">"RestoreStatus"</span><span class="fu">:</span> <span class="fu">{</span></span>+<span id="cb1047-15"><a href="#cb1047-15" aria-hidden="true" tabindex="-1"></a> <span class="dt">"IsRestoreInProgress"</span><span class="fu">:</span> <span class="kw">false</span><span class="fu">,</span></span>+<span id="cb1047-16"><a href="#cb1047-16" aria-hidden="true" tabindex="-1"></a> <span class="dt">"RestoreExpiryDate"</span><span class="fu">:</span> <span class="st">"2023-09-06T12:29:19+01:00"</span></span>+<span id="cb1047-17"><a href="#cb1047-17" aria-hidden="true" tabindex="-1"></a> <span class="fu">},</span></span>+<span id="cb1047-18"><a href="#cb1047-18" aria-hidden="true" tabindex="-1"></a> <span class="dt">"StorageClass"</span><span class="fu">:</span> <span class="st">"DEEP_ARCHIVE"</span></span>+<span id="cb1047-19"><a href="#cb1047-19" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span><span class="ot">,</span></span>+<span id="cb1047-20"><a href="#cb1047-20" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>+<span id="cb1047-21"><a href="#cb1047-21" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Remote"</span><span class="fu">:</span> <span class="st">"test.gz"</span><span class="fu">,</span></span>+<span id="cb1047-22"><a href="#cb1047-22" aria-hidden="true" tabindex="-1"></a> <span class="dt">"VersionID"</span><span class="fu">:</span> <span class="kw">null</span><span class="fu">,</span></span>+<span id="cb1047-23"><a href="#cb1047-23" aria-hidden="true" tabindex="-1"></a> <span class="dt">"RestoreStatus"</span><span class="fu">:</span> <span class="fu">{</span></span>+<span id="cb1047-24"><a href="#cb1047-24" aria-hidden="true" tabindex="-1"></a> <span class="dt">"IsRestoreInProgress"</span><span class="fu">:</span> <span class="kw">true</span><span class="fu">,</span></span>+<span id="cb1047-25"><a href="#cb1047-25" aria-hidden="true" tabindex="-1"></a> <span class="dt">"RestoreExpiryDate"</span><span class="fu">:</span> <span class="st">"null"</span></span>+<span id="cb1047-26"><a href="#cb1047-26" aria-hidden="true" tabindex="-1"></a> <span class="fu">},</span></span>+<span id="cb1047-27"><a href="#cb1047-27" aria-hidden="true" tabindex="-1"></a> <span class="dt">"StorageClass"</span><span class="fu">:</span> <span class="st">"INTELLIGENT_TIERING"</span></span>+<span id="cb1047-28"><a href="#cb1047-28" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>+<span id="cb1047-29"><a href="#cb1047-29" aria-hidden="true" tabindex="-1"></a><span class="ot">]</span></span></code></pre></div><p>Options:</p><ul><li>"all": If set then show all objects, not just ones with restore@@ -32562,27 +34451,27 @@ format.</p>multipart uploads.</p><p>You can call it with no bucket in which case it lists all bucket,with a bucket or with a bucket and path.</p>-<div class="sourceCode" id="cb1020"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb1020-1"><a href="#cb1020-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>-<span id="cb1020-2"><a href="#cb1020-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"rclone"</span><span class="fu">:</span> <span class="ot">[</span></span>-<span id="cb1020-3"><a href="#cb1020-3" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>-<span id="cb1020-4"><a href="#cb1020-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Initiated"</span><span class="fu">:</span> <span class="st">"2020-06-26T14:20:36Z"</span><span class="fu">,</span></span>-<span id="cb1020-5"><a href="#cb1020-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Initiator"</span><span class="fu">:</span> <span class="fu">{</span></span>-<span id="cb1020-6"><a href="#cb1020-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">"DisplayName"</span><span class="fu">:</span> <span class="st">"XXX"</span><span class="fu">,</span></span>-<span id="cb1020-7"><a href="#cb1020-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"ID"</span><span class="fu">:</span> <span class="st">"arn:aws:iam::XXX:user/XXX"</span></span>-<span id="cb1020-8"><a href="#cb1020-8" aria-hidden="true" tabindex="-1"></a> <span class="fu">},</span></span>-<span id="cb1020-9"><a href="#cb1020-9" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Key"</span><span class="fu">:</span> <span class="st">"KEY"</span><span class="fu">,</span></span>-<span id="cb1020-10"><a href="#cb1020-10" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Owner"</span><span class="fu">:</span> <span class="fu">{</span></span>-<span id="cb1020-11"><a href="#cb1020-11" aria-hidden="true" tabindex="-1"></a> <span class="dt">"DisplayName"</span><span class="fu">:</span> <span class="kw">null</span><span class="fu">,</span></span>-<span id="cb1020-12"><a href="#cb1020-12" aria-hidden="true" tabindex="-1"></a> <span class="dt">"ID"</span><span class="fu">:</span> <span class="st">"XXX"</span></span>-<span id="cb1020-13"><a href="#cb1020-13" aria-hidden="true" tabindex="-1"></a> <span class="fu">},</span></span>-<span id="cb1020-14"><a href="#cb1020-14" aria-hidden="true" tabindex="-1"></a> <span class="dt">"StorageClass"</span><span class="fu">:</span> <span class="st">"STANDARD"</span><span class="fu">,</span></span>-<span id="cb1020-15"><a href="#cb1020-15" aria-hidden="true" tabindex="-1"></a> <span class="dt">"UploadId"</span><span class="fu">:</span> <span class="st">"XXX"</span></span>-<span id="cb1020-16"><a href="#cb1020-16" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>-<span id="cb1020-17"><a href="#cb1020-17" aria-hidden="true" tabindex="-1"></a> <span class="ot">]</span><span class="fu">,</span></span>-<span id="cb1020-18"><a href="#cb1020-18" aria-hidden="true" tabindex="-1"></a> <span class="dt">"rclone-1000files"</span><span class="fu">:</span> <span class="ot">[]</span><span class="fu">,</span></span>-<span id="cb1020-19"><a href="#cb1020-19" aria-hidden="true" tabindex="-1"></a> <span class="dt">"rclone-dst"</span><span class="fu">:</span> <span class="ot">[]</span></span>-<span id="cb1020-20"><a href="#cb1020-20" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<div class="sourceCode" id="cb1050"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb1050-1"><a href="#cb1050-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb1050-2"><a href="#cb1050-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"rclone"</span><span class="fu">:</span> <span class="ot">[</span></span>+<span id="cb1050-3"><a href="#cb1050-3" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>+<span id="cb1050-4"><a href="#cb1050-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Initiated"</span><span class="fu">:</span> <span class="st">"2020-06-26T14:20:36Z"</span><span class="fu">,</span></span>+<span id="cb1050-5"><a href="#cb1050-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Initiator"</span><span class="fu">:</span> <span class="fu">{</span></span>+<span id="cb1050-6"><a href="#cb1050-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">"DisplayName"</span><span class="fu">:</span> <span class="st">"XXX"</span><span class="fu">,</span></span>+<span id="cb1050-7"><a href="#cb1050-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"ID"</span><span class="fu">:</span> <span class="st">"arn:aws:iam::XXX:user/XXX"</span></span>+<span id="cb1050-8"><a href="#cb1050-8" aria-hidden="true" tabindex="-1"></a> <span class="fu">},</span></span>+<span id="cb1050-9"><a href="#cb1050-9" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Key"</span><span class="fu">:</span> <span class="st">"KEY"</span><span class="fu">,</span></span>+<span id="cb1050-10"><a href="#cb1050-10" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Owner"</span><span class="fu">:</span> <span class="fu">{</span></span>+<span id="cb1050-11"><a href="#cb1050-11" aria-hidden="true" tabindex="-1"></a> <span class="dt">"DisplayName"</span><span class="fu">:</span> <span class="kw">null</span><span class="fu">,</span></span>+<span id="cb1050-12"><a href="#cb1050-12" aria-hidden="true" tabindex="-1"></a> <span class="dt">"ID"</span><span class="fu">:</span> <span class="st">"XXX"</span></span>+<span id="cb1050-13"><a href="#cb1050-13" aria-hidden="true" tabindex="-1"></a> <span class="fu">},</span></span>+<span id="cb1050-14"><a href="#cb1050-14" aria-hidden="true" tabindex="-1"></a> <span class="dt">"StorageClass"</span><span class="fu">:</span> <span class="st">"STANDARD"</span><span class="fu">,</span></span>+<span id="cb1050-15"><a href="#cb1050-15" aria-hidden="true" tabindex="-1"></a> <span class="dt">"UploadId"</span><span class="fu">:</span> <span class="st">"XXX"</span></span>+<span id="cb1050-16"><a href="#cb1050-16" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>+<span id="cb1050-17"><a href="#cb1050-17" aria-hidden="true" tabindex="-1"></a> <span class="ot">]</span><span class="fu">,</span></span>+<span id="cb1050-18"><a href="#cb1050-18" aria-hidden="true" tabindex="-1"></a> <span class="dt">"rclone-1000files"</span><span class="fu">:</span> <span class="ot">[]</span><span class="fu">,</span></span>+<span id="cb1050-19"><a href="#cb1050-19" aria-hidden="true" tabindex="-1"></a> <span class="dt">"rclone-dst"</span><span class="fu">:</span> <span class="ot">[]</span></span>+<span id="cb1050-20"><a href="#cb1050-20" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div><h3 id="cleanup-2">cleanup</h3><p>Remove unfinished multipart uploads.</p><pre class="console"><code>rclone backend cleanup remote: [options] [<arguments>+]</code></pre>@@ -32638,10 +34527,10 @@ will default to those currently in use.</p><p>If you want to use rclone to access a public bucket, configure with ablank <code>access_key_id</code> and <code>secret_access_key</code>.Your config should end up looking like this:</p>-<div class="sourceCode" id="cb1029"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1029-1"><a href="#cb1029-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[anons3]</span></span>-<span id="cb1029-2"><a href="#cb1029-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>-<span id="cb1029-3"><a href="#cb1029-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> AWS</span></span></code></pre></div>+<div class="sourceCode" id="cb1059"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1059-1"><a href="#cb1059-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[anons3]</span></span>+<span id="cb1059-2"><a href="#cb1059-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>+<span id="cb1059-3"><a href="#cb1059-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> AWS</span></span></code></pre></div><p>Then use it as normal with the name of the public bucket, e.g.</p><pre class="console"><code>rclone lsd anons3:1000genomes</code></pre><p>You will be able to list and copy data but not upload it.</p>@@ -32655,15 +34544,15 @@ href="#configuration">configuration</a> section above.</p><p>From rclone v1.69 <ahref="https://docs.aws.amazon.com/AmazonS3/latest/userguide/directory-buckets-overview.html">DirectoryBuckets</a> are supported.</p>-<p>You will need to set the <code>directory_buckets = true</code> config-parameter or use <code>--s3-directory-buckets</code>.</p>+<p>You will need to set the <code>directory_bucket = true</code> config+parameter or use <code>--s3-directory-bucket</code>.</p><p>Note that rclone cannot yet:</p><ul><li>Create directory buckets</li><li>List directory buckets</li></ul>-<p>See <a href="#s3-directory-buckets">the --s3-directory-buckets-flag</a> for more info</p>+<p>See <a href="#s3-directory-bucket">the --s3-directory-bucket flag</a>+for more info</p><h3 id="aws-snowball-edge">AWS Snowball Edge</h3><p><a href="https://aws.amazon.com/snowball/">AWS Snowball</a> is ahardware appliance used for transferring bulk data back to AWS. Its main@@ -32677,14 +34566,14 @@ query parameter based authentication.</p><p>With rclone v1.59 or later setting <code>upload_cutoff</code> shouldnot be necessary.</p><p>eg.</p>-<div class="sourceCode" id="cb1032"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1032-1"><a href="#cb1032-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[snowball]</span></span>-<span id="cb1032-2"><a href="#cb1032-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>-<span id="cb1032-3"><a href="#cb1032-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Other</span></span>-<span id="cb1032-4"><a href="#cb1032-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> YOUR_ACCESS_KEY</span></span>-<span id="cb1032-5"><a href="#cb1032-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> YOUR_SECRET_KEY</span></span>-<span id="cb1032-6"><a href="#cb1032-6" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> http://[IP of Snowball]:8080</span></span>-<span id="cb1032-7"><a href="#cb1032-7" aria-hidden="true" tabindex="-1"></a><span class="dt">upload_cutoff </span><span class="ot">=</span><span class="st"> </span><span class="dv">0</span></span></code></pre></div>+<div class="sourceCode" id="cb1062"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1062-1"><a href="#cb1062-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[snowball]</span></span>+<span id="cb1062-2"><a href="#cb1062-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>+<span id="cb1062-3"><a href="#cb1062-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Other</span></span>+<span id="cb1062-4"><a href="#cb1062-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> YOUR_ACCESS_KEY</span></span>+<span id="cb1062-5"><a href="#cb1062-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> YOUR_SECRET_KEY</span></span>+<span id="cb1062-6"><a href="#cb1062-6" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> http://[IP of Snowball]:8080</span></span>+<span id="cb1062-7"><a href="#cb1062-7" aria-hidden="true" tabindex="-1"></a><span class="dt">upload_cutoff </span><span class="ot">=</span><span class="st"> </span><span class="dv">0</span></span></code></pre></div><h3 id="alibaba-oss">Alibaba OSS</h3><p>Here is an example of making an <ahref="https://www.alibabacloud.com/product/oss/">Alibaba Cloud (Aliyun)@@ -32879,19 +34768,96 @@ e) Edit this remoted) Delete this remotey/e/d> y</code></pre><p>This will leave the config file looking like this.</p>-<div class="sourceCode" id="cb1036"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1036-1"><a href="#cb1036-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[ArvanCloud]</span></span>-<span id="cb1036-2"><a href="#cb1036-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>-<span id="cb1036-3"><a href="#cb1036-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> ArvanCloud</span></span>-<span id="cb1036-4"><a href="#cb1036-4" aria-hidden="true" tabindex="-1"></a><span class="dt">env_auth </span><span class="ot">=</span><span class="st"> </span><span class="kw">false</span></span>-<span id="cb1036-5"><a href="#cb1036-5" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> YOURACCESSKEY</span></span>-<span id="cb1036-6"><a href="#cb1036-6" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> YOURSECRETACCESSKEY</span></span>-<span id="cb1036-7"><a href="#cb1036-7" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span></span>-<span id="cb1036-8"><a href="#cb1036-8" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> s3.arvanstorage.com</span></span>-<span id="cb1036-9"><a href="#cb1036-9" aria-hidden="true" tabindex="-1"></a><span class="dt">location_constraint </span><span class="ot">=</span></span>-<span id="cb1036-10"><a href="#cb1036-10" aria-hidden="true" tabindex="-1"></a><span class="dt">acl </span><span class="ot">=</span></span>-<span id="cb1036-11"><a href="#cb1036-11" aria-hidden="true" tabindex="-1"></a><span class="dt">server_side_encryption </span><span class="ot">=</span></span>-<span id="cb1036-12"><a href="#cb1036-12" aria-hidden="true" tabindex="-1"></a><span class="dt">storage_class </span><span class="ot">=</span></span></code></pre></div>+<div class="sourceCode" id="cb1066"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1066-1"><a href="#cb1066-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[ArvanCloud]</span></span>+<span id="cb1066-2"><a href="#cb1066-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>+<span id="cb1066-3"><a href="#cb1066-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> ArvanCloud</span></span>+<span id="cb1066-4"><a href="#cb1066-4" aria-hidden="true" tabindex="-1"></a><span class="dt">env_auth </span><span class="ot">=</span><span class="st"> </span><span class="kw">false</span></span>+<span id="cb1066-5"><a href="#cb1066-5" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> YOURACCESSKEY</span></span>+<span id="cb1066-6"><a href="#cb1066-6" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> YOURSECRETACCESSKEY</span></span>+<span id="cb1066-7"><a href="#cb1066-7" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span></span>+<span id="cb1066-8"><a href="#cb1066-8" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> s3.arvanstorage.com</span></span>+<span id="cb1066-9"><a href="#cb1066-9" aria-hidden="true" tabindex="-1"></a><span class="dt">location_constraint </span><span class="ot">=</span></span>+<span id="cb1066-10"><a href="#cb1066-10" aria-hidden="true" tabindex="-1"></a><span class="dt">acl </span><span class="ot">=</span></span>+<span id="cb1066-11"><a href="#cb1066-11" aria-hidden="true" tabindex="-1"></a><span class="dt">server_side_encryption </span><span class="ot">=</span></span>+<span id="cb1066-12"><a href="#cb1066-12" aria-hidden="true" tabindex="-1"></a><span class="dt">storage_class </span><span class="ot">=</span></span></code></pre></div>+<h3 id="autonomys">Autonomys Auto Drive</h3>+<p><a href="https://ai3.storage/">Auto Drive</a> is permanent storage on+the Autonomys Network, with an S3-compatible API. Deleting an object+hides it, but its data and every earlier version are kept. Buckets are+created by the first upload into them.</p>+<p>Use your Auto Drive API key as the <code>access_key_id</code>, and+any value as the <code>secret_access_key</code>.</p>+<p>Here is an example of making a configuration. First run:</p>+<pre class="console"><code>rclone config</code></pre>+<p>This will guide you through an interactive setup process.</p>+<pre class="text"><code>No remotes found, make a new one?+n) New remote+s) Set configuration password+q) Quit config+n/s/q> n++Enter name for new remote.+name> autodrive++Option Storage.+Type of storage to configure.+Storage> s3++Option provider.+Choose your S3 provider.+provider> Autonomys++Option env_auth.+Get AWS credentials from runtime (environment variables or EC2/ECS meta data if no env vars).+Only applies if access_key_id and secret_access_key is blank.+env_auth> false++Option access_key_id.+AWS Access Key ID.+access_key_id> YOUR_API_KEY++Option secret_access_key.+AWS Secret Access Key (password).+secret_access_key> ANY_VALUE++Option endpoint.+Endpoint for S3 API.+Choose a number from below, or type in your own value.+ 1 / Auto Drive mainnet S3 endpoint+ \ (s3.auto-drive.autonomys.xyz)+endpoint> 1++Option bucket_object_lock_enabled.+Enable Object Lock when creating new buckets.+Enter a boolean value (true or false). Press Enter for the default (false).+bucket_object_lock_enabled>++Edit advanced config?+y) Yes+n) No (default)+y/n> n++Configuration complete.+Options:+- type: s3+- provider: Autonomys+- access_key_id: YOUR_API_KEY+- secret_access_key: ANY_VALUE+- endpoint: s3.auto-drive.autonomys.xyz+Keep this "autodrive" remote?+y) Yes this is OK (default)+e) Edit this remote+d) Delete this remote+y/e/d> y</code></pre>+<p>The resulting configuration file should look like:</p>+<div class="sourceCode" id="cb1069"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1069-1"><a href="#cb1069-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[autodrive]</span></span>+<span id="cb1069-2"><a href="#cb1069-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>+<span id="cb1069-3"><a href="#cb1069-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Autonomys</span></span>+<span id="cb1069-4"><a href="#cb1069-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> YOUR_API_KEY</span></span>+<span id="cb1069-5"><a href="#cb1069-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> ANY_VALUE</span></span>+<span id="cb1069-6"><a href="#cb1069-6" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> s3.auto-drive.autonomys.xyz</span></span></code></pre></div><h3 id="bizflycloud">BizflyCloud</h3><p><a href="https://bizflycloud.vn/simple-storage">Bizfly Cloud SimpleStorage</a> is an S3-compatible service with regions in Hanoi (HN) and@@ -32902,19 +34868,19 @@ Ho Chi Minh City (HCM).</p><li>HCM: <code>hcm.ss.bfcplatform.vn</code></li></ul><p>A minimal configuration looks like this.</p>-<div class="sourceCode" id="cb1037"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1037-1"><a href="#cb1037-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[bizfly]</span></span>-<span id="cb1037-2"><a href="#cb1037-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>-<span id="cb1037-3"><a href="#cb1037-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> BizflyCloud</span></span>-<span id="cb1037-4"><a href="#cb1037-4" aria-hidden="true" tabindex="-1"></a><span class="dt">env_auth </span><span class="ot">=</span><span class="st"> </span><span class="kw">false</span></span>-<span id="cb1037-5"><a href="#cb1037-5" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> YOUR_ACCESS_KEY</span></span>-<span id="cb1037-6"><a href="#cb1037-6" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> YOUR_SECRET_KEY</span></span>-<span id="cb1037-7"><a href="#cb1037-7" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span><span class="st"> HN</span></span>-<span id="cb1037-8"><a href="#cb1037-8" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> hn.ss.bfcplatform.vn</span></span>-<span id="cb1037-9"><a href="#cb1037-9" aria-hidden="true" tabindex="-1"></a><span class="dt">location_constraint </span><span class="ot">=</span></span>-<span id="cb1037-10"><a href="#cb1037-10" aria-hidden="true" tabindex="-1"></a><span class="dt">acl </span><span class="ot">=</span></span>-<span id="cb1037-11"><a href="#cb1037-11" aria-hidden="true" tabindex="-1"></a><span class="dt">server_side_encryption </span><span class="ot">=</span></span>-<span id="cb1037-12"><a href="#cb1037-12" aria-hidden="true" tabindex="-1"></a><span class="dt">storage_class </span><span class="ot">=</span></span></code></pre></div>+<div class="sourceCode" id="cb1070"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1070-1"><a href="#cb1070-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[bizfly]</span></span>+<span id="cb1070-2"><a href="#cb1070-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>+<span id="cb1070-3"><a href="#cb1070-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> BizflyCloud</span></span>+<span id="cb1070-4"><a href="#cb1070-4" aria-hidden="true" tabindex="-1"></a><span class="dt">env_auth </span><span class="ot">=</span><span class="st"> </span><span class="kw">false</span></span>+<span id="cb1070-5"><a href="#cb1070-5" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> YOUR_ACCESS_KEY</span></span>+<span id="cb1070-6"><a href="#cb1070-6" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> YOUR_SECRET_KEY</span></span>+<span id="cb1070-7"><a href="#cb1070-7" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span><span class="st"> HN</span></span>+<span id="cb1070-8"><a href="#cb1070-8" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> hn.ss.bfcplatform.vn</span></span>+<span id="cb1070-9"><a href="#cb1070-9" aria-hidden="true" tabindex="-1"></a><span class="dt">location_constraint </span><span class="ot">=</span></span>+<span id="cb1070-10"><a href="#cb1070-10" aria-hidden="true" tabindex="-1"></a><span class="dt">acl </span><span class="ot">=</span></span>+<span id="cb1070-11"><a href="#cb1070-11" aria-hidden="true" tabindex="-1"></a><span class="dt">server_side_encryption </span><span class="ot">=</span></span>+<span id="cb1070-12"><a href="#cb1070-12" aria-hidden="true" tabindex="-1"></a><span class="dt">storage_class </span><span class="ot">=</span></span></code></pre></div><p>Switch <code>region</code> and <code>endpoint</code> to<code>HCM</code> and <code>hcm.ss.bfcplatform.vn</code> for Ho Chi MinhCity.</p>@@ -32926,24 +34892,30 @@ interface.</p><p>To use rclone with Ceph, configure as above but leave the regionblank and set the endpoint. You should end up with something like thisin your config:</p>-<div class="sourceCode" id="cb1038"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1038-1"><a href="#cb1038-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[ceph]</span></span>-<span id="cb1038-2"><a href="#cb1038-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>-<span id="cb1038-3"><a href="#cb1038-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Ceph</span></span>-<span id="cb1038-4"><a href="#cb1038-4" aria-hidden="true" tabindex="-1"></a><span class="dt">env_auth </span><span class="ot">=</span><span class="st"> </span><span class="kw">false</span></span>-<span id="cb1038-5"><a href="#cb1038-5" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> XXX</span></span>-<span id="cb1038-6"><a href="#cb1038-6" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> YYY</span></span>-<span id="cb1038-7"><a href="#cb1038-7" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span></span>-<span id="cb1038-8"><a href="#cb1038-8" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> https://ceph.endpoint.example.com</span></span>-<span id="cb1038-9"><a href="#cb1038-9" aria-hidden="true" tabindex="-1"></a><span class="dt">location_constraint </span><span class="ot">=</span></span>-<span id="cb1038-10"><a href="#cb1038-10" aria-hidden="true" tabindex="-1"></a><span class="dt">acl </span><span class="ot">=</span></span>-<span id="cb1038-11"><a href="#cb1038-11" aria-hidden="true" tabindex="-1"></a><span class="dt">server_side_encryption </span><span class="ot">=</span></span>-<span id="cb1038-12"><a href="#cb1038-12" aria-hidden="true" tabindex="-1"></a><span class="dt">storage_class </span><span class="ot">=</span></span></code></pre></div>+<div class="sourceCode" id="cb1071"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1071-1"><a href="#cb1071-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[ceph]</span></span>+<span id="cb1071-2"><a href="#cb1071-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>+<span id="cb1071-3"><a href="#cb1071-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Ceph</span></span>+<span id="cb1071-4"><a href="#cb1071-4" aria-hidden="true" tabindex="-1"></a><span class="dt">env_auth </span><span class="ot">=</span><span class="st"> </span><span class="kw">false</span></span>+<span id="cb1071-5"><a href="#cb1071-5" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> XXX</span></span>+<span id="cb1071-6"><a href="#cb1071-6" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> YYY</span></span>+<span id="cb1071-7"><a href="#cb1071-7" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span></span>+<span id="cb1071-8"><a href="#cb1071-8" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> https://ceph.endpoint.example.com</span></span>+<span id="cb1071-9"><a href="#cb1071-9" aria-hidden="true" tabindex="-1"></a><span class="dt">location_constraint </span><span class="ot">=</span></span>+<span id="cb1071-10"><a href="#cb1071-10" aria-hidden="true" tabindex="-1"></a><span class="dt">acl </span><span class="ot">=</span></span>+<span id="cb1071-11"><a href="#cb1071-11" aria-hidden="true" tabindex="-1"></a><span class="dt">server_side_encryption </span><span class="ot">=</span></span>+<span id="cb1071-12"><a href="#cb1071-12" aria-hidden="true" tabindex="-1"></a><span class="dt">storage_class </span><span class="ot">=</span></span></code></pre></div><p>If you are using an older version of CEPH (e.g. 10.2.x Jewel) and aversion of rclone before v1.59 then you may need to supply the parameter<code>--s3-upload-cutoff 0</code> or put this in the config file as<code>upload_cutoff 0</code> to work around a bug which causes uploadingof small files to fail.</p>+<p>Rclone defaults <code>sign_accept_encoding</code> to false for the+Ceph provider. That avoids <code>SignatureDoesNotMatch</code> errors+when a reverse proxy (for example Cloudflare) rewrites the+<code>Accept-Encoding</code> request header. You can override this with+<code>--s3-sign-accept-encoding</code> /+<code>sign_accept_encoding</code> if needed.</p><p>Note also that Ceph sometimes puts <code>/</code> in the passwords itgives users. If you read the secret access key using the command linetools you will get a JSON blob with the <code>/</code> escaped as@@ -32951,25 +34923,25 @@ tools you will get a JSON blob with the <code>/</code> escaped asaccess key.</p><p>Eg the dump from Ceph looks something like this (irrelevant keysremoved).</p>-<div class="sourceCode" id="cb1039"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb1039-1"><a href="#cb1039-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>-<span id="cb1039-2"><a href="#cb1039-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"user_id"</span><span class="fu">:</span> <span class="st">"xxx"</span><span class="fu">,</span></span>-<span id="cb1039-3"><a href="#cb1039-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"display_name"</span><span class="fu">:</span> <span class="st">"xxxx"</span><span class="fu">,</span></span>-<span id="cb1039-4"><a href="#cb1039-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"keys"</span><span class="fu">:</span> <span class="ot">[</span></span>-<span id="cb1039-5"><a href="#cb1039-5" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>-<span id="cb1039-6"><a href="#cb1039-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">"user"</span><span class="fu">:</span> <span class="st">"xxx"</span><span class="fu">,</span></span>-<span id="cb1039-7"><a href="#cb1039-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"access_key"</span><span class="fu">:</span> <span class="st">"xxxxxx"</span><span class="fu">,</span></span>-<span id="cb1039-8"><a href="#cb1039-8" aria-hidden="true" tabindex="-1"></a> <span class="dt">"secret_key"</span><span class="fu">:</span> <span class="st">"xxxxxx</span><span class="ch">\/</span><span class="st">xxxx"</span></span>-<span id="cb1039-9"><a href="#cb1039-9" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>-<span id="cb1039-10"><a href="#cb1039-10" aria-hidden="true" tabindex="-1"></a> <span class="ot">]</span><span class="fu">,</span></span>-<span id="cb1039-11"><a href="#cb1039-11" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<div class="sourceCode" id="cb1072"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb1072-1"><a href="#cb1072-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb1072-2"><a href="#cb1072-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"user_id"</span><span class="fu">:</span> <span class="st">"xxx"</span><span class="fu">,</span></span>+<span id="cb1072-3"><a href="#cb1072-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"display_name"</span><span class="fu">:</span> <span class="st">"xxxx"</span><span class="fu">,</span></span>+<span id="cb1072-4"><a href="#cb1072-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"keys"</span><span class="fu">:</span> <span class="ot">[</span></span>+<span id="cb1072-5"><a href="#cb1072-5" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>+<span id="cb1072-6"><a href="#cb1072-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">"user"</span><span class="fu">:</span> <span class="st">"xxx"</span><span class="fu">,</span></span>+<span id="cb1072-7"><a href="#cb1072-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"access_key"</span><span class="fu">:</span> <span class="st">"xxxxxx"</span><span class="fu">,</span></span>+<span id="cb1072-8"><a href="#cb1072-8" aria-hidden="true" tabindex="-1"></a> <span class="dt">"secret_key"</span><span class="fu">:</span> <span class="st">"xxxxxx</span><span class="ch">\/</span><span class="st">xxxx"</span></span>+<span id="cb1072-9"><a href="#cb1072-9" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>+<span id="cb1072-10"><a href="#cb1072-10" aria-hidden="true" tabindex="-1"></a> <span class="ot">]</span><span class="fu">,</span></span>+<span id="cb1072-11"><a href="#cb1072-11" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div><p>Because this is a json dump, it is encoding the <code>/</code> as<code>\/</code>, so if you use the secret key as<code>xxxxxx/xxxx</code> it will work fine.</p><h3 id="china-mobile-ecloud-eos">China Mobile Ecloud Elastic ObjectStorage (EOS)</h3><p>Here is an example of making an <a-href="https:///ecloud.10086.cn/home/product-introduction/eos/">China+href="https://ecloud.10086.cn/home/product-introduction/eos/">ChinaMobile Ecloud Elastic Object Storage (EOS)</a> configuration. Firstrun:</p><pre class="console"><code>rclone config</code></pre>@@ -33212,6 +35184,107 @@ y) Yes this is OK (default)e) Edit this remoted) Delete this remotey/e/d> y</code></pre>+<h3 id="clever-cloud-cellar">Clever Cloud Cellar</h3>+<p><a+href="https://www.clever.cloud/product/cellar-object-storage/">Cellar</a>+is the S3-compatible object storage service of <a+href="https://www.clever.cloud/">Clever Cloud</a>, hosted in France.+Credentials and the endpoint are shown in the Cellar add-on dashboard,+or in the <code>CELLAR_ADDON_KEY_ID</code>,+<code>CELLAR_ADDON_KEY_SECRET</code> and <code>CELLAR_ADDON_HOST</code>+environment variables of applications linked to it.</p>+<p>Cellar does not apply Object Lock headers sent when initiating a+multipart upload. When setting Object Lock retention or legal hold+options, also set <code>--s3-object-lock-set-after-upload</code>: the+lock is then applied with separate API calls after the upload, so the+object is not protected by these settings until those calls succeed.</p>+<p>Cellar supports server-side encryption with customer-provided keys+(SSE-C), set with <code>sse_customer_algorithm</code> and+<code>sse_customer_key_base64</code>. It rejects server-side copies of+SSE-C objects unless they are done as multipart copies, so use+<code>--no-update-modtime</code> to avoid modification time updates, and+expect server-side moves and renames of encrypted objects smaller than+<code>--s3-copy-cutoff</code> to fail. See the <a+href="https://www.clever.cloud/developers/doc/deploy/storage/cellar/#server-side-encryption-with-customer-keys-sse-c">Cellar+SSE-C documentation</a>.</p>+<p>First run:</p>+<pre class="console"><code>rclone config</code></pre>+<pre class="text"><code>This will guide you through an interactive setup process:++n) New remote+n/s/q> n++Enter name for new remote.+name> cellar++Option Storage.+Type of storage to configure.+Choose a number from below, or type in your own value.+[snip]+XX / Amazon S3 Compliant Storage Providers including AWS, ..., Cellar, ... and others+ \ (s3)+Storage> s3++Option provider.+Choose your S3 provider.+[snip]+XX / Clever Cloud Cellar Object Storage+ \ (Cellar)+provider> Cellar++Option env_auth.+Get AWS credentials from runtime (environment variables or EC2/ECS meta data if no env vars).+ 1 / Enter AWS credentials in the next step.+ \ (false)+ 2 / Get AWS credentials from the environment (env vars or IAM).+ \ (true)+env_auth> 1++Option access_key_id.+AWS Access Key ID.+access_key_id> ACCESS_KEY++Option secret_access_key.+AWS Secret Access Key (password).+secret_access_key> SECRET_KEY++Option endpoint.+Endpoint for S3 API.+ 1 / Paris, France+ \ (cellar-c2.services.clever-cloud.com)+endpoint> 1++Option acl.+Canned ACL used when creating buckets and storing or copying objects.+[snip]+acl> 1++Option bucket_object_lock_enabled.+Enable Object Lock when creating new buckets.+bucket_object_lock_enabled>++Edit advanced config?+y/n> n++Configuration complete.+Options:+- type: s3+- provider: Cellar+- access_key_id: ACCESS_KEY+- secret_access_key: SECRET_KEY+- endpoint: cellar-c2.services.clever-cloud.com+- acl: private+Keep this "cellar" remote?+y/e/d> y</code></pre>+<p>This will leave the config file looking like this.</p>+<div class="sourceCode" id="cb1077"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1077-1"><a href="#cb1077-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[cellar]</span></span>+<span id="cb1077-2"><a href="#cb1077-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>+<span id="cb1077-3"><a href="#cb1077-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Cellar</span></span>+<span id="cb1077-4"><a href="#cb1077-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> ACCESS_KEY</span></span>+<span id="cb1077-5"><a href="#cb1077-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> SECRET_KEY</span></span>+<span id="cb1077-6"><a href="#cb1077-6" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> cellar-c2.services.clever-cloud.com</span></span>+<span id="cb1077-7"><a href="#cb1077-7" aria-hidden="true" tabindex="-1"></a><span class="dt">acl </span><span class="ot">=</span><span class="st"> private</span></span></code></pre></div><h3 id="cloudflare-r2">Cloudflare R2</h3><p><a href="https://blog.cloudflare.com/r2-open-beta/">Cloudflare R2</a>Storage allows developers to store large amounts of unstructured data@@ -33289,15 +35362,15 @@ e) Edit this remoted) Delete this remotey/e/d> y</code></pre><p>This will leave your config looking something like:</p>-<div class="sourceCode" id="cb1044"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1044-1"><a href="#cb1044-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[r2]</span></span>-<span id="cb1044-2"><a href="#cb1044-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>-<span id="cb1044-3"><a href="#cb1044-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Cloudflare</span></span>-<span id="cb1044-4"><a href="#cb1044-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> ACCESS_KEY</span></span>-<span id="cb1044-5"><a href="#cb1044-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> SECRET_ACCESS_KEY</span></span>-<span id="cb1044-6"><a href="#cb1044-6" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span><span class="st"> auto</span></span>-<span id="cb1044-7"><a href="#cb1044-7" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> https://ACCOUNT_ID.r2.cloudflarestorage.com</span></span>-<span id="cb1044-8"><a href="#cb1044-8" aria-hidden="true" tabindex="-1"></a><span class="dt">acl </span><span class="ot">=</span><span class="st"> private</span></span></code></pre></div>+<div class="sourceCode" id="cb1080"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1080-1"><a href="#cb1080-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[r2]</span></span>+<span id="cb1080-2"><a href="#cb1080-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>+<span id="cb1080-3"><a href="#cb1080-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Cloudflare</span></span>+<span id="cb1080-4"><a href="#cb1080-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> ACCESS_KEY</span></span>+<span id="cb1080-5"><a href="#cb1080-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> SECRET_ACCESS_KEY</span></span>+<span id="cb1080-6"><a href="#cb1080-6" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span><span class="st"> auto</span></span>+<span id="cb1080-7"><a href="#cb1080-7" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> https://ACCOUNT_ID.r2.cloudflarestorage.com</span></span>+<span id="cb1080-8"><a href="#cb1080-8" aria-hidden="true" tabindex="-1"></a><span class="dt">acl </span><span class="ot">=</span><span class="st"> private</span></span></code></pre></div><p>Now run <code>rclone lsf r2:</code> to see your buckets and<code>rclone lsf r2:bucket</code> to look within a bucket.</p><p>For R2 tokens with the "Object Read & Write" permission, you may@@ -33335,14 +35408,14 @@ region> eu-west-1 (or leave empty)endpoint> s3.cubbit.eu (or s3.{tenant_name}.cubbit.eu)acl></code></pre><p>The resulting configuration file should look like:</p>-<div class="sourceCode" id="cb1046"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1046-1"><a href="#cb1046-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[cubbit-ds3]</span></span>-<span id="cb1046-2"><a href="#cb1046-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>-<span id="cb1046-3"><a href="#cb1046-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Cubbit</span></span>-<span id="cb1046-4"><a href="#cb1046-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> ACCESS_KEY</span></span>-<span id="cb1046-5"><a href="#cb1046-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> SECRET_KEY</span></span>-<span id="cb1046-6"><a href="#cb1046-6" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span><span class="st"> eu-west-1</span></span>-<span id="cb1046-7"><a href="#cb1046-7" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> s3.cubbit.eu</span></span></code></pre></div>+<div class="sourceCode" id="cb1082"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1082-1"><a href="#cb1082-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[cubbit-ds3]</span></span>+<span id="cb1082-2"><a href="#cb1082-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>+<span id="cb1082-3"><a href="#cb1082-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Cubbit</span></span>+<span id="cb1082-4"><a href="#cb1082-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> ACCESS_KEY</span></span>+<span id="cb1082-5"><a href="#cb1082-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> SECRET_KEY</span></span>+<span id="cb1082-6"><a href="#cb1082-6" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span><span class="st"> eu-west-1</span></span>+<span id="cb1082-7"><a href="#cb1082-7" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> s3.cubbit.eu</span></span></code></pre></div><p>You can then start using Cubbit DS3 with rclone. For example, tocreate a new bucket and copy files into it, you can run:</p><pre class="console"><code>rclone mkdir cubbit-ds3:my-bucket@@ -33377,19 +35450,19 @@ location_constraint>acl>storage_class></code></pre><p>The resulting configuration file should look like:</p>-<div class="sourceCode" id="cb1049"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1049-1"><a href="#cb1049-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[spaces]</span></span>-<span id="cb1049-2"><a href="#cb1049-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>-<span id="cb1049-3"><a href="#cb1049-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> DigitalOcean</span></span>-<span id="cb1049-4"><a href="#cb1049-4" aria-hidden="true" tabindex="-1"></a><span class="dt">env_auth </span><span class="ot">=</span><span class="st"> </span><span class="kw">false</span></span>-<span id="cb1049-5"><a href="#cb1049-5" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> YOUR_ACCESS_KEY</span></span>-<span id="cb1049-6"><a href="#cb1049-6" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> YOUR_SECRET_KEY</span></span>-<span id="cb1049-7"><a href="#cb1049-7" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span></span>-<span id="cb1049-8"><a href="#cb1049-8" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> nyc3.digitaloceanspaces.com</span></span>-<span id="cb1049-9"><a href="#cb1049-9" aria-hidden="true" tabindex="-1"></a><span class="dt">location_constraint </span><span class="ot">=</span></span>-<span id="cb1049-10"><a href="#cb1049-10" aria-hidden="true" tabindex="-1"></a><span class="dt">acl </span><span class="ot">=</span></span>-<span id="cb1049-11"><a href="#cb1049-11" aria-hidden="true" tabindex="-1"></a><span class="dt">server_side_encryption </span><span class="ot">=</span></span>-<span id="cb1049-12"><a href="#cb1049-12" aria-hidden="true" tabindex="-1"></a><span class="dt">storage_class </span><span class="ot">=</span></span></code></pre></div>+<div class="sourceCode" id="cb1085"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1085-1"><a href="#cb1085-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[spaces]</span></span>+<span id="cb1085-2"><a href="#cb1085-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>+<span id="cb1085-3"><a href="#cb1085-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> DigitalOcean</span></span>+<span id="cb1085-4"><a href="#cb1085-4" aria-hidden="true" tabindex="-1"></a><span class="dt">env_auth </span><span class="ot">=</span><span class="st"> </span><span class="kw">false</span></span>+<span id="cb1085-5"><a href="#cb1085-5" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> YOUR_ACCESS_KEY</span></span>+<span id="cb1085-6"><a href="#cb1085-6" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> YOUR_SECRET_KEY</span></span>+<span id="cb1085-7"><a href="#cb1085-7" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span></span>+<span id="cb1085-8"><a href="#cb1085-8" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> nyc3.digitaloceanspaces.com</span></span>+<span id="cb1085-9"><a href="#cb1085-9" aria-hidden="true" tabindex="-1"></a><span class="dt">location_constraint </span><span class="ot">=</span></span>+<span id="cb1085-10"><a href="#cb1085-10" aria-hidden="true" tabindex="-1"></a><span class="dt">acl </span><span class="ot">=</span></span>+<span id="cb1085-11"><a href="#cb1085-11" aria-hidden="true" tabindex="-1"></a><span class="dt">server_side_encryption </span><span class="ot">=</span></span>+<span id="cb1085-12"><a href="#cb1085-12" aria-hidden="true" tabindex="-1"></a><span class="dt">storage_class </span><span class="ot">=</span></span></code></pre></div><p>Once configured, you can create a new Space and begin copying files.For example:</p><pre class="console"><code>rclone mkdir spaces:my-new-space@@ -33401,19 +35474,19 @@ object storage system based on CEPH.</p><p>To use rclone with Dreamhost, configure as above but leave the regionblank and set the endpoint. You should end up with something like thisin your config:</p>-<div class="sourceCode" id="cb1051"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1051-1"><a href="#cb1051-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[dreamobjects]</span></span>-<span id="cb1051-2"><a href="#cb1051-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>-<span id="cb1051-3"><a href="#cb1051-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> DreamHost</span></span>-<span id="cb1051-4"><a href="#cb1051-4" aria-hidden="true" tabindex="-1"></a><span class="dt">env_auth </span><span class="ot">=</span><span class="st"> </span><span class="kw">false</span></span>-<span id="cb1051-5"><a href="#cb1051-5" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> your_access_key</span></span>-<span id="cb1051-6"><a href="#cb1051-6" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> your_secret_key</span></span>-<span id="cb1051-7"><a href="#cb1051-7" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span></span>-<span id="cb1051-8"><a href="#cb1051-8" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> objects-us-west-1.dream.io</span></span>-<span id="cb1051-9"><a href="#cb1051-9" aria-hidden="true" tabindex="-1"></a><span class="dt">location_constraint </span><span class="ot">=</span></span>-<span id="cb1051-10"><a href="#cb1051-10" aria-hidden="true" tabindex="-1"></a><span class="dt">acl </span><span class="ot">=</span><span class="st"> private</span></span>-<span id="cb1051-11"><a href="#cb1051-11" aria-hidden="true" tabindex="-1"></a><span class="dt">server_side_encryption </span><span class="ot">=</span></span>-<span id="cb1051-12"><a href="#cb1051-12" aria-hidden="true" tabindex="-1"></a><span class="dt">storage_class </span><span class="ot">=</span></span></code></pre></div>+<div class="sourceCode" id="cb1087"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1087-1"><a href="#cb1087-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[dreamobjects]</span></span>+<span id="cb1087-2"><a href="#cb1087-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>+<span id="cb1087-3"><a href="#cb1087-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> DreamHost</span></span>+<span id="cb1087-4"><a href="#cb1087-4" aria-hidden="true" tabindex="-1"></a><span class="dt">env_auth </span><span class="ot">=</span><span class="st"> </span><span class="kw">false</span></span>+<span id="cb1087-5"><a href="#cb1087-5" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> your_access_key</span></span>+<span id="cb1087-6"><a href="#cb1087-6" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> your_secret_key</span></span>+<span id="cb1087-7"><a href="#cb1087-7" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span></span>+<span id="cb1087-8"><a href="#cb1087-8" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> objects-us-west-1.dream.io</span></span>+<span id="cb1087-9"><a href="#cb1087-9" aria-hidden="true" tabindex="-1"></a><span class="dt">location_constraint </span><span class="ot">=</span></span>+<span id="cb1087-10"><a href="#cb1087-10" aria-hidden="true" tabindex="-1"></a><span class="dt">acl </span><span class="ot">=</span><span class="st"> private</span></span>+<span id="cb1087-11"><a href="#cb1087-11" aria-hidden="true" tabindex="-1"></a><span class="dt">server_side_encryption </span><span class="ot">=</span></span>+<span id="cb1087-12"><a href="#cb1087-12" aria-hidden="true" tabindex="-1"></a><span class="dt">storage_class </span><span class="ot">=</span></span></code></pre></div><h3 id="exaba">Exaba</h3><p><a href="https://exaba.com/">Exaba</a> is an on-premises,S3-compatible storage for service providers and large enterprises. It is@@ -33474,13 +35547,13 @@ y) Yesn) No (default)y/n> n</code></pre><p>And the config generated will end up looking like this:</p>-<div class="sourceCode" id="cb1053"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1053-1"><a href="#cb1053-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[exaba]</span></span>-<span id="cb1053-2"><a href="#cb1053-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>-<span id="cb1053-3"><a href="#cb1053-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Exaba</span></span>-<span id="cb1053-4"><a href="#cb1053-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> XXX</span></span>-<span id="cb1053-5"><a href="#cb1053-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> XXX</span></span>-<span id="cb1053-6"><a href="#cb1053-6" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> http://127.0.0.1:9000/</span></span></code></pre></div>+<div class="sourceCode" id="cb1089"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1089-1"><a href="#cb1089-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[exaba]</span></span>+<span id="cb1089-2"><a href="#cb1089-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>+<span id="cb1089-3"><a href="#cb1089-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Exaba</span></span>+<span id="cb1089-4"><a href="#cb1089-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> XXX</span></span>+<span id="cb1089-5"><a href="#cb1089-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> XXX</span></span>+<span id="cb1089-6"><a href="#cb1089-6" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> http://127.0.0.1:9000/</span></span></code></pre></div><h3 id="fastly">Fastly Object Storage</h3><p><a href="https://www.fastly.com/products/storage">Fastly ObjectStorage</a> is an S3-compatible object storage service from Fastly. It@@ -33546,14 +35619,14 @@ e) Edit this remoted) Delete this remotey/e/d> y</code></pre><p>The resulting configuration file should look like:</p>-<div class="sourceCode" id="cb1056"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1056-1"><a href="#cb1056-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[fastly]</span></span>-<span id="cb1056-2"><a href="#cb1056-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>-<span id="cb1056-3"><a href="#cb1056-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Fastly</span></span>-<span id="cb1056-4"><a href="#cb1056-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> YOUR_ACCESS_KEY</span></span>-<span id="cb1056-5"><a href="#cb1056-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> YOUR_SECRET_KEY</span></span>-<span id="cb1056-6"><a href="#cb1056-6" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span><span class="st"> us-east</span></span>-<span id="cb1056-7"><a href="#cb1056-7" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> us-east.object.fastlystorage.app</span></span></code></pre></div>+<div class="sourceCode" id="cb1092"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1092-1"><a href="#cb1092-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[fastly]</span></span>+<span id="cb1092-2"><a href="#cb1092-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>+<span id="cb1092-3"><a href="#cb1092-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Fastly</span></span>+<span id="cb1092-4"><a href="#cb1092-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> YOUR_ACCESS_KEY</span></span>+<span id="cb1092-5"><a href="#cb1092-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> YOUR_SECRET_KEY</span></span>+<span id="cb1092-6"><a href="#cb1092-6" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span><span class="st"> us-east</span></span>+<span id="cb1092-7"><a href="#cb1092-7" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> us-east.object.fastlystorage.app</span></span></code></pre></div><h3 id="google-cloud-storage">Google Cloud Storage</h3><p><ahref="https://cloud.google.com/storage/docs">GoogleCloudStorage</a> is@@ -33564,13 +35637,13 @@ object storage service from Google Cloud Platform.</p>secret key. These can be retrieved by creating an <ahref="https://cloud.google.com/storage/docs/authentication/managing-hmackeys">HMACkey</a>.</p>-<div class="sourceCode" id="cb1057"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1057-1"><a href="#cb1057-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[gs]</span></span>-<span id="cb1057-2"><a href="#cb1057-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>-<span id="cb1057-3"><a href="#cb1057-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> GCS</span></span>-<span id="cb1057-4"><a href="#cb1057-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> your_access_key</span></span>-<span id="cb1057-5"><a href="#cb1057-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> your_secret_key</span></span>-<span id="cb1057-6"><a href="#cb1057-6" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> https://storage.googleapis.com</span></span></code></pre></div>+<div class="sourceCode" id="cb1093"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1093-1"><a href="#cb1093-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[gs]</span></span>+<span id="cb1093-2"><a href="#cb1093-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>+<span id="cb1093-3"><a href="#cb1093-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> GCS</span></span>+<span id="cb1093-4"><a href="#cb1093-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> your_access_key</span></span>+<span id="cb1093-5"><a href="#cb1093-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> your_secret_key</span></span>+<span id="cb1093-6"><a href="#cb1093-6" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> https://storage.googleapis.com</span></span></code></pre></div><p><strong>Note</strong> that <code>--s3-versions</code> does not workwith GCS when it needs to do directory paging. Rclone will return theerror:</p>@@ -33702,15 +35775,15 @@ s) Set configuration passwordq) Quit confige/n/d/r/c/s/q> </code></pre><p>This will leave the config file looking like this.</p>-<div class="sourceCode" id="cb1061"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1061-1"><a href="#cb1061-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[my-hetzner]</span></span>-<span id="cb1061-2"><a href="#cb1061-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>-<span id="cb1061-3"><a href="#cb1061-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Hetzner</span></span>-<span id="cb1061-4"><a href="#cb1061-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> ACCESS_KEY</span></span>-<span id="cb1061-5"><a href="#cb1061-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> SECRET_KEY</span></span>-<span id="cb1061-6"><a href="#cb1061-6" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span><span class="st"> hel1</span></span>-<span id="cb1061-7"><a href="#cb1061-7" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> hel1.your-objectstorage.com</span></span>-<span id="cb1061-8"><a href="#cb1061-8" aria-hidden="true" tabindex="-1"></a><span class="dt">acl </span><span class="ot">=</span><span class="st"> private</span></span></code></pre></div>+<div class="sourceCode" id="cb1097"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1097-1"><a href="#cb1097-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[my-hetzner]</span></span>+<span id="cb1097-2"><a href="#cb1097-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>+<span id="cb1097-3"><a href="#cb1097-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Hetzner</span></span>+<span id="cb1097-4"><a href="#cb1097-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> ACCESS_KEY</span></span>+<span id="cb1097-5"><a href="#cb1097-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> SECRET_KEY</span></span>+<span id="cb1097-6"><a href="#cb1097-6" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span><span class="st"> hel1</span></span>+<span id="cb1097-7"><a href="#cb1097-7" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> hel1.your-objectstorage.com</span></span>+<span id="cb1097-8"><a href="#cb1097-8" aria-hidden="true" tabindex="-1"></a><span class="dt">acl </span><span class="ot">=</span><span class="st"> private</span></span></code></pre></div><h3 id="hcp">Hitachi Content Platform (HCP)</h3><p>Here is an example of making a <ahref="https://www.hitachivantara.com/en-us/products/storage/content-platform.html">Hitachi@@ -33797,13 +35870,13 @@ s) Set configuration passwordq) Quit confige/n/d/r/c/s/q> </code></pre><p>This will leave the config file looking like this.</p>-<div class="sourceCode" id="cb1064"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1064-1"><a href="#cb1064-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[my-hcp]</span></span>-<span id="cb1064-2"><a href="#cb1064-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>-<span id="cb1064-3"><a href="#cb1064-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> HCP</span></span>-<span id="cb1064-4"><a href="#cb1064-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> ACCESS_KEY</span></span>-<span id="cb1064-5"><a href="#cb1064-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> SECRET_KEY</span></span>-<span id="cb1064-6"><a href="#cb1064-6" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> https://hcp.example.com</span></span></code></pre></div>+<div class="sourceCode" id="cb1100"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1100-1"><a href="#cb1100-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[my-hcp]</span></span>+<span id="cb1100-2"><a href="#cb1100-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>+<span id="cb1100-3"><a href="#cb1100-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> HCP</span></span>+<span id="cb1100-4"><a href="#cb1100-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> ACCESS_KEY</span></span>+<span id="cb1100-5"><a href="#cb1100-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> SECRET_KEY</span></span>+<span id="cb1100-6"><a href="#cb1100-6" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> https://hcp.example.com</span></span></code></pre></div><h4 id="known-limitations">Known limitations</h4><p>These limitations have been observed with HCP version 9.7.x.</p><p><strong>Automatic directory objects</strong></p>@@ -33831,15 +35904,15 @@ easy-to-use cloud storage that lets you store virtually any volume ofunstructured data in any format and access it from anywhere.</p><p>OBS provides an S3 interface, you can copy and modify the followingconfiguration and add it to your rclone configuration file.</p>-<div class="sourceCode" id="cb1065"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1065-1"><a href="#cb1065-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[obs]</span></span>-<span id="cb1065-2"><a href="#cb1065-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>-<span id="cb1065-3"><a href="#cb1065-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> HuaweiOBS</span></span>-<span id="cb1065-4"><a href="#cb1065-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> your-access-key-id</span></span>-<span id="cb1065-5"><a href="#cb1065-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> your-secret-access-key</span></span>-<span id="cb1065-6"><a href="#cb1065-6" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span><span class="st"> af-south-1</span></span>-<span id="cb1065-7"><a href="#cb1065-7" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> obs.af-south-1.myhuaweicloud.com</span></span>-<span id="cb1065-8"><a href="#cb1065-8" aria-hidden="true" tabindex="-1"></a><span class="dt">acl </span><span class="ot">=</span><span class="st"> private</span></span></code></pre></div>+<div class="sourceCode" id="cb1101"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1101-1"><a href="#cb1101-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[obs]</span></span>+<span id="cb1101-2"><a href="#cb1101-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>+<span id="cb1101-3"><a href="#cb1101-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> HuaweiOBS</span></span>+<span id="cb1101-4"><a href="#cb1101-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> your-access-key-id</span></span>+<span id="cb1101-5"><a href="#cb1101-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> your-secret-access-key</span></span>+<span id="cb1101-6"><a href="#cb1101-6" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span><span class="st"> af-south-1</span></span>+<span id="cb1101-7"><a href="#cb1101-7" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> obs.af-south-1.myhuaweicloud.com</span></span>+<span id="cb1101-8"><a href="#cb1101-8" aria-hidden="true" tabindex="-1"></a><span class="dt">acl </span><span class="ot">=</span><span class="st"> private</span></span></code></pre></div><p>Or you can also configure via the interactive command line:</p><pre class="text"><code>No remotes found, make a new one\?n) New remote@@ -34068,15 +36141,15 @@ Choose a number from below, or type in your own valueacl> 1</code></pre></li><li><p>Review the displayed configuration and accept to save the"remote" then quit. The config file should look like this</p>-<div class="sourceCode" id="cb1075"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1075-1"><a href="#cb1075-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[xxx]</span></span>-<span id="cb1075-2"><a href="#cb1075-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>-<span id="cb1075-3"><a href="#cb1075-3" aria-hidden="true" tabindex="-1"></a><span class="dt">Provider </span><span class="ot">=</span><span class="st"> IBMCOS</span></span>-<span id="cb1075-4"><a href="#cb1075-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> xxx</span></span>-<span id="cb1075-5"><a href="#cb1075-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> yyy</span></span>-<span id="cb1075-6"><a href="#cb1075-6" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> s3-api.us-geo.objectstorage.softlayer.net</span></span>-<span id="cb1075-7"><a href="#cb1075-7" aria-hidden="true" tabindex="-1"></a><span class="dt">location_constraint </span><span class="ot">=</span><span class="st"> us-standard</span></span>-<span id="cb1075-8"><a href="#cb1075-8" aria-hidden="true" tabindex="-1"></a><span class="dt">acl </span><span class="ot">=</span><span class="st"> private</span></span></code></pre></div></li>+<div class="sourceCode" id="cb1111"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1111-1"><a href="#cb1111-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[xxx]</span></span>+<span id="cb1111-2"><a href="#cb1111-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>+<span id="cb1111-3"><a href="#cb1111-3" aria-hidden="true" tabindex="-1"></a><span class="dt">Provider </span><span class="ot">=</span><span class="st"> IBMCOS</span></span>+<span id="cb1111-4"><a href="#cb1111-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> xxx</span></span>+<span id="cb1111-5"><a href="#cb1111-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> yyy</span></span>+<span id="cb1111-6"><a href="#cb1111-6" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> s3-api.us-geo.objectstorage.softlayer.net</span></span>+<span id="cb1111-7"><a href="#cb1111-7" aria-hidden="true" tabindex="-1"></a><span class="dt">location_constraint </span><span class="ot">=</span><span class="st"> us-standard</span></span>+<span id="cb1111-8"><a href="#cb1111-8" aria-hidden="true" tabindex="-1"></a><span class="dt">acl </span><span class="ot">=</span><span class="st"> private</span></span></code></pre></div></li><li><p>Execute rclone commands</p></li></ol><pre class="text"><code>1) Create a bucket.@@ -34377,15 +36450,121 @@ s) Set configuration passwordq) Quit confige/n/d/r/c/s/q> </code></pre><p>This will leave the config file looking like this.</p>-<div class="sourceCode" id="cb1084"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1084-1"><a href="#cb1084-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[my-impossible-cloud]</span></span>-<span id="cb1084-2"><a href="#cb1084-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>-<span id="cb1084-3"><a href="#cb1084-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> ImpossibleCloud</span></span>-<span id="cb1084-4"><a href="#cb1084-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> ACCESS_KEY</span></span>-<span id="cb1084-5"><a href="#cb1084-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> SECRET_KEY</span></span>-<span id="cb1084-6"><a href="#cb1084-6" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span><span class="st"> eu-central-2</span></span>-<span id="cb1084-7"><a href="#cb1084-7" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> eu-central-2.storage.impossibleapi.net</span></span>-<span id="cb1084-8"><a href="#cb1084-8" aria-hidden="true" tabindex="-1"></a><span class="dt">acl </span><span class="ot">=</span><span class="st"> private</span></span></code></pre></div>+<div class="sourceCode" id="cb1120"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1120-1"><a href="#cb1120-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[my-impossible-cloud]</span></span>+<span id="cb1120-2"><a href="#cb1120-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>+<span id="cb1120-3"><a href="#cb1120-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> ImpossibleCloud</span></span>+<span id="cb1120-4"><a href="#cb1120-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> ACCESS_KEY</span></span>+<span id="cb1120-5"><a href="#cb1120-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> SECRET_KEY</span></span>+<span id="cb1120-6"><a href="#cb1120-6" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span><span class="st"> eu-central-2</span></span>+<span id="cb1120-7"><a href="#cb1120-7" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> eu-central-2.storage.impossibleapi.net</span></span>+<span id="cb1120-8"><a href="#cb1120-8" aria-hidden="true" tabindex="-1"></a><span class="dt">acl </span><span class="ot">=</span><span class="st"> private</span></span></code></pre></div>+<h3 id="infomaniak">Infomaniak Swiss Backup</h3>+<p><a+href="https://www.infomaniak.com/en/swiss-backup/cloud-storage">Infomaniak+Swiss Backup</a> is a cloud backup space in Switzerland that you can use+however you want with the script or software of your choice that is+compatible with Swift, S3 or SFTP.</p>+<h4 id="prerequisites">Prerequisites</h4>+<p>Create and setup your Swiss Backup product following <a+href="https://www.infomaniak.com/en/support/faq/2546/create-an-s3-location-on-swiss-backup">this+documentation</a>. Memorize the generated <code>Endpoint</code>,+<code>Access Key</code>, and <code>Secret Key</code>:</p>+<h4 id="configuration-3">Configuration</h4>+<p>To configure access to an S3 location on Swiss Backup, follow the+steps below:</p>+<ol type="1">+<li>Run <code>rclone config</code> and select <code>n</code> for a new+remote.</li>+</ol>+<div class="sourceCode" id="cb1121"><pre+class="sourceCode sh"><code class="sourceCode bash"><span id="cb1121-1"><a href="#cb1121-1" aria-hidden="true" tabindex="-1"></a> <span class="ex">rclone</span> config</span>+<span id="cb1121-2"><a href="#cb1121-2" aria-hidden="true" tabindex="-1"></a> <span class="ex">No</span> remotes found, make a new one<span class="pp">?</span></span>+<span id="cb1121-3"><a href="#cb1121-3" aria-hidden="true" tabindex="-1"></a> <span class="ex">n</span><span class="er">)</span> <span class="ex">New</span> remote</span>+<span id="cb1121-4"><a href="#cb1121-4" aria-hidden="true" tabindex="-1"></a> <span class="ex">s</span><span class="er">)</span> <span class="ex">Set</span> configuration password</span>+<span id="cb1121-5"><a href="#cb1121-5" aria-hidden="true" tabindex="-1"></a> <span class="ex">q</span><span class="er">)</span> <span class="ex">Quit</span> config</span>+<span id="cb1121-6"><a href="#cb1121-6" aria-hidden="true" tabindex="-1"></a> <span class="ex">n/s/q</span><span class="op">></span> n</span></code></pre></div>+<ol start="2" type="1">+<li>Give the name of the configuration. For example, name it+'swissbackup'.</li>+</ol>+<div class="sourceCode" id="cb1122"><pre+class="sourceCode sh"><code class="sourceCode bash"><span id="cb1122-1"><a href="#cb1122-1" aria-hidden="true" tabindex="-1"></a> <span class="ex">name</span><span class="op">></span> swissbackup</span></code></pre></div>+<ol start="3" type="1">+<li>Select <code>s3</code> storage.</li>+</ol>+<div class="sourceCode" id="cb1123"><pre+class="sourceCode sh"><code class="sourceCode bash"><span id="cb1123-1"><a href="#cb1123-1" aria-hidden="true" tabindex="-1"></a> <span class="ex">Type</span> of storage to configure.</span>+<span id="cb1123-2"><a href="#cb1123-2" aria-hidden="true" tabindex="-1"></a> <span class="ex">XX</span> / Amazon S3 Compliant Storage Providers including AWS, ... <span class="at">-</span> Tier 1</span>+<span id="cb1123-3"><a href="#cb1123-3" aria-hidden="true" tabindex="-1"></a> <span class="ex">\ </span><span class="er">(</span><span class="ex">s3</span><span class="kw">)</span></span>+<span id="cb1123-4"><a href="#cb1123-4" aria-hidden="true" tabindex="-1"></a> <span class="ex">Storage</span><span class="op">></span> s3</span></code></pre></div>+<ol start="4" type="1">+<li>Select <code>Infomaniak</code> provider.</li>+</ol>+<div class="sourceCode" id="cb1124"><pre+class="sourceCode sh"><code class="sourceCode bash"><span id="cb1124-1"><a href="#cb1124-1" aria-hidden="true" tabindex="-1"></a> <span class="ex">Choose</span> your S3 provider.</span>+<span id="cb1124-2"><a href="#cb1124-2" aria-hidden="true" tabindex="-1"></a> <span class="ex">xx</span> / Infomaniak Swiss Backup</span>+<span id="cb1124-3"><a href="#cb1124-3" aria-hidden="true" tabindex="-1"></a> <span class="ex">\ </span><span class="er">(</span><span class="ex">Infomaniak</span><span class="kw">)</span></span>+<span id="cb1124-4"><a href="#cb1124-4" aria-hidden="true" tabindex="-1"></a> <span class="ex">provider</span><span class="op">></span> Infomaniak</span></code></pre></div>+<ol start="5" type="1">+<li>Enter your Access Key and Secret Key. <em>You can generate new ones+by clicking 'Generate new keys' on your Swiss Backup product+dashboard.</em></li>+</ol>+<div class="sourceCode" id="cb1125"><pre+class="sourceCode sh"><code class="sourceCode bash"><span id="cb1125-1"><a href="#cb1125-1" aria-hidden="true" tabindex="-1"></a> <span class="ex">Option</span> env_auth.</span>+<span id="cb1125-2"><a href="#cb1125-2" aria-hidden="true" tabindex="-1"></a> <span class="ex">1</span> / Enter AWS credentials in the next step</span>+<span id="cb1125-3"><a href="#cb1125-3" aria-hidden="true" tabindex="-1"></a> <span class="ex">\ </span><span class="er">(</span><span class="fu">false</span><span class="kw">)</span></span>+<span id="cb1125-4"><a href="#cb1125-4" aria-hidden="true" tabindex="-1"></a> <span class="ex">2</span> / Get AWS credentials from the environment <span class="er">(</span><span class="fu">env</span> vars or IAM<span class="kw">)</span></span>+<span id="cb1125-5"><a href="#cb1125-5" aria-hidden="true" tabindex="-1"></a> <span class="ex">\ </span><span class="er">(</span><span class="fu">true</span><span class="kw">)</span></span>+<span id="cb1125-6"><a href="#cb1125-6" aria-hidden="true" tabindex="-1"></a> <span class="ex">env_auth</span><span class="op">></span> 1</span>+<span id="cb1125-7"><a href="#cb1125-7" aria-hidden="true" tabindex="-1"></a></span>+<span id="cb1125-8"><a href="#cb1125-8" aria-hidden="true" tabindex="-1"></a> <span class="ex">AWS</span> Access Key ID.</span>+<span id="cb1125-9"><a href="#cb1125-9" aria-hidden="true" tabindex="-1"></a> <span class="ex">access_key_id</span><span class="op">></span> {YOUR-ACCESS-ID}</span>+<span id="cb1125-10"><a href="#cb1125-10" aria-hidden="true" tabindex="-1"></a></span>+<span id="cb1125-11"><a href="#cb1125-11" aria-hidden="true" tabindex="-1"></a> <span class="ex">AWS</span> Secret Access Key <span class="er">(</span><span class="ex">password</span><span class="kw">)</span></span>+<span id="cb1125-12"><a href="#cb1125-12" aria-hidden="true" tabindex="-1"></a> <span class="ex">secret_access_key</span><span class="op">></span> {YOUR-ACCESS-SECRET}</span></code></pre></div>+<ol start="6" type="1">+<li>Select <code>us-east-1</code> region for compatibility.</li>+</ol>+<div class="sourceCode" id="cb1126"><pre+class="sourceCode sh"><code class="sourceCode bash"><span id="cb1126-1"><a href="#cb1126-1" aria-hidden="true" tabindex="-1"></a> <span class="ex">Region</span> to connect to.</span>+<span id="cb1126-2"><a href="#cb1126-2" aria-hidden="true" tabindex="-1"></a> <span class="ex">/</span> Default region for compatibility.</span>+<span id="cb1126-3"><a href="#cb1126-3" aria-hidden="true" tabindex="-1"></a> <span class="ex">1</span> <span class="kw">|</span> <span class="ex">All</span> data is stored in Switzerland.</span>+<span id="cb1126-4"><a href="#cb1126-4" aria-hidden="true" tabindex="-1"></a> <span class="ex">\ </span><span class="er">(</span><span class="ex">us-east-1</span><span class="kw">)</span></span>+<span id="cb1126-5"><a href="#cb1126-5" aria-hidden="true" tabindex="-1"></a> <span class="ex">region</span><span class="op">></span> 1</span></code></pre></div>+<ol start="7" type="1">+<li>Select the endpoint corresponding to the one generated with your+Access and Secret Keys.</li>+</ol>+<div class="sourceCode" id="cb1127"><pre+class="sourceCode sh"><code class="sourceCode bash"><span id="cb1127-1"><a href="#cb1127-1" aria-hidden="true" tabindex="-1"></a> <span class="ex">Endpoint</span> for S3 API.</span>+<span id="cb1127-2"><a href="#cb1127-2" aria-hidden="true" tabindex="-1"></a> <span class="ex">1</span> / Swiss Backup Location 01</span>+<span id="cb1127-3"><a href="#cb1127-3" aria-hidden="true" tabindex="-1"></a> <span class="ex">\ </span><span class="er">(</span><span class="ex">s3.swiss-backup01.infomaniak.com</span><span class="kw">)</span></span>+<span id="cb1127-4"><a href="#cb1127-4" aria-hidden="true" tabindex="-1"></a> <span class="ex">2</span> / Swiss Backup Location 02</span>+<span id="cb1127-5"><a href="#cb1127-5" aria-hidden="true" tabindex="-1"></a> <span class="ex">\ </span><span class="er">(</span><span class="ex">s3.swiss-backup02.infomaniak.com</span><span class="kw">)</span></span>+<span id="cb1127-6"><a href="#cb1127-6" aria-hidden="true" tabindex="-1"></a> <span class="ex">3</span> / Swiss Backup Location 03</span>+<span id="cb1127-7"><a href="#cb1127-7" aria-hidden="true" tabindex="-1"></a> <span class="ex">\ </span><span class="er">(</span><span class="ex">s3.swiss-backup03.infomaniak.com</span><span class="kw">)</span></span>+<span id="cb1127-8"><a href="#cb1127-8" aria-hidden="true" tabindex="-1"></a> <span class="ex">4</span> / Swiss Backup Location 04</span>+<span id="cb1127-9"><a href="#cb1127-9" aria-hidden="true" tabindex="-1"></a> <span class="ex">\ </span><span class="er">(</span><span class="ex">s3.swiss-backup04.infomaniak.com</span><span class="kw">)</span></span>+<span id="cb1127-10"><a href="#cb1127-10" aria-hidden="true" tabindex="-1"></a> <span class="ex">endpoint</span><span class="op">></span> 1</span></code></pre></div>+<ol start="8" type="1">+<li><p>Adjust other parameters and advanced config at your convenience.+Leave blank if not sure.</p></li>+<li><p>You should obtain something like this.</p></li>+</ol>+<div class="sourceCode" id="cb1128"><pre+class="sourceCode sh"><code class="sourceCode bash"><span id="cb1128-1"><a href="#cb1128-1" aria-hidden="true" tabindex="-1"></a> <span class="ex">Configuration</span> complete.</span>+<span id="cb1128-2"><a href="#cb1128-2" aria-hidden="true" tabindex="-1"></a> <span class="ex">Options:</span></span>+<span id="cb1128-3"><a href="#cb1128-3" aria-hidden="true" tabindex="-1"></a> <span class="ex">-</span> type: s3</span>+<span id="cb1128-4"><a href="#cb1128-4" aria-hidden="true" tabindex="-1"></a> <span class="ex">-</span> provider: Infomaniak</span>+<span id="cb1128-5"><a href="#cb1128-5" aria-hidden="true" tabindex="-1"></a> <span class="ex">-</span> access_key_id: xxxxxxxxxx</span>+<span id="cb1128-6"><a href="#cb1128-6" aria-hidden="true" tabindex="-1"></a> <span class="ex">-</span> secret_access_key: xxxxxxxxxx</span>+<span id="cb1128-7"><a href="#cb1128-7" aria-hidden="true" tabindex="-1"></a> <span class="ex">-</span> region: us-east-1</span>+<span id="cb1128-8"><a href="#cb1128-8" aria-hidden="true" tabindex="-1"></a> <span class="ex">-</span> endpoint: s3.swiss-backup01.infomaniak.com</span></code></pre></div>+<ol start="10" type="1">+<li>Press <code>y</code> to validate. Then <code>q</code> to quit.</li>+</ol><h3 id="intercolo">Intercolo Object Storage</h3><p><a href="https://intercolo.de/object-storage">Intercolo ObjectStorage</a> offers GDPR-compliant, transparently priced, S3-compatible@@ -34496,14 +36675,14 @@ e) Edit this remoted) Delete this remotey/e/d> y</code></pre><p>This will leave the config file looking like this.</p>-<div class="sourceCode" id="cb1087"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1087-1"><a href="#cb1087-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[intercolo]</span></span>-<span id="cb1087-2"><a href="#cb1087-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>-<span id="cb1087-3"><a href="#cb1087-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Intercolo</span></span>-<span id="cb1087-4"><a href="#cb1087-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> ACCESS_KEY</span></span>-<span id="cb1087-5"><a href="#cb1087-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> SECRET_KEY</span></span>-<span id="cb1087-6"><a href="#cb1087-6" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span><span class="st"> de-fra</span></span>-<span id="cb1087-7"><a href="#cb1087-7" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> de-fra.i3storage.com</span></span></code></pre></div>+<div class="sourceCode" id="cb1131"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1131-1"><a href="#cb1131-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[intercolo]</span></span>+<span id="cb1131-2"><a href="#cb1131-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>+<span id="cb1131-3"><a href="#cb1131-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Intercolo</span></span>+<span id="cb1131-4"><a href="#cb1131-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> ACCESS_KEY</span></span>+<span id="cb1131-5"><a href="#cb1131-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> SECRET_KEY</span></span>+<span id="cb1131-6"><a href="#cb1131-6" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span><span class="st"> de-fra</span></span>+<span id="cb1131-7"><a href="#cb1131-7" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> de-fra.i3storage.com</span></span></code></pre></div><h3 id="ionos">IONOS Cloud</h3><p><a href="https://cloud.ionos.com/storage/object-storage">IONOS S3Object Storage</a> is a service offered by IONOS for storing and@@ -34811,19 +36990,19 @@ e) Edit this remoted) Delete this remotey/e/d> y</code></pre><p>This will leave the config file looking like this.</p>-<div class="sourceCode" id="cb1112"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1112-1"><a href="#cb1112-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[Liara]</span></span>-<span id="cb1112-2"><a href="#cb1112-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>-<span id="cb1112-3"><a href="#cb1112-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Liara</span></span>-<span id="cb1112-4"><a href="#cb1112-4" aria-hidden="true" tabindex="-1"></a><span class="dt">env_auth </span><span class="ot">=</span><span class="st"> </span><span class="kw">false</span></span>-<span id="cb1112-5"><a href="#cb1112-5" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> YOURACCESSKEY</span></span>-<span id="cb1112-6"><a href="#cb1112-6" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> YOURSECRETACCESSKEY</span></span>-<span id="cb1112-7"><a href="#cb1112-7" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span></span>-<span id="cb1112-8"><a href="#cb1112-8" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> storage.iran.liara.space</span></span>-<span id="cb1112-9"><a href="#cb1112-9" aria-hidden="true" tabindex="-1"></a><span class="dt">location_constraint </span><span class="ot">=</span></span>-<span id="cb1112-10"><a href="#cb1112-10" aria-hidden="true" tabindex="-1"></a><span class="dt">acl </span><span class="ot">=</span></span>-<span id="cb1112-11"><a href="#cb1112-11" aria-hidden="true" tabindex="-1"></a><span class="dt">server_side_encryption </span><span class="ot">=</span></span>-<span id="cb1112-12"><a href="#cb1112-12" aria-hidden="true" tabindex="-1"></a><span class="dt">storage_class </span><span class="ot">=</span></span></code></pre></div>+<div class="sourceCode" id="cb1156"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1156-1"><a href="#cb1156-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[Liara]</span></span>+<span id="cb1156-2"><a href="#cb1156-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>+<span id="cb1156-3"><a href="#cb1156-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Liara</span></span>+<span id="cb1156-4"><a href="#cb1156-4" aria-hidden="true" tabindex="-1"></a><span class="dt">env_auth </span><span class="ot">=</span><span class="st"> </span><span class="kw">false</span></span>+<span id="cb1156-5"><a href="#cb1156-5" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> YOURACCESSKEY</span></span>+<span id="cb1156-6"><a href="#cb1156-6" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> YOURSECRETACCESSKEY</span></span>+<span id="cb1156-7"><a href="#cb1156-7" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span></span>+<span id="cb1156-8"><a href="#cb1156-8" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> storage.iran.liara.space</span></span>+<span id="cb1156-9"><a href="#cb1156-9" aria-hidden="true" tabindex="-1"></a><span class="dt">location_constraint </span><span class="ot">=</span></span>+<span id="cb1156-10"><a href="#cb1156-10" aria-hidden="true" tabindex="-1"></a><span class="dt">acl </span><span class="ot">=</span></span>+<span id="cb1156-11"><a href="#cb1156-11" aria-hidden="true" tabindex="-1"></a><span class="dt">server_side_encryption </span><span class="ot">=</span></span>+<span id="cb1156-12"><a href="#cb1156-12" aria-hidden="true" tabindex="-1"></a><span class="dt">storage_class </span><span class="ot">=</span></span></code></pre></div><h3 id="linode">Linode</h3><p>Here is an example of making a <ahref="https://www.linode.com/products/object-storage/">Linode Object@@ -34963,13 +37142,17 @@ e) Edit this remoted) Delete this remotey/e/d> y</code></pre><p>This will leave the config file looking like this.</p>-<div class="sourceCode" id="cb1115"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1115-1"><a href="#cb1115-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[linode]</span></span>-<span id="cb1115-2"><a href="#cb1115-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>-<span id="cb1115-3"><a href="#cb1115-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Linode</span></span>-<span id="cb1115-4"><a href="#cb1115-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> ACCESS_KEY</span></span>-<span id="cb1115-5"><a href="#cb1115-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> SECRET_ACCESS_KEY</span></span>-<span id="cb1115-6"><a href="#cb1115-6" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> eu-central-1.linodeobjects.com</span></span></code></pre></div>+<div class="sourceCode" id="cb1159"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1159-1"><a href="#cb1159-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[linode]</span></span>+<span id="cb1159-2"><a href="#cb1159-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>+<span id="cb1159-3"><a href="#cb1159-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Linode</span></span>+<span id="cb1159-4"><a href="#cb1159-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> ACCESS_KEY</span></span>+<span id="cb1159-5"><a href="#cb1159-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> SECRET_ACCESS_KEY</span></span>+<span id="cb1159-6"><a href="#cb1159-6" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> eu-central-1.linodeobjects.com</span></span></code></pre></div>+<p>Linode Object Storage is Ceph-backed. Rclone defaults+<code>sign_accept_encoding</code> to false for this provider so SigV4+requests do not fail with <code>SignatureDoesNotMatch</code> (see the <a+href="#ceph">Ceph</a> section).</p><h3 id="magalu">Magalu</h3><p>Here is an example of making a <ahref="https://magalu.cloud/object-storage/">Magalu Object Storage</a>@@ -35071,13 +37254,13 @@ e) Edit this remoted) Delete this remotey/e/d> y</code></pre><p>This will leave the config file looking like this.</p>-<div class="sourceCode" id="cb1118"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1118-1"><a href="#cb1118-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[magalu]</span></span>-<span id="cb1118-2"><a href="#cb1118-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>-<span id="cb1118-3"><a href="#cb1118-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Magalu</span></span>-<span id="cb1118-4"><a href="#cb1118-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> ACCESS_KEY</span></span>-<span id="cb1118-5"><a href="#cb1118-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> SECRET_ACCESS_KEY</span></span>-<span id="cb1118-6"><a href="#cb1118-6" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> br-ne1.magaluobjects.com</span></span></code></pre></div>+<div class="sourceCode" id="cb1162"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1162-1"><a href="#cb1162-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[magalu]</span></span>+<span id="cb1162-2"><a href="#cb1162-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>+<span id="cb1162-3"><a href="#cb1162-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Magalu</span></span>+<span id="cb1162-4"><a href="#cb1162-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> ACCESS_KEY</span></span>+<span id="cb1162-5"><a href="#cb1162-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> SECRET_ACCESS_KEY</span></span>+<span id="cb1162-6"><a href="#cb1162-6" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> br-ne1.magaluobjects.com</span></span></code></pre></div><h3 id="mega">MEGA S4</h3><p><a href="https://mega.io/objectstorage">MEGA S4 Object Storage</a> isan S3 compatible object storage system. It has a single pricing tier@@ -35170,13 +37353,13 @@ e) Edit this remoted) Delete this remotey/e/d> y</code></pre><p>This will leave the config file looking like this.</p>-<div class="sourceCode" id="cb1121"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1121-1"><a href="#cb1121-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[megas4]</span></span>-<span id="cb1121-2"><a href="#cb1121-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>-<span id="cb1121-3"><a href="#cb1121-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Mega</span></span>-<span id="cb1121-4"><a href="#cb1121-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> XXX</span></span>-<span id="cb1121-5"><a href="#cb1121-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> XXX</span></span>-<span id="cb1121-6"><a href="#cb1121-6" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> s3.eu-central-1.s4.mega.io</span></span></code></pre></div>+<div class="sourceCode" id="cb1165"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1165-1"><a href="#cb1165-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[megas4]</span></span>+<span id="cb1165-2"><a href="#cb1165-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>+<span id="cb1165-3"><a href="#cb1165-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Mega</span></span>+<span id="cb1165-4"><a href="#cb1165-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> XXX</span></span>+<span id="cb1165-5"><a href="#cb1165-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> XXX</span></span>+<span id="cb1165-6"><a href="#cb1165-6" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> s3.eu-central-1.s4.mega.io</span></span></code></pre></div><h3 id="minio">Minio</h3><p><a href="https://minio.io/">Minio</a> is an object storage serverbuilt for cloud application developers and devops.</p>@@ -35215,19 +37398,163 @@ endpoint> http://192.168.1.106:9000location_constraint>server_side_encryption></code></pre><p>Which makes the config file look like this</p>-<div class="sourceCode" id="cb1124"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1124-1"><a href="#cb1124-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[minio]</span></span>-<span id="cb1124-2"><a href="#cb1124-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>-<span id="cb1124-3"><a href="#cb1124-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Minio</span></span>-<span id="cb1124-4"><a href="#cb1124-4" aria-hidden="true" tabindex="-1"></a><span class="dt">env_auth </span><span class="ot">=</span><span class="st"> </span><span class="kw">false</span></span>-<span id="cb1124-5"><a href="#cb1124-5" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> USWUXHGYZQYFYFFIT3RE</span></span>-<span id="cb1124-6"><a href="#cb1124-6" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> MOJRH0mkL1IPauahWITSVvyDrQbEEIwljvmxdq03</span></span>-<span id="cb1124-7"><a href="#cb1124-7" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span><span class="st"> us-east-1</span></span>-<span id="cb1124-8"><a href="#cb1124-8" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> http://192.168.1.106:9000</span></span>-<span id="cb1124-9"><a href="#cb1124-9" aria-hidden="true" tabindex="-1"></a><span class="dt">location_constraint </span><span class="ot">=</span></span>-<span id="cb1124-10"><a href="#cb1124-10" aria-hidden="true" tabindex="-1"></a><span class="dt">server_side_encryption </span><span class="ot">=</span></span></code></pre></div>+<div class="sourceCode" id="cb1168"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1168-1"><a href="#cb1168-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[minio]</span></span>+<span id="cb1168-2"><a href="#cb1168-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>+<span id="cb1168-3"><a href="#cb1168-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Minio</span></span>+<span id="cb1168-4"><a href="#cb1168-4" aria-hidden="true" tabindex="-1"></a><span class="dt">env_auth </span><span class="ot">=</span><span class="st"> </span><span class="kw">false</span></span>+<span id="cb1168-5"><a href="#cb1168-5" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> USWUXHGYZQYFYFFIT3RE</span></span>+<span id="cb1168-6"><a href="#cb1168-6" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> MOJRH0mkL1IPauahWITSVvyDrQbEEIwljvmxdq03</span></span>+<span id="cb1168-7"><a href="#cb1168-7" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span><span class="st"> us-east-1</span></span>+<span id="cb1168-8"><a href="#cb1168-8" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> http://192.168.1.106:9000</span></span>+<span id="cb1168-9"><a href="#cb1168-9" aria-hidden="true" tabindex="-1"></a><span class="dt">location_constraint </span><span class="ot">=</span></span>+<span id="cb1168-10"><a href="#cb1168-10" aria-hidden="true" tabindex="-1"></a><span class="dt">server_side_encryption </span><span class="ot">=</span></span></code></pre></div><p>So once set up, for example, to copy files into a bucket</p><pre class="console"><code>rclone copy /path/to/files minio:bucket</code></pre>+<h3 id="neon">Neon Object Storage</h3>+<p><a href="https://neon.com/docs/storage/overview">Neon Object+Storage</a> is S3 compatible object storage in a Neon project. Each+branch of the project has its own copy of the buckets and its own S3+endpoint. The endpoint looks like+<code>https://<branch_id>.storage.<cell>.<region>.aws.neon.tech</code>.</p>+<p>In the Neon Console, click <strong>Connect</strong>, open the+<strong>Storage</strong> tab and choose <strong>Parameters+only</strong>. Use <code>AWS_ACCESS_KEY_ID</code>,+<code>AWS_SECRET_ACCESS_KEY</code>, <code>AWS_ENDPOINT_URL_S3</code> and+<code>AWS_REGION</code> as <code>access_key_id</code>,+<code>secret_access_key</code>, <code>endpoint</code> and+<code>region</code>. To make a separate credential, use+<strong>Settings</strong> > <strong>Credentials</strong> or run+<code>neon credentials create --scope storage:read --scope storage:write</code>.</p>+<p>A credential works on the branch it was made on and on that branch's+child branches. To use a child branch, change only+<code>endpoint</code>.</p>+<p>Neon also has these limits (see the <a+href="https://neon.com/docs/storage/s3-compatibility">S3 compatibility+docs</a>):</p>+<ul>+<li>The maximum object size is 5 GiB for a single upload and 10 GiB for+a multipart upload.</li>+<li>Neon rejects object keys containing control characters with+<code>400 InvalidURI</code>. To store files with these names, add+<code>Ctl,Del</code> to the encoding, for example+<code>encoding = Slash,InvalidUtf8,Dot,Ctl,Del</code>.</li>+<li>Neon doesn't support ACLs over the S3 API. Leave <code>acl</code>+and <code>bucket_acl</code> empty and set bucket access+(<code>private</code> or <code>public_read</code>) in Neon.</li>+<li>Neon stores versioning and lifecycle configurations but doesn't+enforce them.</li>+<li>Neon may throttle heavy use with <code>503 SlowDown</code> errors.+rclone retries these.</li>+</ul>+<p>Here is an example of making a Neon Object Storage configuration.+First run:</p>+<pre class="console"><code>rclone config</code></pre>+<p>This will guide you through an interactive setup process.</p>+<pre class="text"><code>No remotes found, make a new one?+n) New remote+s) Set configuration password+q) Quit config+n/s/q> n++Enter name for new remote.+name> neon++Option Storage.+Type of storage to configure.+Choose a number from below, or type in your own value.+[snip]+XX / Amazon S3 Compliant Storage Providers including AWS, ... Neon, ...+ \ (s3)+[snip]+Storage> s3++Option provider.+Choose your S3 provider.+Choose a number from below, or type in your own value.+Press Enter to leave empty.+[snip]+XX / Neon Object Storage+ \ (Neon)+[snip]+provider> Neon++Option env_auth.+Get AWS credentials from runtime (environment variables or EC2/ECS meta data if no env vars).+Only applies if access_key_id and secret_access_key is blank.+Choose a number from below, or type in your own boolean value (true or false).+Press Enter for the default (false).+ 1 / Enter AWS credentials in the next step.+ \ (false)+ 2 / Get AWS credentials from the environment (env vars or IAM).+ \ (true)+env_auth>++Option access_key_id.+AWS Access Key ID.+Leave blank for anonymous access or runtime credentials.+Enter a value. Press Enter to leave empty.+access_key_id> NEON_ACCESS_KEY_ID++Option secret_access_key.+AWS Secret Access Key (password).+Leave blank for anonymous access or runtime credentials.+Enter a value. Press Enter to leave empty.+secret_access_key> NEON_SECRET_ACCESS_KEY++Option region.+Region to connect to.+Leave blank if you are using an S3 clone and you don't have a region.+Choose a number from below, or type in your own value.+Press Enter to leave empty.+ 1 / US East (N. Virginia)+ \ (us-east-1)+ 2 / US East (Ohio)+ \ (us-east-2)+ 3 / Europe (Frankfurt)+ \ (eu-central-1)+ 4 / Asia Pacific (Singapore)+ \ (ap-southeast-1)+region> us-east-2++Option endpoint.+Endpoint for S3 API.+Required when using an S3 clone.+Enter a value. Press Enter to leave empty.+endpoint> https://<branch_id>.storage.<cell>.<region>.aws.neon.tech++Option bucket_object_lock_enabled.+Enable Object Lock when creating new buckets.+Enter a boolean value (true or false). Press Enter for the default (false).+bucket_object_lock_enabled>++Edit advanced config?+y) Yes+n) No (default)+y/n> n++Configuration complete.+Options:+- type: s3+- provider: Neon+- access_key_id: NEON_ACCESS_KEY_ID+- secret_access_key: NEON_SECRET_ACCESS_KEY+- region: us-east-2+- endpoint: https://<branch_id>.storage.<cell>.<region>.aws.neon.tech+Keep this "neon" remote?+y) Yes this is OK (default)+e) Edit this remote+d) Delete this remote+y/e/d> y</code></pre>+<p>This will leave the config file looking like this.</p>+<div class="sourceCode" id="cb1172"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1172-1"><a href="#cb1172-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[neon]</span></span>+<span id="cb1172-2"><a href="#cb1172-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>+<span id="cb1172-3"><a href="#cb1172-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Neon</span></span>+<span id="cb1172-4"><a href="#cb1172-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> NEON_ACCESS_KEY_ID</span></span>+<span id="cb1172-5"><a href="#cb1172-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> NEON_SECRET_ACCESS_KEY</span></span>+<span id="cb1172-6"><a href="#cb1172-6" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span><span class="st"> us-east-2</span></span>+<span id="cb1172-7"><a href="#cb1172-7" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> https://<branch_id>.storage.<cell>.<region>.aws.neon.tech</span></span></code></pre></div><h3 id="netease-nos">Netease NOS</h3><p>For Netease NOS configure as per the configurator<code>rclone config</code> setting the provider <code>Netease</code>.@@ -35243,16 +37570,16 @@ href="https://docs.outscale.com/en/userguide/OUTSCALE-Object-Storage-OOS.html">odocumentation</a>.</p><p>Here is an example of an OOS configuration that you can paste intoyour rclone configuration file:</p>-<div class="sourceCode" id="cb1126"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1126-1"><a href="#cb1126-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[outscale]</span></span>-<span id="cb1126-2"><a href="#cb1126-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>-<span id="cb1126-3"><a href="#cb1126-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Outscale</span></span>-<span id="cb1126-4"><a href="#cb1126-4" aria-hidden="true" tabindex="-1"></a><span class="dt">env_auth </span><span class="ot">=</span><span class="st"> </span><span class="kw">false</span></span>-<span id="cb1126-5"><a href="#cb1126-5" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> ABCDEFGHIJ0123456789</span></span>-<span id="cb1126-6"><a href="#cb1126-6" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX</span></span>-<span id="cb1126-7"><a href="#cb1126-7" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span><span class="st"> eu-west-2</span></span>-<span id="cb1126-8"><a href="#cb1126-8" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> oos.eu-west-2.outscale.com</span></span>-<span id="cb1126-9"><a href="#cb1126-9" aria-hidden="true" tabindex="-1"></a><span class="dt">acl </span><span class="ot">=</span><span class="st"> private</span></span></code></pre></div>+<div class="sourceCode" id="cb1173"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1173-1"><a href="#cb1173-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[outscale]</span></span>+<span id="cb1173-2"><a href="#cb1173-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>+<span id="cb1173-3"><a href="#cb1173-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Outscale</span></span>+<span id="cb1173-4"><a href="#cb1173-4" aria-hidden="true" tabindex="-1"></a><span class="dt">env_auth </span><span class="ot">=</span><span class="st"> </span><span class="kw">false</span></span>+<span id="cb1173-5"><a href="#cb1173-5" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> ABCDEFGHIJ0123456789</span></span>+<span id="cb1173-6"><a href="#cb1173-6" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX</span></span>+<span id="cb1173-7"><a href="#cb1173-7" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span><span class="st"> eu-west-2</span></span>+<span id="cb1173-8"><a href="#cb1173-8" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> oos.eu-west-2.outscale.com</span></span>+<span id="cb1173-9"><a href="#cb1173-9" aria-hidden="true" tabindex="-1"></a><span class="dt">acl </span><span class="ot">=</span><span class="st"> private</span></span></code></pre></div><p>You can also run <code>rclone config</code> to go through theinteractive setup process:</p><pre class="text"><code>No remotes found, make a new one\?@@ -35546,15 +37873,15 @@ e) Edit this remoted) Delete this remotey/e/d> y</code></pre><p>Your configuration file should now look like this:</p>-<div class="sourceCode" id="cb1140"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1140-1"><a href="#cb1140-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[ovhcloud-rbx]</span></span>-<span id="cb1140-2"><a href="#cb1140-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>-<span id="cb1140-3"><a href="#cb1140-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> OVHcloud</span></span>-<span id="cb1140-4"><a href="#cb1140-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> my_access</span></span>-<span id="cb1140-5"><a href="#cb1140-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> my_secret</span></span>-<span id="cb1140-6"><a href="#cb1140-6" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span><span class="st"> rbx</span></span>-<span id="cb1140-7"><a href="#cb1140-7" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> s3.rbx.io.cloud.ovh.net</span></span>-<span id="cb1140-8"><a href="#cb1140-8" aria-hidden="true" tabindex="-1"></a><span class="dt">acl </span><span class="ot">=</span><span class="st"> private</span></span></code></pre></div>+<div class="sourceCode" id="cb1187"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1187-1"><a href="#cb1187-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[ovhcloud-rbx]</span></span>+<span id="cb1187-2"><a href="#cb1187-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>+<span id="cb1187-3"><a href="#cb1187-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> OVHcloud</span></span>+<span id="cb1187-4"><a href="#cb1187-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> my_access</span></span>+<span id="cb1187-5"><a href="#cb1187-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> my_secret</span></span>+<span id="cb1187-6"><a href="#cb1187-6" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span><span class="st"> rbx</span></span>+<span id="cb1187-7"><a href="#cb1187-7" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> s3.rbx.io.cloud.ovh.net</span></span>+<span id="cb1187-8"><a href="#cb1187-8" aria-hidden="true" tabindex="-1"></a><span class="dt">acl </span><span class="ot">=</span><span class="st"> private</span></span></code></pre></div><h3 id="petabox">Petabox</h3><p>Here is an example of making a <ahref="https://petabox.io/">Petabox</a> configuration. First run:</p>@@ -35695,14 +38022,14 @@ e) Edit this remoted) Delete this remotey/e/d> y</code></pre><p>This will leave the config file looking like this.</p>-<div class="sourceCode" id="cb1143"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1143-1"><a href="#cb1143-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[My Petabox Storage]</span></span>-<span id="cb1143-2"><a href="#cb1143-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>-<span id="cb1143-3"><a href="#cb1143-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Petabox</span></span>-<span id="cb1143-4"><a href="#cb1143-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> YOUR_ACCESS_KEY_ID</span></span>-<span id="cb1143-5"><a href="#cb1143-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> YOUR_SECRET_ACCESS_KEY</span></span>-<span id="cb1143-6"><a href="#cb1143-6" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span><span class="st"> us-east-1</span></span>-<span id="cb1143-7"><a href="#cb1143-7" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> s3.petabox.io</span></span></code></pre></div>+<div class="sourceCode" id="cb1190"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1190-1"><a href="#cb1190-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[My Petabox Storage]</span></span>+<span id="cb1190-2"><a href="#cb1190-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>+<span id="cb1190-3"><a href="#cb1190-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Petabox</span></span>+<span id="cb1190-4"><a href="#cb1190-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> YOUR_ACCESS_KEY_ID</span></span>+<span id="cb1190-5"><a href="#cb1190-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> YOUR_SECRET_ACCESS_KEY</span></span>+<span id="cb1190-6"><a href="#cb1190-6" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span><span class="st"> us-east-1</span></span>+<span id="cb1190-7"><a href="#cb1190-7" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> s3.petabox.io</span></span></code></pre></div><h3 id="pure-storage-flashblade">Pure Storage FlashBlade</h3><p><ahref="https://www.purestorage.com/products/unstructured-data-storage.html">Pure@@ -35797,13 +38124,13 @@ d) Delete this remotey/e/d> y</code></pre><p>This results in the following configuration being stored in<code>~/.config/rclone/rclone.conf</code>:</p>-<div class="sourceCode" id="cb1146"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1146-1"><a href="#cb1146-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[flashblade]</span></span>-<span id="cb1146-2"><a href="#cb1146-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>-<span id="cb1146-3"><a href="#cb1146-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> FlashBlade</span></span>-<span id="cb1146-4"><a href="#cb1146-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> ACCESS_KEY_ID</span></span>-<span id="cb1146-5"><a href="#cb1146-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> SECRET_ACCESS_KEY</span></span>-<span id="cb1146-6"><a href="#cb1146-6" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> https://s3.flashblade.example.com</span></span></code></pre></div>+<div class="sourceCode" id="cb1193"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1193-1"><a href="#cb1193-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[flashblade]</span></span>+<span id="cb1193-2"><a href="#cb1193-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>+<span id="cb1193-3"><a href="#cb1193-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> FlashBlade</span></span>+<span id="cb1193-4"><a href="#cb1193-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> ACCESS_KEY_ID</span></span>+<span id="cb1193-5"><a href="#cb1193-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> SECRET_ACCESS_KEY</span></span>+<span id="cb1193-6"><a href="#cb1193-6" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> https://s3.flashblade.example.com</span></span></code></pre></div><p>Note: The FlashBlade endpoint should be the S3 data VIP. Forvirtual-hosted style requests, ensure proper DNS configuration:subdomains of the endpoint hostname should resolve to a FlashBlade data@@ -36080,13 +38407,13 @@ e) Edit this remoted) Delete this remotey/e/d> y</code></pre><p>This will leave the config file looking like this.</p>-<div class="sourceCode" id="cb1156"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1156-1"><a href="#cb1156-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[s5lu]</span></span>-<span id="cb1156-2"><a href="#cb1156-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>-<span id="cb1156-3"><a href="#cb1156-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> FileLu</span></span>-<span id="cb1156-4"><a href="#cb1156-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> XXX</span></span>-<span id="cb1156-5"><a href="#cb1156-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> XXX</span></span>-<span id="cb1156-6"><a href="#cb1156-6" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> s5lu.com</span></span></code></pre></div>+<div class="sourceCode" id="cb1203"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1203-1"><a href="#cb1203-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[s5lu]</span></span>+<span id="cb1203-2"><a href="#cb1203-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>+<span id="cb1203-3"><a href="#cb1203-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> FileLu</span></span>+<span id="cb1203-4"><a href="#cb1203-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> XXX</span></span>+<span id="cb1203-5"><a href="#cb1203-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> XXX</span></span>+<span id="cb1203-6"><a href="#cb1203-6" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> s5lu.com</span></span></code></pre></div><h3 id="Rabata">Rabata</h3><p><a href="https://rabata.io">Rabata</a> is an S3-compatible securecloud storage service that offers flat, transparent pricing (no API@@ -36223,16 +38550,16 @@ details are required for the next steps of configuration, when<code>rclone config</code> asks for your <code>access_key_id</code> and<code>secret_access_key</code>.</p><p>Your config should end up looking a bit like this:</p>-<div class="sourceCode" id="cb1158"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1158-1"><a href="#cb1158-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[RCS3-demo-config]</span></span>-<span id="cb1158-2"><a href="#cb1158-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>-<span id="cb1158-3"><a href="#cb1158-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> RackCorp</span></span>-<span id="cb1158-4"><a href="#cb1158-4" aria-hidden="true" tabindex="-1"></a><span class="dt">env_auth </span><span class="ot">=</span><span class="st"> </span><span class="kw">true</span></span>-<span id="cb1158-5"><a href="#cb1158-5" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> YOURACCESSKEY</span></span>-<span id="cb1158-6"><a href="#cb1158-6" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> YOURSECRETACCESSKEY</span></span>-<span id="cb1158-7"><a href="#cb1158-7" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span><span class="st"> au-nsw</span></span>-<span id="cb1158-8"><a href="#cb1158-8" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> s3.rackcorp.com</span></span>-<span id="cb1158-9"><a href="#cb1158-9" aria-hidden="true" tabindex="-1"></a><span class="dt">location_constraint </span><span class="ot">=</span><span class="st"> au-nsw</span></span></code></pre></div>+<div class="sourceCode" id="cb1205"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1205-1"><a href="#cb1205-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[RCS3-demo-config]</span></span>+<span id="cb1205-2"><a href="#cb1205-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>+<span id="cb1205-3"><a href="#cb1205-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> RackCorp</span></span>+<span id="cb1205-4"><a href="#cb1205-4" aria-hidden="true" tabindex="-1"></a><span class="dt">env_auth </span><span class="ot">=</span><span class="st"> </span><span class="kw">true</span></span>+<span id="cb1205-5"><a href="#cb1205-5" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> YOURACCESSKEY</span></span>+<span id="cb1205-6"><a href="#cb1205-6" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> YOURSECRETACCESSKEY</span></span>+<span id="cb1205-7"><a href="#cb1205-7" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span><span class="st"> au-nsw</span></span>+<span id="cb1205-8"><a href="#cb1205-8" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> s3.rackcorp.com</span></span>+<span id="cb1205-9"><a href="#cb1205-9" aria-hidden="true" tabindex="-1"></a><span class="dt">location_constraint </span><span class="ot">=</span><span class="st"> au-nsw</span></span></code></pre></div><h3 id="rclone">Rclone Serve S3</h3><p>Rclone can serve any remote over the S3 protocol. For details see the<a href="https://rclone.org/commands/rclone_serve_s3/">rclone serve@@ -36242,18 +38569,133 @@ server like this:</p><pre class="console"><code>rclone serve s3 --auth-key ACCESS_KEY_ID,SECRET_ACCESS_KEY remote:path</code></pre><p>This will be compatible with an rclone remote which is defined likethis:</p>-<div class="sourceCode" id="cb1160"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1160-1"><a href="#cb1160-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[serves3]</span></span>-<span id="cb1160-2"><a href="#cb1160-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>-<span id="cb1160-3"><a href="#cb1160-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Rclone</span></span>-<span id="cb1160-4"><a href="#cb1160-4" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> http://127.0.0.1:8080/</span></span>-<span id="cb1160-5"><a href="#cb1160-5" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> ACCESS_KEY_ID</span></span>-<span id="cb1160-6"><a href="#cb1160-6" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> SECRET_ACCESS_KEY</span></span>-<span id="cb1160-7"><a href="#cb1160-7" aria-hidden="true" tabindex="-1"></a><span class="dt">use_multipart_uploads </span><span class="ot">=</span><span class="st"> </span><span class="kw">false</span></span></code></pre></div>+<div class="sourceCode" id="cb1207"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1207-1"><a href="#cb1207-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[serves3]</span></span>+<span id="cb1207-2"><a href="#cb1207-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>+<span id="cb1207-3"><a href="#cb1207-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Rclone</span></span>+<span id="cb1207-4"><a href="#cb1207-4" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> http://127.0.0.1:8080/</span></span>+<span id="cb1207-5"><a href="#cb1207-5" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> ACCESS_KEY_ID</span></span>+<span id="cb1207-6"><a href="#cb1207-6" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> SECRET_ACCESS_KEY</span></span>+<span id="cb1207-7"><a href="#cb1207-7" aria-hidden="true" tabindex="-1"></a><span class="dt">use_multipart_uploads </span><span class="ot">=</span><span class="st"> </span><span class="kw">false</span></span></code></pre></div><p>Note that setting <code>use_multipart_uploads = false</code> is towork around <ahref="https://rclone.org/commands/rclone_serve_s3/#bugs">a bug</a> whichwill be fixed in due course.</p>+<h3 id="relaix">RelAix.Storage</h3>+<p>RelAix.Storage is S3 compatible object storage from <a+href="https://www.relaix.net/">RelAix Networks</a>, operated in its own+data centres in Aachen, Germany. There are two locations, Aachen-1 and+Aachen-2, each with its own endpoint. Buckets and data stay at the+location of the endpoint you use.</p>+<p>You get an access key and a secret key from RelAix Networks. Then+run:</p>+<pre class="console"><code>rclone config</code></pre>+<p>This will guide you through an interactive setup process.</p>+<pre class="text"><code>No remotes found, make a new one?+n) New remote+s) Set configuration password+q) Quit config+n/s/q> n++Enter name for new remote.+name> my-relaix++Option Storage.+Type of storage to configure.+Choose a number from below, or type in your own value.+[snip]+XX / Amazon S3 Compliant Storage Providers including AWS, ...+ \ (s3)+[snip]+Storage> s3++Option provider.+Choose your S3 provider.+Choose a number from below, or type in your own value.+Press Enter to leave empty.+[snip]+XX / RelAix.Storage+ \ (RelAix)+[snip]+provider> RelAix++Option env_auth.+Get AWS credentials from runtime (environment variables or EC2/ECS meta data if no env vars).+Only applies if access_key_id and secret_access_key is blank.+Choose a number from below, or type in your own boolean value (true or false).+Press Enter for the default (false).+ 1 / Enter AWS credentials in the next step.+ \ (false)+ 2 / Get AWS credentials from the environment (env vars or IAM).+ \ (true)+env_auth>++Option access_key_id.+AWS Access Key ID.+Leave blank for anonymous access or runtime credentials.+Enter a value. Press Enter to leave empty.+access_key_id> ACCESS_KEY++Option secret_access_key.+AWS Secret Access Key (password).+Leave blank for anonymous access or runtime credentials.+Enter a value. Press Enter to leave empty.+secret_access_key> SECRET_KEY++Option endpoint.+Endpoint for S3 API.+Required when using an S3 clone.+Choose a number from below, or type in your own value.+Press Enter to leave empty.+ 1 / Aachen-1, Germany+ \ (s3-gw1.relaix.net)+ 2 / Aachen-2, Germany+ \ (s3-gw2.relaix.net)+endpoint> 1++Option acl.+Canned ACL used when creating buckets and storing or copying objects.+[snip]+acl>++Option bucket_object_lock_enabled.+Enable Object Lock when creating new buckets.+Enter a boolean value (true or false). Press Enter for the default (false).+bucket_object_lock_enabled>++Edit advanced config?+y) Yes+n) No (default)+y/n> n++Configuration complete.+Options:+- type: s3+- provider: RelAix+- access_key_id: ACCESS_KEY+- secret_access_key: SECRET_KEY+- endpoint: s3-gw1.relaix.net+Keep this "my-relaix" remote?+y) Yes this is OK (default)+e) Edit this remote+d) Delete this remote+y/e/d> y</code></pre>+<p>This will leave the config file looking like this.</p>+<div class="sourceCode" id="cb1210"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1210-1"><a href="#cb1210-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[my-relaix]</span></span>+<span id="cb1210-2"><a href="#cb1210-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>+<span id="cb1210-3"><a href="#cb1210-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> RelAix</span></span>+<span id="cb1210-4"><a href="#cb1210-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> ACCESS_KEY</span></span>+<span id="cb1210-5"><a href="#cb1210-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> SECRET_KEY</span></span>+<span id="cb1210-6"><a href="#cb1210-6" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> s3-gw1.relaix.net</span></span></code></pre></div>+<p>Object Lock can only be enabled when a bucket is created. To create+buckets with Object Lock, set+<code>bucket_object_lock_enabled = true</code> before running+<code>rclone mkdir</code>.</p>+<p>To lock objects on upload with <code>object_lock_mode</code>, also+set <code>object_lock_set_after_upload = true</code>. RelAix.Storage+does not yet apply Object Lock headers sent with multipart uploads, so+rclone needs to set the lock in a separate call after the upload.</p><h3 id="scaleway">Scaleway</h3><p><a href="https://www.scaleway.com/object-storage/">Scaleway</a> TheObject Storage platform allows you to store anything from backups, logs@@ -36262,20 +38704,20 @@ Scaleway console or transferred through our API and CLI or using anyS3-compatible tool.</p><p>Scaleway provides an S3 interface which can be configured for usewith rclone like this:</p>-<div class="sourceCode" id="cb1161"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1161-1"><a href="#cb1161-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[scaleway]</span></span>-<span id="cb1161-2"><a href="#cb1161-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>-<span id="cb1161-3"><a href="#cb1161-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Scaleway</span></span>-<span id="cb1161-4"><a href="#cb1161-4" aria-hidden="true" tabindex="-1"></a><span class="dt">env_auth </span><span class="ot">=</span><span class="st"> </span><span class="kw">false</span></span>-<span id="cb1161-5"><a href="#cb1161-5" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> s3.nl-ams.scw.cloud</span></span>-<span id="cb1161-6"><a href="#cb1161-6" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> SCWXXXXXXXXXXXXXX</span></span>-<span id="cb1161-7"><a href="#cb1161-7" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> 1111111-2222-3333-44444-55555555555555</span></span>-<span id="cb1161-8"><a href="#cb1161-8" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span><span class="st"> nl-ams</span></span>-<span id="cb1161-9"><a href="#cb1161-9" aria-hidden="true" tabindex="-1"></a><span class="dt">location_constraint </span><span class="ot">=</span><span class="st"> nl-ams</span></span>-<span id="cb1161-10"><a href="#cb1161-10" aria-hidden="true" tabindex="-1"></a><span class="dt">acl </span><span class="ot">=</span><span class="st"> private</span></span>-<span id="cb1161-11"><a href="#cb1161-11" aria-hidden="true" tabindex="-1"></a><span class="dt">upload_cutoff </span><span class="ot">=</span><span class="st"> 5M</span></span>-<span id="cb1161-12"><a href="#cb1161-12" aria-hidden="true" tabindex="-1"></a><span class="dt">chunk_size </span><span class="ot">=</span><span class="st"> 5M</span></span>-<span id="cb1161-13"><a href="#cb1161-13" aria-hidden="true" tabindex="-1"></a><span class="dt">copy_cutoff </span><span class="ot">=</span><span class="st"> 5M</span></span></code></pre></div>+<div class="sourceCode" id="cb1211"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1211-1"><a href="#cb1211-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[scaleway]</span></span>+<span id="cb1211-2"><a href="#cb1211-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>+<span id="cb1211-3"><a href="#cb1211-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Scaleway</span></span>+<span id="cb1211-4"><a href="#cb1211-4" aria-hidden="true" tabindex="-1"></a><span class="dt">env_auth </span><span class="ot">=</span><span class="st"> </span><span class="kw">false</span></span>+<span id="cb1211-5"><a href="#cb1211-5" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> s3.nl-ams.scw.cloud</span></span>+<span id="cb1211-6"><a href="#cb1211-6" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> SCWXXXXXXXXXXXXXX</span></span>+<span id="cb1211-7"><a href="#cb1211-7" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> 1111111-2222-3333-44444-55555555555555</span></span>+<span id="cb1211-8"><a href="#cb1211-8" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span><span class="st"> nl-ams</span></span>+<span id="cb1211-9"><a href="#cb1211-9" aria-hidden="true" tabindex="-1"></a><span class="dt">location_constraint </span><span class="ot">=</span><span class="st"> nl-ams</span></span>+<span id="cb1211-10"><a href="#cb1211-10" aria-hidden="true" tabindex="-1"></a><span class="dt">acl </span><span class="ot">=</span><span class="st"> private</span></span>+<span id="cb1211-11"><a href="#cb1211-11" aria-hidden="true" tabindex="-1"></a><span class="dt">upload_cutoff </span><span class="ot">=</span><span class="st"> 5M</span></span>+<span id="cb1211-12"><a href="#cb1211-12" aria-hidden="true" tabindex="-1"></a><span class="dt">chunk_size </span><span class="ot">=</span><span class="st"> 5M</span></span>+<span id="cb1211-13"><a href="#cb1211-13" aria-hidden="true" tabindex="-1"></a><span class="dt">copy_cutoff </span><span class="ot">=</span><span class="st"> 5M</span></span></code></pre></div><p><a href="https://www.scaleway.com/en/glacier-cold-storage/">ScalewayGlacier</a> is the low-cost S3 Glacier alternative from Scaleway and itworks the same way as on S3 by accepting the "GLACIER"@@ -36390,13 +38832,13 @@ e) Edit this remoted) Delete this remotey/e/d> y</code></pre><p>This will leave the config file looking like this.</p>-<div class="sourceCode" id="cb1164"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1164-1"><a href="#cb1164-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[scality]</span></span>-<span id="cb1164-2"><a href="#cb1164-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>-<span id="cb1164-3"><a href="#cb1164-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Scality</span></span>-<span id="cb1164-4"><a href="#cb1164-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> S3_ACCESS_KEY</span></span>-<span id="cb1164-5"><a href="#cb1164-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> S3_SECRET_KEY</span></span>-<span id="cb1164-6"><a href="#cb1164-6" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> https://s3.example.com</span></span></code></pre></div>+<div class="sourceCode" id="cb1214"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1214-1"><a href="#cb1214-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[scality]</span></span>+<span id="cb1214-2"><a href="#cb1214-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>+<span id="cb1214-3"><a href="#cb1214-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Scality</span></span>+<span id="cb1214-4"><a href="#cb1214-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> S3_ACCESS_KEY</span></span>+<span id="cb1214-5"><a href="#cb1214-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> S3_SECRET_KEY</span></span>+<span id="cb1214-6"><a href="#cb1214-6" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> https://s3.example.com</span></span></code></pre></div><h3 id="lyve">Seagate Lyve Cloud</h3><p><ahref="https://www.seagate.com/gb/en/services/cloud/storage/">Seagate@@ -36489,13 +38931,13 @@ Press Enter to leave empty.[snip]acl></code></pre><p>And the config file should end up looking like this:</p>-<div class="sourceCode" id="cb1175"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1175-1"><a href="#cb1175-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[remote]</span></span>-<span id="cb1175-2"><a href="#cb1175-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>-<span id="cb1175-3"><a href="#cb1175-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> LyveCloud</span></span>-<span id="cb1175-4"><a href="#cb1175-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> XXX</span></span>-<span id="cb1175-5"><a href="#cb1175-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> YYY</span></span>-<span id="cb1175-6"><a href="#cb1175-6" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> s3.us-east-1.lyvecloud.seagate.com</span></span></code></pre></div>+<div class="sourceCode" id="cb1225"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1225-1"><a href="#cb1225-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[remote]</span></span>+<span id="cb1225-2"><a href="#cb1225-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>+<span id="cb1225-3"><a href="#cb1225-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> LyveCloud</span></span>+<span id="cb1225-4"><a href="#cb1225-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> XXX</span></span>+<span id="cb1225-5"><a href="#cb1225-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> YYY</span></span>+<span id="cb1225-6"><a href="#cb1225-6" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> s3.us-east-1.lyvecloud.seagate.com</span></span></code></pre></div><h3 id="seaweedfs">SeaweedFS</h3><p><a href="https://github.com/chrislusf/seaweedfs/">SeaweedFS</a> is adistributed storage system for blobs, objects, files, and data lake,@@ -36531,13 +38973,13 @@ such:</p>}</code></pre><p>To use rclone with SeaweedFS, above configuration should end up withsomething like this in your config:</p>-<div class="sourceCode" id="cb1177"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1177-1"><a href="#cb1177-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[seaweedfs_s3]</span></span>-<span id="cb1177-2"><a href="#cb1177-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>-<span id="cb1177-3"><a href="#cb1177-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> SeaweedFS</span></span>-<span id="cb1177-4"><a href="#cb1177-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> any</span></span>-<span id="cb1177-5"><a href="#cb1177-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> any</span></span>-<span id="cb1177-6"><a href="#cb1177-6" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> localhost:8333</span></span></code></pre></div>+<div class="sourceCode" id="cb1227"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1227-1"><a href="#cb1227-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[seaweedfs_s3]</span></span>+<span id="cb1227-2"><a href="#cb1227-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>+<span id="cb1227-3"><a href="#cb1227-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> SeaweedFS</span></span>+<span id="cb1227-4"><a href="#cb1227-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> any</span></span>+<span id="cb1227-5"><a href="#cb1227-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> any</span></span>+<span id="cb1227-6"><a href="#cb1227-6" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> localhost:8333</span></span></code></pre></div><p>So once set up, for example to copy files into a bucket</p><pre class="console"><code>rclone copy /path/to/files seaweedfs_s3:foo</code></pre><h3 id="selectel">Selectel</h3>@@ -36640,14 +39082,14 @@ e) Edit this remoted) Delete this remotey/e/d> y</code></pre><p>And your config should end up looking like this:</p>-<div class="sourceCode" id="cb1180"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1180-1"><a href="#cb1180-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[selectel]</span></span>-<span id="cb1180-2"><a href="#cb1180-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>-<span id="cb1180-3"><a href="#cb1180-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Selectel</span></span>-<span id="cb1180-4"><a href="#cb1180-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> ACCESS_KEY</span></span>-<span id="cb1180-5"><a href="#cb1180-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> SECRET_ACCESS_KEY</span></span>-<span id="cb1180-6"><a href="#cb1180-6" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span><span class="st"> ru-1</span></span>-<span id="cb1180-7"><a href="#cb1180-7" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> s3.ru-1.storage.selcloud.ru</span></span></code></pre></div>+<div class="sourceCode" id="cb1230"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1230-1"><a href="#cb1230-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[selectel]</span></span>+<span id="cb1230-2"><a href="#cb1230-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>+<span id="cb1230-3"><a href="#cb1230-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Selectel</span></span>+<span id="cb1230-4"><a href="#cb1230-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> ACCESS_KEY</span></span>+<span id="cb1230-5"><a href="#cb1230-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> SECRET_ACCESS_KEY</span></span>+<span id="cb1230-6"><a href="#cb1230-6" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span><span class="st"> ru-1</span></span>+<span id="cb1230-7"><a href="#cb1230-7" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> s3.ru-1.storage.selcloud.ru</span></span></code></pre></div><h3 id="servercore">Servercore</h3><p><a href="https://servercore.com/services/object-storage/">ServercoreObject Storage</a> is an S3 compatible object storage system that@@ -36840,13 +39282,13 @@ e) Edit this remoted) Delete this remotey/e/d> y</code></pre><p>And your config should end up looking like this:</p>-<div class="sourceCode" id="cb1183"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1183-1"><a href="#cb1183-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[spectratest]</span></span>-<span id="cb1183-2"><a href="#cb1183-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>-<span id="cb1183-3"><a href="#cb1183-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> SpectraLogic</span></span>-<span id="cb1183-4"><a href="#cb1183-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> ACCESS_KEY</span></span>-<span id="cb1183-5"><a href="#cb1183-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> SECRET_ACCESS_KEY</span></span>-<span id="cb1183-6"><a href="#cb1183-6" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> https://bp.example.com</span></span></code></pre></div>+<div class="sourceCode" id="cb1233"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1233-1"><a href="#cb1233-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[spectratest]</span></span>+<span id="cb1233-2"><a href="#cb1233-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>+<span id="cb1233-3"><a href="#cb1233-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> SpectraLogic</span></span>+<span id="cb1233-4"><a href="#cb1233-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> ACCESS_KEY</span></span>+<span id="cb1233-5"><a href="#cb1233-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> SECRET_ACCESS_KEY</span></span>+<span id="cb1233-6"><a href="#cb1233-6" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> https://bp.example.com</span></span></code></pre></div><h3 id="storj">Storj</h3><p>Storj is a decentralized cloud storage which can be used through itsnative protocol or an S3 compatible gateway.</p>@@ -37395,19 +39837,19 @@ e) Edit this remoted) Delete this remotey/e/d> y</code></pre><p>This will leave the config file looking like this.</p>-<div class="sourceCode" id="cb1201"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1201-1"><a href="#cb1201-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[wasabi]</span></span>-<span id="cb1201-2"><a href="#cb1201-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>-<span id="cb1201-3"><a href="#cb1201-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Wasabi</span></span>-<span id="cb1201-4"><a href="#cb1201-4" aria-hidden="true" tabindex="-1"></a><span class="dt">env_auth </span><span class="ot">=</span><span class="st"> </span><span class="kw">false</span></span>-<span id="cb1201-5"><a href="#cb1201-5" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> YOURACCESSKEY</span></span>-<span id="cb1201-6"><a href="#cb1201-6" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> YOURSECRETACCESSKEY</span></span>-<span id="cb1201-7"><a href="#cb1201-7" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span></span>-<span id="cb1201-8"><a href="#cb1201-8" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> s3.wasabisys.com</span></span>-<span id="cb1201-9"><a href="#cb1201-9" aria-hidden="true" tabindex="-1"></a><span class="dt">location_constraint </span><span class="ot">=</span></span>-<span id="cb1201-10"><a href="#cb1201-10" aria-hidden="true" tabindex="-1"></a><span class="dt">acl </span><span class="ot">=</span></span>-<span id="cb1201-11"><a href="#cb1201-11" aria-hidden="true" tabindex="-1"></a><span class="dt">server_side_encryption </span><span class="ot">=</span></span>-<span id="cb1201-12"><a href="#cb1201-12" aria-hidden="true" tabindex="-1"></a><span class="dt">storage_class </span><span class="ot">=</span></span></code></pre></div>+<div class="sourceCode" id="cb1251"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1251-1"><a href="#cb1251-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[wasabi]</span></span>+<span id="cb1251-2"><a href="#cb1251-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>+<span id="cb1251-3"><a href="#cb1251-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Wasabi</span></span>+<span id="cb1251-4"><a href="#cb1251-4" aria-hidden="true" tabindex="-1"></a><span class="dt">env_auth </span><span class="ot">=</span><span class="st"> </span><span class="kw">false</span></span>+<span id="cb1251-5"><a href="#cb1251-5" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> YOURACCESSKEY</span></span>+<span id="cb1251-6"><a href="#cb1251-6" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> YOURSECRETACCESSKEY</span></span>+<span id="cb1251-7"><a href="#cb1251-7" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span></span>+<span id="cb1251-8"><a href="#cb1251-8" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> s3.wasabisys.com</span></span>+<span id="cb1251-9"><a href="#cb1251-9" aria-hidden="true" tabindex="-1"></a><span class="dt">location_constraint </span><span class="ot">=</span></span>+<span id="cb1251-10"><a href="#cb1251-10" aria-hidden="true" tabindex="-1"></a><span class="dt">acl </span><span class="ot">=</span></span>+<span id="cb1251-11"><a href="#cb1251-11" aria-hidden="true" tabindex="-1"></a><span class="dt">server_side_encryption </span><span class="ot">=</span></span>+<span id="cb1251-12"><a href="#cb1251-12" aria-hidden="true" tabindex="-1"></a><span class="dt">storage_class </span><span class="ot">=</span></span></code></pre></div><h3 id="zadara">Zadara Object Storage</h3><p><a href="https://www.zadara.com">Zadara Object Storage</a> is afully-managed, enterprise-grade, S3-compatible storage solution that@@ -37507,13 +39949,13 @@ e) Edit this remoted) Delete this remotey/e/d> y</code></pre><p>This will leave the config file looking like this.</p>-<div class="sourceCode" id="cb1204"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1204-1"><a href="#cb1204-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[Zadara-Object-Storage]</span></span>-<span id="cb1204-2"><a href="#cb1204-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>-<span id="cb1204-3"><a href="#cb1204-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Zadara</span></span>-<span id="cb1204-4"><a href="#cb1204-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> S3_ACCESS_KEY</span></span>-<span id="cb1204-5"><a href="#cb1204-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> S3_SECRET_KEY</span></span>-<span id="cb1204-6"><a href="#cb1204-6" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> https://vsa-00000001-public-zadara-cloud-01.zadarazios.com</span></span></code></pre></div>+<div class="sourceCode" id="cb1254"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1254-1"><a href="#cb1254-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[Zadara-Object-Storage]</span></span>+<span id="cb1254-2"><a href="#cb1254-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>+<span id="cb1254-3"><a href="#cb1254-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Zadara</span></span>+<span id="cb1254-4"><a href="#cb1254-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> S3_ACCESS_KEY</span></span>+<span id="cb1254-5"><a href="#cb1254-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> S3_SECRET_KEY</span></span>+<span id="cb1254-6"><a href="#cb1254-6" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> https://vsa-00000001-public-zadara-cloud-01.zadarazios.com</span></span></code></pre></div><h3 id="Zata">Zata Object Storage</h3><p><a href="https://zata.ai/">Zata Object Storage</a> provides a secure,S3-compatible cloud storage solution designed for scalability and@@ -37649,14 +40091,14 @@ e) Edit this remoted) Delete this remotey/e/d></code></pre><p>This will leave the config file looking like this.</p>-<div class="sourceCode" id="cb1207"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1207-1"><a href="#cb1207-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[my zata storage]</span></span>-<span id="cb1207-2"><a href="#cb1207-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>-<span id="cb1207-3"><a href="#cb1207-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Zata</span></span>-<span id="cb1207-4"><a href="#cb1207-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> xxx</span></span>-<span id="cb1207-5"><a href="#cb1207-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> xxx</span></span>-<span id="cb1207-6"><a href="#cb1207-6" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span><span class="st"> us-east-1</span></span>-<span id="cb1207-7"><a href="#cb1207-7" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> idr01.zata.ai</span></span></code></pre></div>+<div class="sourceCode" id="cb1257"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1257-1"><a href="#cb1257-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[my zata storage]</span></span>+<span id="cb1257-2"><a href="#cb1257-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> s3</span></span>+<span id="cb1257-3"><a href="#cb1257-3" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> Zata</span></span>+<span id="cb1257-4"><a href="#cb1257-4" aria-hidden="true" tabindex="-1"></a><span class="dt">access_key_id </span><span class="ot">=</span><span class="st"> xxx</span></span>+<span id="cb1257-5"><a href="#cb1257-5" aria-hidden="true" tabindex="-1"></a><span class="dt">secret_access_key </span><span class="ot">=</span><span class="st"> xxx</span></span>+<span id="cb1257-6"><a href="#cb1257-6" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span><span class="st"> us-east-1</span></span>+<span id="cb1257-7"><a href="#cb1257-7" aria-hidden="true" tabindex="-1"></a><span class="dt">endpoint </span><span class="ot">=</span><span class="st"> idr01.zata.ai</span></span></code></pre></div><h3 id="zero-z3">Zero Services (ZERO-Z3)</h3><p><a href="https://lp.zeroservices.eu/zero-z3/">Zero Services GmbH</a>offers ZERO-Z3, S3-compatible object storage hosted in the EU on its own@@ -37805,7 +40247,7 @@ archive extract</a></li><p>These commands supports a wider range of non cloud friendly archives(but not squashfs) but can't be used for <code>rclone mount</code> orany other rclone commands (eg <code>rclone check</code>).</p>-<h2 id="configuration-3">Configuration</h2>+<h2 id="configuration-4">Configuration</h2><p>This backend is best used without configuration.</p><p>Use it by putting the string <code>:archive:</code> in front ofanother remote, say <code>remote:dir</code> to make@@ -37851,7 +40293,7 @@ gigihevupaz5zkacak/...</code></pre>-<h2 id="modification-times-2">Modification times</h2>+<h2 id="modification-times-3">Modification times</h2><p>Modification times are preserved with an accuracy depending on thearchive type.</p><pre><code>$ rclone lsl --max-depth 1 :archive:s3:rclone/dir/100files.zip@@ -37928,6 +40370,21 @@ number"</li><p>Rclone works fastest with large squashfs block sizes. Forexample:</p><pre><code>mksquashfs 100files 100files.sqfs -comp zstd -b 1M</code></pre>+<h2 id="symlinks-10">Symlinks</h2>+<p>Symlinks stored in an archive are skipped unless the+<code>-l/--links</code> flag is supplied, as they are on the other+backends. With the flag they are listed with a <code>.rclonelink</code>+suffix and read back as the path they point at, the same convention the+<a href="https://rclone.org/local/#symlinks-junction-points">local</a>+backend uses, so they can be copied back as symlinks:</p>+<pre><code>$ rclone lsl --links :archive:/tmp/test.zip+ 5 2026-09-28 14:16:59.000000000 src/file.txt+ 8 2026-09-28 14:16:59.000000000 src/link.txt.rclonelink+$ rclone cat --links :archive:/tmp/test.zip/src/link.txt.rclonelink+file.txt+$ rclone copy --links :archive:/tmp/test.zip /tmp/out+$ ls -l /tmp/out/src/link.txt+lrwxr-xr-x 1 user user 8 Sep 28 14:16 /tmp/out/src/link.txt -> file.txt</code></pre><h2 id="limitations-7">Limitations</h2><p>Files in the archive backend are read only. It isn't possible tocreate archives with the archive backend yet. However you@@ -38098,7 +40555,7 @@ cloud storage system</a>.</p><p>Paths are specified as <code>remote:bucket</code> (or<code>remote:</code> for the <code>lsd</code> command.) You may putsubdirectories in too, e.g. <code>remote:bucket/path/to/dir</code>.</p>-<h2 id="configuration-4">Configuration</h2>+<h2 id="configuration-5">Configuration</h2><p>Here is an example of making a b2 configuration. First run</p><pre class="console"><code>rclone config</code></pre><p>This will guide you through an interactive setup process. To@@ -38165,7 +40622,7 @@ then B2 will return 401 errors.</p>fewer transactions in exchange for more memory. See the <ahref="https://rclone.org/docs/#fast-list">rclone docs</a> for moredetails.</p>-<h3 id="modification-times-3">Modification times</h3>+<h3 id="modification-times-4">Modification times</h3><p>The modification time is stored as metadata on the object as<code>X-Bz-Info-src_last_modified_millis</code> as milliseconds since1970-01-01 in the Backblaze standard. Other tools should be able to use@@ -38528,6 +40985,10 @@ with private buckets by sending an "Authorization" header. If the customendpoint rewrites the requests for authentication, e.g., in CloudflareWorkers, this header needs to be handled properly. Leave blank if youwant to use the endpoint provided by Backblaze.</p>+<p>Rclone will not follow a redirect from this endpoint which downgrades+HTTPS to HTTP, and will not send the "Authorization" header or the SSE-C+key headers once a redirect has left the host and port of this+endpoint.</p><p>The URL provided here SHOULD have the protocol and SHOULD NOT have atrailing slash or specify the /file/bucket subpath as rclone willrequest files with "{download_url}/file/{bucket_name}/{path}".</p>@@ -38614,7 +41075,7 @@ storing this object in B2.</p><h4 id="b2-sse-customer-key">--b2-sse-customer-key</h4><p>To use SSE-C, you may provide the secret encryption key encoded in aUTF-8 compatible string to encrypt/decrypt your data</p>-<p>Alternatively you can provide --sse-customer-key-base64.</p>+<p>Alternatively you can provide --b2-sse-customer-key-base64.</p><p>Properties:</p><ul><li>Config: sse_customer_key</li>@@ -38632,7 +41093,7 @@ UTF-8 compatible string to encrypt/decrypt your data</p><h4 id="b2-sse-customer-key-base64">--b2-sse-customer-key-base64</h4><p>To use SSE-C, you may provide the secret encryption key encoded inBase64 format to encrypt/decrypt your data</p>-<p>Alternatively you can provide --sse-customer-key.</p>+<p>Alternatively you can provide --b2-sse-customer-key.</p><p>Properties:</p><ul><li>Config: sse_customer_key_base64</li>@@ -38693,15 +41154,15 @@ bucket.</p><p>To show the current lifecycle rules:</p><pre class="console"><code>rclone backend lifecycle b2:bucket</code></pre><p>This will dump something like this showing the lifecycle rules.</p>-<div class="sourceCode" id="cb1240"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb1240-1"><a href="#cb1240-1" aria-hidden="true" tabindex="-1"></a><span class="ot">[</span></span>-<span id="cb1240-2"><a href="#cb1240-2" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>-<span id="cb1240-3"><a href="#cb1240-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"daysFromHidingToDeleting"</span><span class="fu">:</span> <span class="dv">1</span><span class="fu">,</span></span>-<span id="cb1240-4"><a href="#cb1240-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"daysFromUploadingToHiding"</span><span class="fu">:</span> <span class="kw">null</span><span class="fu">,</span></span>-<span id="cb1240-5"><a href="#cb1240-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"daysFromStartingToCancelingUnfinishedLargeFiles"</span><span class="fu">:</span> <span class="kw">null</span><span class="fu">,</span></span>-<span id="cb1240-6"><a href="#cb1240-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">"fileNamePrefix"</span><span class="fu">:</span> <span class="st">""</span></span>-<span id="cb1240-7"><a href="#cb1240-7" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>-<span id="cb1240-8"><a href="#cb1240-8" aria-hidden="true" tabindex="-1"></a><span class="ot">]</span></span></code></pre></div>+<div class="sourceCode" id="cb1291"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb1291-1"><a href="#cb1291-1" aria-hidden="true" tabindex="-1"></a><span class="ot">[</span></span>+<span id="cb1291-2"><a href="#cb1291-2" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>+<span id="cb1291-3"><a href="#cb1291-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"daysFromHidingToDeleting"</span><span class="fu">:</span> <span class="dv">1</span><span class="fu">,</span></span>+<span id="cb1291-4"><a href="#cb1291-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"daysFromUploadingToHiding"</span><span class="fu">:</span> <span class="kw">null</span><span class="fu">,</span></span>+<span id="cb1291-5"><a href="#cb1291-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"daysFromStartingToCancelingUnfinishedLargeFiles"</span><span class="fu">:</span> <span class="kw">null</span><span class="fu">,</span></span>+<span id="cb1291-6"><a href="#cb1291-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">"fileNamePrefix"</span><span class="fu">:</span> <span class="st">""</span></span>+<span id="cb1291-7"><a href="#cb1291-7" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>+<span id="cb1291-8"><a href="#cb1291-8" aria-hidden="true" tabindex="-1"></a><span class="ot">]</span></span></code></pre></div><p>If there are no lifecycle rules (the default) then it will justreturn <code>[]</code>.</p><p>To reset the current lifecycle rules:</p>@@ -38765,7 +41226,7 @@ href="https://rclone.org/commands/rclone_about/">rclone about</a>.</p>can do either in your browser, or with a config.json downloaded from Boxto use JWT authentication. <code>rclone config</code> walks you throughit.</p>-<h2 id="configuration-5">Configuration</h2>+<h2 id="configuration-6">Configuration</h2><p>Here is an example of how to make a remote called<code>remote</code>. First run:</p><pre class="console"><code>rclone config</code></pre>@@ -38853,10 +41314,11 @@ using the same procedure detailed above in the, using the password youhave just set.</p><h3 id="invalid-refresh-token">Invalid refresh token</h3><p>According to the <a-href="https://developer.box.com/v2.0/docs/oauth-20#section-6-using-the-access-and-refresh-tokens">box+href="https://developer.box.com/guides/authentication/tokens/refresh/">boxdocs</a>:</p><blockquote>-<p>Each refresh_token is valid for one use in 60 days.</p>+<p>A Refresh Token is valid for 60 days and can be used to obtain a new+Access Token and Refresh Token only once.</p></blockquote><p>This means that if you</p><ul>@@ -39266,7 +41728,7 @@ find you can't work without it. There are many docs online describingthe use of the cache backend to minimize API hits and by-and-large theseare out of date and the cache backend isn't needed in those scenariosany more.</p>-<h2 id="configuration-6">Configuration</h2>+<h2 id="configuration-7">Configuration</h2><p>To get started you just need to have an existing remote which can beconfigured with <code>cache</code>.</p><p>Here is an example of how to make a remote called@@ -39878,7 +42340,7 @@ href="https://rclone.org/rc/#backend-command">backend/command</a>.</p>into smaller chunks during upload to wrapped remote and transparentlyassembles them back when the file is downloaded. This allows toeffectively overcome size limits imposed by storage providers.</p>-<h2 id="configuration-7">Configuration</h2>+<h2 id="configuration-8">Configuration</h2><p>To use it, first set up the underlying remote following theconfiguration instructions for that remote. You can also use a localpathname instead of a remote.</p>@@ -40090,7 +42552,7 @@ secondary type. This will save CPU and bandwidth but can result in emptyhashsums at destination. Beware of consequences: the <code>sync</code>command will revert (sometimes silently) to time/size comparison ifcompatible hashsums between source and target are not found.</p>-<h3 id="modification-times-4">Modification times</h3>+<h3 id="modification-times-5">Modification times</h3><p>Chunker stores modification times using the wrapped remote so supportdepends on that. For a small non-chunked file the chunker overlay simplymanipulates modification time of the wrapped remote file. For a@@ -40366,7 +42828,7 @@ pricing details</a>.</p><h2 id="securing-your-credentials">Securing Your Credentials</h2><p>Please refer to the <ahref="https://rclone.org/docs/#configuration-encryption-cheatsheet">docs</a></p>-<h2 id="configuration-8">Configuration</h2>+<h2 id="configuration-9">Configuration</h2><p>Here is an example of making a Cloudinary configuration.</p><p>First, create a <ahref="https://cloudinary.com/users/register_free">cloudinary.com</a>@@ -40561,7 +43023,7 @@ ts u3ma usdz wdp webm webp wmv]</li><h1 id="citrix-sharefile">Citrix ShareFile</h1><p><a href="https://sharefile.com">Citrix ShareFile</a> is a secure filesharing and transfer service aimed as business.</p>-<h2 id="configuration-9">Configuration</h2>+<h2 id="configuration-10">Configuration</h2><p>The initial setup for Citrix ShareFile involves getting a token fromCitrix ShareFile which you can in your browser.<code>rclone config</code> walks you through it.</p>@@ -40970,7 +43432,7 @@ SecretBox</a>, based on XSalsa20 cipher and Poly1305 for integrity. <ahref="#name-encryption">Names</a> (file- and directory names) are alsoencrypted by default, but this has some implications and is thereforepossible to be turned off.</p>-<h2 id="configuration-10">Configuration</h2>+<h2 id="configuration-11">Configuration</h2><p>Here is an example of how to make a remote called<code>secret</code>.</p><p>To use <code>crypt</code>, first set up the underlying remote. Follow@@ -41620,6 +44082,14 @@ on the cloud storage system.</p><li>filenames with the same name will encrypt the same</li><li>filenames which start the same won't have a common prefix</li></ul>+<p>A version string of the form <code>-vYYYY-MM-DD-HHMMSS-NNN</code> on+the end of a file name (as added by <code>--b2-versions</code> /+<code>--s3-versions</code>) is left in plain text so that versioned+files can be found. Directory names are encrypted in full. Rclone before+v1.76 left such a suffix in plain text on directory names too, so a+directory named like this created by an older rclone will appear in+listings with a warning but can't be opened or removed until renamed on+the underlying remote to the name given in the warning.</p><p>This uses a 32 byte key (256 bits) and a 16 byte (128 bits) IV bothof which are derived from the user password.</p><p>After encryption they are written out using a modified version of@@ -41640,7 +44110,7 @@ If the user doesn't supply a salt then rclone uses an internal one.</p><p><code>scrypt</code> makes it impractical to mount a dictionary attackon rclone encrypted data. For full protection against this you shouldalways use a salt.</p>-<h2 id="see-also-99">See Also</h2>+<h2 id="see-also-100">See Also</h2><ul><li><a href="https://rclone.org/commands/rclone_cryptdecode/">rclonecryptdecode</a> - Show forward/reverse mapping of encrypted@@ -41655,7 +44125,7 @@ code and don't use this remote in critical applications.</p><p>The <code>Compress</code> remote adds compression to another remote.It is best used with remotes containing many large compressiblefiles.</p>-<h2 id="configuration-11">Configuration</h2>+<h2 id="configuration-12">Configuration</h2><p>To use this remote, all you need to do is specify another remote anda compression mode to use:</p><pre class="text"><code>$ rclone config@@ -41875,7 +44345,7 @@ the config like this</p><p>During the initial setup with <code>rclone config</code> you willspecify the upstreams remotes as a space separated list. The upstreamremotes can either be a local paths or other remotes.</p>-<h2 id="configuration-12">Configuration</h2>+<h2 id="configuration-13">Configuration</h2><p>Here is an example of how to make a combine called<code>remote</code> for the example above. First run:</p><pre class="console"><code>rclone config</code></pre>@@ -41921,18 +44391,18 @@ the shared drives you have access to.</p><code>drive:</code> you would run</p><pre class="console"><code>rclone backend -o config drives drive:</code></pre><p>This would produce something like this:</p>-<div class="sourceCode" id="cb1292"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1292-1"><a href="#cb1292-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[My Drive]</span></span>-<span id="cb1292-2"><a href="#cb1292-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> alias</span></span>-<span id="cb1292-3"><a href="#cb1292-3" aria-hidden="true" tabindex="-1"></a><span class="dt">remote </span><span class="ot">=</span><span class="st"> drive,team_drive=0ABCDEF-01234567890,root_folder_id=:</span></span>-<span id="cb1292-4"><a href="#cb1292-4" aria-hidden="true" tabindex="-1"></a></span>-<span id="cb1292-5"><a href="#cb1292-5" aria-hidden="true" tabindex="-1"></a><span class="kw">[Test Drive]</span></span>-<span id="cb1292-6"><a href="#cb1292-6" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> alias</span></span>-<span id="cb1292-7"><a href="#cb1292-7" aria-hidden="true" tabindex="-1"></a><span class="dt">remote </span><span class="ot">=</span><span class="st"> drive,team_drive=0ABCDEFabcdefghijkl,root_folder_id=:</span></span>-<span id="cb1292-8"><a href="#cb1292-8" aria-hidden="true" tabindex="-1"></a></span>-<span id="cb1292-9"><a href="#cb1292-9" aria-hidden="true" tabindex="-1"></a><span class="kw">[AllDrives]</span></span>-<span id="cb1292-10"><a href="#cb1292-10" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> combine</span></span>-<span id="cb1292-11"><a href="#cb1292-11" aria-hidden="true" tabindex="-1"></a><span class="dt">upstreams </span><span class="ot">=</span><span class="st"> "My Drive=My Drive:" "Test Drive=Test Drive:"</span></span></code></pre></div>+<div class="sourceCode" id="cb1343"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1343-1"><a href="#cb1343-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[My Drive]</span></span>+<span id="cb1343-2"><a href="#cb1343-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> alias</span></span>+<span id="cb1343-3"><a href="#cb1343-3" aria-hidden="true" tabindex="-1"></a><span class="dt">remote </span><span class="ot">=</span><span class="st"> drive,team_drive=0ABCDEF-01234567890,root_folder_id=:</span></span>+<span id="cb1343-4"><a href="#cb1343-4" aria-hidden="true" tabindex="-1"></a></span>+<span id="cb1343-5"><a href="#cb1343-5" aria-hidden="true" tabindex="-1"></a><span class="kw">[Test Drive]</span></span>+<span id="cb1343-6"><a href="#cb1343-6" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> alias</span></span>+<span id="cb1343-7"><a href="#cb1343-7" aria-hidden="true" tabindex="-1"></a><span class="dt">remote </span><span class="ot">=</span><span class="st"> drive,team_drive=0ABCDEFabcdefghijkl,root_folder_id=:</span></span>+<span id="cb1343-8"><a href="#cb1343-8" aria-hidden="true" tabindex="-1"></a></span>+<span id="cb1343-9"><a href="#cb1343-9" aria-hidden="true" tabindex="-1"></a><span class="kw">[AllDrives]</span></span>+<span id="cb1343-10"><a href="#cb1343-10" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> combine</span></span>+<span id="cb1343-11"><a href="#cb1343-11" aria-hidden="true" tabindex="-1"></a><span class="dt">upstreams </span><span class="ot">=</span><span class="st"> "My Drive=My Drive:" "Test Drive=Test Drive:"</span></span></code></pre></div><p>If you then add that config to your config file (find it with<code>rclone config file</code>) then you can access all the shareddrives in one place with the <code>AllDrives:</code> remote.</p>@@ -41998,7 +44468,7 @@ repositories</a></li><p>Paths are specified as <code>remote:path</code></p><p>Paths may be as deep as required, e.g.<code>remote:directory/subdirectory</code>.</p>-<h2 id="configuration-13">Configuration</h2>+<h2 id="configuration-14">Configuration</h2><p>Here is an example of how to make a remote called<code>remote</code>. First run:</p><pre class="console"><code>rclone config</code></pre>@@ -42133,6 +44603,255 @@ with the new parameters. Only new parameters need be passed as thevalues will default to those currently in use.</p><p>It doesn't return anything.</p><!-- autogenerated options stop -->+<h1 id="dosya.dev">dosya.dev</h1>+<p><a href="https://dosya.dev">dosya.dev</a> is a cloud file storage,+sharing, and synchronization platform with workspace-based+multi-tenancy, file versioning, and team collaboration features.</p>+<h2 id="configuration-15">Configuration</h2>+<p>Here is an example of how to make a remote called+<code>remote</code>. First run:</p>+<pre class="console"><code>rclone config</code></pre>+<p>This will guide you through an interactive setup process:</p>+<pre class="text"><code>No remotes found, make a new one?+n) New remote+s) Set configuration password+q) Quit config+n/s/q> n++Enter name for new remote.+name> remote++Option Storage.+Type of storage to configure.+Choose a number from below, or type in your own value.+XX / dosya.dev+ \ (dosya)+Storage> dosya++Option api_key.+Your API Key, get it from https://dosya.dev/profile under "API keys".+Enter a value.+api_key> dos_YOURAPIKEY++Option workspace_id.+Your workspace ID.++Leave this blank to use the one workspace the API key can reach, which+is what a key restricted to a single workspace does. If the key can+reach several, rclone will refuse and list their IDs for you to choose+from.+Enter a value. Press Enter to leave empty.+workspace_id>++Edit advanced config?+y) Yes+n) No (default)+y/n> n++Configuration complete.+Options:+- type: dosya+- api_key: dos_YOURAPIKEY+Keep this "remote" remote?+y) Yes this is OK (default)+e) Edit this remote+d) Delete this remote+y/e/d> y</code></pre>+<p>Once configured you can then use <code>rclone</code> like this+(replace <code>remote</code> with the name you gave your remote):</p>+<p>List directories and files in the top level of your dosya+workspace</p>+<pre class="console"><code>rclone lsf remote:</code></pre>+<p>To copy a local directory to a dosya directory called backup</p>+<pre class="console"><code>rclone copy /home/source remote:backup</code></pre>+<p>The dosya backend uploads large files in parts and stores them as+single files that remain accessible through the web app and other+clients. If you previously used a WebDAV remote with a chunker overlay+for large uploads, you can use this backend without the overlay.</p>+<h3 id="getting-your-api-key">Getting your API key</h3>+<ol type="1">+<li>Log in to <a href="https://dosya.dev">dosya.dev</a></li>+<li>Go to <a href="https://dosya.dev/profile">Profile</a> >+<strong>API keys</strong></li>+<li>Create a new API key</li>+<li>Copy the key (starts with <code>dos_</code>)</li>+</ol>+<p>A key can be restricted to a single workspace when you create it,+which is the simplest thing to do for rclone: <code>workspace_id</code>+can then be left out of the config entirely.</p>+<h3 id="workspaces">Workspaces</h3>+<p>Leave <code>workspace_id</code> blank and rclone asks which+workspaces the key can reach. A key restricted to one workspace, and an+account that has only one, both resolve without any configuration. If+the key can reach several, rclone refuses and lists their IDs so you can+set the one you want:</p>+<pre class="text"><code>this API key can reach 2 workspaces, so workspace_id must be set to one of:+ ws_AAAAAAAA (Work)+ ws_BBBBBBBB (Personal)</code></pre>+<p>To access several workspaces at once, create a separate remote for+each:</p>+<pre class="console"><code>rclone config create work dosya api_key=dos_XXX workspace_id=ws_WORK+rclone config create personal dosya api_key=dos_XXX workspace_id=ws_PERSONAL</code></pre>+<p>Then use them independently:</p>+<pre class="console"><code>rclone ls work:+rclone copy work:docs/ personal:backup/</code></pre>+<h3 id="modification-times-and-hashes-5">Modification times and+hashes</h3>+<p>dosya.dev supports modification times with a resolution of 1 second.+The modification time is sent with the upload, so a copied or synced+file keeps its original timestamp. Modification times cannot be changed+after upload without re-uploading the file.</p>+<p>dosya.dev supports SHA-256 checksums. The server computes the hash of+a single-shot (small file) upload from the bytes it receives, so+<code>rclone check</code> and hash-based sync work without downloading.+Large files uploaded in multiple parts have no server-side hash (the+parts are never re-read to hash the whole object), the same limitation+the S3 backend has for multipart objects; those objects report an empty+hash.</p>+<h3 id="restricted-filename-characters-5">Restricted filename+characters</h3>+<p>In addition to the <a+href="https://rclone.org/overview/#restricted-characters">default+restricted characters set</a> the following characters are also+replaced:</p>+<table>+<thead>+<tr class="header">+<th>Character</th>+<th style="text-align: center;">Value</th>+<th style="text-align: center;">Replacement</th>+</tr>+</thead>+<tbody>+<tr class="odd">+<td>"</td>+<td style="text-align: center;">0x22</td>+<td style="text-align: center;">"</td>+</tr>+<tr class="even">+<td>'</td>+<td style="text-align: center;">0x27</td>+<td style="text-align: center;">'</td>+</tr>+<tr class="odd">+<td><</td>+<td style="text-align: center;">0x3C</td>+<td style="text-align: center;"><</td>+</tr>+<tr class="even">+<td>></td>+<td style="text-align: center;">0x3E</td>+<td style="text-align: center;">></td>+</tr>+<tr class="odd">+<td>\</td>+<td style="text-align: center;">0x5C</td>+<td style="text-align: center;">\</td>+</tr>+</tbody>+</table>+<p>Invalid UTF-8 bytes will also be <a+href="https://rclone.org/overview/#invalid-utf8">replaced</a>, as they+can't be used in JSON strings.</p>+<h3 id="public-links">Public Links</h3>+<p>dosya.dev supports <code>rclone link</code> to create public share+links to files. You can use the <code>--expire</code> flag to specify+how long the link should be valid.</p>+<pre class="console"><code>rclone link remote:path/to/file.txt+rclone link --expire 7d remote:path/to/file.txt</code></pre>+<!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/dosya/dosya.go and run make backenddocs to verify -->+<!-- markdownlint-disable-line line-length -->+<h3 id="standard-options-14">Standard options</h3>+<p>Here are the Standard options specific to dosya (dosya.dev).</p>+<h4 id="dosya-api-key">--dosya-api-key</h4>+<p>Your API Key, get it from https://dosya.dev/profile under "API+keys".</p>+<p>Properties:</p>+<ul>+<li>Config: api_key</li>+<li>Env Var: RCLONE_DOSYA_API_KEY</li>+<li>Type: string</li>+<li>Required: true</li>+</ul>+<h4 id="dosya-workspace-id">--dosya-workspace-id</h4>+<p>Your workspace ID.</p>+<p>Leave this blank to use the one workspace the API key can reach,+which is what a key restricted to a single workspace does. If the key+can reach several, rclone will refuse and list their IDs for you to+choose from.</p>+<p>Properties:</p>+<ul>+<li>Config: workspace_id</li>+<li>Env Var: RCLONE_DOSYA_WORKSPACE_ID</li>+<li>Type: string</li>+<li>Required: false</li>+</ul>+<h3 id="advanced-options-14">Advanced options</h3>+<p>Here are the Advanced options specific to dosya (dosya.dev).</p>+<h4 id="dosya-api-url">--dosya-api-url</h4>+<p>API base URL.</p>+<p>Properties:</p>+<ul>+<li>Config: api_url</li>+<li>Env Var: RCLONE_DOSYA_API_URL</li>+<li>Type: string</li>+<li>Default: "https://api.dosya.dev"</li>+</ul>+<h4 id="dosya-upload-concurrency">--dosya-upload-concurrency</h4>+<p>Concurrency for multipart uploads.</p>+<p>This is the number of parts of the same file that are uploaded+concurrently. The first part is always uploaded on its own, because the+server creates the upload when it receives it.</p>+<p>Each part in flight is held in memory, so memory use grows with this+value times the part size the server chooses.</p>+<p>If you are uploading large files over high-speed links and these+uploads do not fully utilize your bandwidth, then increasing this may+help to speed up the transfers.</p>+<p>Properties:</p>+<ul>+<li>Config: upload_concurrency</li>+<li>Env Var: RCLONE_DOSYA_UPLOAD_CONCURRENCY</li>+<li>Type: int</li>+<li>Default: 4</li>+</ul>+<h4 id="dosya-encoding">--dosya-encoding</h4>+<p>The encoding for the backend.</p>+<p>See the <a href="https://rclone.org/overview/#encoding">encoding+section in the overview</a> for more info.</p>+<p>Properties:</p>+<ul>+<li>Config: encoding</li>+<li>Env Var: RCLONE_DOSYA_ENCODING</li>+<li>Type: Encoding</li>+<li>Default:+Slash,LtGt,DoubleQuote,SingleQuote,BackSlash,Del,Ctl,LeftSpace,RightSpace,InvalidUtf8,Dot</li>+</ul>+<h4 id="dosya-description">--dosya-description</h4>+<p>Description of the remote.</p>+<p>Properties:</p>+<ul>+<li>Config: description</li>+<li>Env Var: RCLONE_DOSYA_DESCRIPTION</li>+<li>Type: string</li>+<li>Required: false</li>+</ul>+<!-- autogenerated options stop -->+<h2 id="limitations-11">Limitations</h2>+<p>dosya.dev only supports filenames up to 255 characters in length.</p>+<p>Empty files (0 bytes) cannot be uploaded to dosya.dev.</p>+<p>Upload requires knowing the file size in advance, so streaming+uploads of unknown-size files are not supported.</p>+<p>Each workspace limits the number of uploads a user can have in flight+at once (5 by default, configurable in the workspace settings). When+<code>--transfers</code> is set higher than that limit, rclone waits for+a free upload slot rather than failing the transfer, retrying every 5+seconds for up to 10 minutes.</p>+<h3 id="deleting-files-and-directories">Deleting files and+directories</h3>+<p>rclone deletes files and directories permanently, in a single request+(<code>DELETE ?permanent=1</code>), bypassing the workspace trash.+Removing a directory also asks the server whether it is empty+(<code>?if_empty=1</code>) rather than listing it first.</p><h1 id="drime">Drime</h1><p><a href="https://drime.cloud/">Drime</a> is a cloud storage andtransfer service focused on fast, resilient file delivery. It offers@@ -42141,7 +44860,7 @@ sharing.</p><p>To setup Drime you need to log in, navigate to Settings, Developer,and create a token to use as an API access key. Give it a sensible nameand copy the token for use in the config.</p>-<h2 id="configuration-14">Configuration</h2>+<h2 id="configuration-16">Configuration</h2><p>Here is a run through of <code>rclone config</code> to make a remotecalled <code>remote</code>.</p><p>Firstly run:</p>@@ -42189,7 +44908,7 @@ y/e/d> y</code></pre><pre class="console"><code>rclone lsf remote:</code></pre><p>To copy a local directory to a Drime directory called backup</p><pre class="console"><code>rclone copy /home/source remote:backup</code></pre>-<h3 id="modification-times-and-hashes-5">Modification times and+<h3 id="modification-times-and-hashes-6">Modification times andhashes</h3><p>Drime does not support modification times or hashes.</p><p>This means that by default syncs will only use the size of the file@@ -42198,7 +44917,7 @@ to determine if it needs updating.</p>the object was uploaded. For many operations this is sufficient todetermine if it has changed. However files created with timestamps inthe past will be missed by the sync if using <code>--update</code>.</p>-<h3 id="restricted-filename-characters-5">Restricted filename+<h3 id="restricted-filename-characters-6">Restricted filenamecharacters</h3><p>In addition to the <ahref="https://rclone.org/overview/#restricted-characters">default@@ -42263,7 +44982,7 @@ set</p><p>To restrict rclone to the <code>Files</code> directory.</p><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/drime/drime.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-14">Standard options</h3>+<h3 id="standard-options-15">Standard options</h3><p>Here are the Standard options specific to drime (Drime).</p><h4 id="drime-access-token">--drime-access-token</h4><p>API Access token</p>@@ -42275,7 +44994,7 @@ set</p><li>Type: string</li><li>Required: false</li></ul>-<h3 id="advanced-options-14">Advanced options</h3>+<h3 id="advanced-options-15">Advanced options</h3><p>Here are the Advanced options specific to drime (Drime).</p><h4 id="drime-root-folder-id">--drime-root-folder-id</h4><p>ID of the root folder</p>@@ -42390,14 +45109,14 @@ Slash,BackSlash,Del,Ctl,LeftSpace,RightSpace,InvalidUtf8,Dot</li><li>Required: false</li></ul><!-- autogenerated options stop -->-<h2 id="limitations-11">Limitations</h2>+<h2 id="limitations-12">Limitations</h2><p>Drime only supports filenames up to 255 bytes in length, wherefilenames are encoded in UTF8.</p><h1 id="dropbox">Dropbox</h1><p>Paths are specified as <code>remote:path</code></p><p>Dropbox paths may be as deep as required, e.g.<code>remote:directory/subdirectory</code>.</p>-<h2 id="configuration-15">Configuration</h2>+<h2 id="configuration-17">Configuration</h2><p>The initial setup for dropbox involves getting a token from Dropboxwhich you need to do in your browser. <code>rclone config</code> walksyou through it.</p>@@ -42465,7 +45184,7 @@ and show you all Team Folders and your User Folder.</p><p>A leading <code>/</code> for a Dropbox personal account will donothing, but it will take an extra HTTP transaction so it should beavoided.</p>-<h3 id="modification-times-and-hashes-6">Modification times and+<h3 id="modification-times-and-hashes-7">Modification times andhashes</h3><p>Dropbox supports modified times, but the only way to set amodification time is to re-upload the file.</p>@@ -42477,7 +45196,7 @@ don't want this to happen use <code>--size-only</code> or<p>Dropbox supports <ahref="https://www.dropbox.com/developers/reference/content-hash">its ownhash type</a> which is checked for all transfers.</p>-<h3 id="restricted-filename-characters-6">Restricted filename+<h3 id="restricted-filename-characters-7">Restricted filenamecharacters</h3><table><thead>@@ -42641,7 +45360,7 @@ supported formats at any time.</p></table><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/dropbox/dropbox.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-15">Standard options</h3>+<h3 id="standard-options-16">Standard options</h3><p>Here are the Standard options specific to dropbox (Dropbox).</p><h4 id="dropbox-client-id">--dropbox-client-id</h4><p>OAuth Client Id.</p>@@ -42663,7 +45382,7 @@ supported formats at any time.</p><li>Type: string</li><li>Required: false</li></ul>-<h3 id="advanced-options-15">Advanced options</h3>+<h3 id="advanced-options-16">Advanced options</h3><p>Here are the Advanced options specific to dropbox (Dropbox).</p><h4 id="dropbox-token">--dropbox-token</h4><p>OAuth Access Token as a JSON blob.</p>@@ -42974,7 +45693,7 @@ default based on the batch_mode in use.</p><li>Required: false</li></ul><!-- autogenerated options stop -->-<h2 id="limitations-12">Limitations</h2>+<h2 id="limitations-13">Limitations</h2><p>Note that Dropbox is case insensitive so you can't have a file called"Hello.doc" and one called "hello.doc".</p><p>There are some file names such as <code>thumbs.db</code> which@@ -43048,7 +45767,7 @@ href="https://storagemadeeasy.com/about/">Storage Made Easy's EnterpriseFile Fabric™</a> which provides a software solution to integrate andunify File and Object Storage accessible through a global filesystem.</p>-<h2 id="configuration-16">Configuration</h2>+<h2 id="configuration-18">Configuration</h2><p>The initial setup for the Enterprise File Fabric backend involvesgetting a token from the Enterprise File Fabric which you need to do inyour browser. <code>rclone config</code> walks you through it.</p>@@ -43126,14 +45845,14 @@ y/e/d> y</code></pre><p>To copy a local directory to an Enterprise File Fabric directorycalled backup</p><pre class="console"><code>rclone copy /home/source remote:backup</code></pre>-<h3 id="modification-times-and-hashes-7">Modification times and+<h3 id="modification-times-and-hashes-8">Modification times andhashes</h3><p>The Enterprise File Fabric allows modification times to be set onfiles accurate to 1 second. These will be used to detect whether objectsneed syncing or not.</p><p>The Enterprise File Fabric does not support any data hashes at thistime.</p>-<h3 id="restricted-filename-characters-7">Restricted filename+<h3 id="restricted-filename-characters-8">Restricted filenamecharacters</h3><p>The <ahref="https://rclone.org/overview/#restricted-characters">default@@ -43168,7 +45887,7 @@ for example</p><p>The ID for "S3 Storage" would be <code>120673761</code>.</p><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/filefabric/filefabric.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-16">Standard options</h3>+<h3 id="standard-options-17">Standard options</h3><p>Here are the Standard options specific to filefabric (Enterprise FileFabric).</p><h4 id="filefabric-url">--filefabric-url</h4>@@ -43222,7 +45941,7 @@ https://docs.storagemadeeasy.com/organisationcloud/api-tokens</p><li>Type: string</li><li>Required: false</li></ul>-<h3 id="advanced-options-16">Advanced options</h3>+<h3 id="advanced-options-17">Advanced options</h3><p>Here are the Advanced options specific to filefabric (Enterprise FileFabric).</p><h4 id="filefabric-token">--filefabric-token</h4>@@ -43288,7 +46007,7 @@ upload and back up files from any internet-connected device.</p><p><strong>Note</strong> FileLu now has a fully featured S3 backend <ahref="/s3#filelu-s5">FileLu S5</a>, an industry standard S3 compatibleobject store.</p>-<h2 id="configuration-17">Configuration</h2>+<h2 id="configuration-19">Configuration</h2><p>Here is an example of how to make a remote called<code>filelu</code>. First, run:</p><pre class="console"><code>rclone config</code></pre>@@ -43366,7 +46085,7 @@ two or three folders named "test_folders," the system may becomeconfused and won't know which folder to move. In large storage systems,some clients have hundred of thousands of folders and a few millions offiles, duplicate folder names or paths are quite common.</p>-<h3 id="modification-times-and-hashes-8">Modification Times and+<h3 id="modification-times-and-hashes-9">Modification Times andHashes</h3><p>FileLu supports both modification times and MD5 hashes.</p><p>FileLu only supports filenames and folder names up to 255 characters@@ -43394,7 +46113,7 @@ running <code>rclone</code> has sufficient memory and CPU to handlethese operations.</p><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/filelu/filelu.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-17">Standard options</h3>+<h3 id="standard-options-18">Standard options</h3><p>Here are the Standard options specific to filelu (FileLu CloudStorage).</p><h4 id="filelu-key">--filelu-key</h4>@@ -43406,7 +46125,7 @@ Storage).</p><li>Type: string</li><li>Required: true</li></ul>-<h3 id="advanced-options-17">Advanced options</h3>+<h3 id="advanced-options-18">Advanced options</h3><p>Here are the Advanced options specific to filelu (FileLu CloudStorage).</p><h4 id="filelu-upload-cutoff">--filelu-upload-cutoff</h4>@@ -43450,7 +46169,7 @@ Slash,LtGt,DoubleQuote,SingleQuote,BackQuote,Dollar,Colon,Question,Asterisk,Pipe<li>Required: false</li></ul><!-- autogenerated options stop -->-<h2 id="limitations-13">Limitations</h2>+<h2 id="limitations-14">Limitations</h2><p>This backend uses a custom library implementing the FileLu API. Whileit supports file transfers, some advanced features may not yet beavailable. Please report any issues to the <a@@ -43459,7 +46178,7 @@ and updates.</p><p>For further information, visit <a href="https://filelu.com/">FileLu'swebsite</a>.</p><h1 id="filen">Filen</h1>-<h2 id="configuration-18">Configuration</h2>+<h2 id="configuration-20">Configuration</h2><p>The initial setup for Filen requires that you get an API key for youraccount, currently this is only possible using the <ahref="https://github.com/FilenCloudDienste/filen-cli">Filen CLI</a>.@@ -43534,19 +46253,19 @@ y) Yes this is OK (default)e) Edit this remoted) Delete this remotey/e/d> y</code></pre>-<h3 id="modification-times-and-hashes-9">Modification times and+<h3 id="modification-times-and-hashes-10">Modification times andhashes</h3><p>Modification times are fully supported for files, for directories,only the creation time matters.</p><p>Filen supports Blake3 hashes.</p>-<h3 id="restricted-filename-characters-8">Restricted filename+<h3 id="restricted-filename-characters-9">Restricted filenamecharacters</h3><p>Invalid UTF-8 bytes will be <ahref="https://rclone.org/overview/#invalid-utf8">replaced</a></p><h3 id="api-key">API Key</h3><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/filen/filen.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-18">Standard options</h3>+<h3 id="standard-options-19">Standard options</h3><p>Here are the Standard options specific to filen (Filen).</p><h4 id="filen-email">--filen-email</h4><p>Email of your Filen account</p>@@ -43583,7 +46302,7 @@ obscure</a>.</p><li>Type: string</li><li>Required: true</li></ul>-<h3 id="advanced-options-18">Advanced options</h3>+<h3 id="advanced-options-19">Advanced options</h3><p>Here are the Advanced options specific to filen (Filen).</p><h4 id="filen-upload-concurrency">--filen-upload-concurrency</h4><p>Concurrency for chunked uploads.</p>@@ -43676,7 +46395,7 @@ username, and password. Alternatively, you can authenticate using an APIKey from <ahref="https://www.files.com/docs/sdk-and-apis/api-keys/">Files.com</a>.<code>rclone config</code> walks you through it.</p>-<h2 id="configuration-19">Configuration</h2>+<h2 id="configuration-21">Configuration</h2><p>Here is an example of how to make a remote called<code>remote</code>. First run:</p><pre class="console"><code>rclone config</code></pre>@@ -43747,6 +46466,75 @@ y/e/d> y</code></pre><p>Sync <code>/home/local/directory</code> to the remote directory,deleting any excess files in the directory.</p><pre class="console"><code>rclone sync --interactive /home/local/directory remote:dir</code></pre>+<h3 id="restricted-filename-characters-10">Restricted filename+characters</h3>+<p>In addition to the <a+href="https://rclone.org/overview/#restricted-characters">default+restricted characters set</a> the following characters are also+replaced:</p>+<table>+<thead>+<tr class="header">+<th>Character</th>+<th style="text-align: center;">Value</th>+<th style="text-align: center;">Replacement</th>+</tr>+</thead>+<tbody>+<tr class="odd">+<td>\</td>+<td style="text-align: center;">0x5C</td>+<td style="text-align: center;">\</td>+</tr>+</tbody>+</table>+<p>File names can also not end with the following characters. These only+get replaced if they are the last character in the name:</p>+<table>+<thead>+<tr class="header">+<th>Character</th>+<th style="text-align: center;">Value</th>+<th style="text-align: center;">Replacement</th>+</tr>+</thead>+<tbody>+<tr class="odd">+<td>SP</td>+<td style="text-align: center;">0x20</td>+<td style="text-align: center;">␠</td>+</tr>+<tr class="even">+<td>LF</td>+<td style="text-align: center;">0x0A</td>+<td style="text-align: center;">␊</td>+</tr>+<tr class="odd">+<td>CR</td>+<td style="text-align: center;">0x0D</td>+<td style="text-align: center;">␍</td>+</tr>+<tr class="even">+<td>HT</td>+<td style="text-align: center;">0x09</td>+<td style="text-align: center;">␉</td>+</tr>+<tr class="odd">+<td>VT</td>+<td style="text-align: center;">0x0B</td>+<td style="text-align: center;">␋</td>+</tr>+</tbody>+</table>+<p>Invalid UTF-8 bytes will also be <a+href="https://rclone.org/overview/#invalid-utf8">replaced</a>.</p>+<p>Files.com treats the replacements <code>/</code> and <code>\</code>+as if they were <code>/</code> and <code>\</code>, and the names+<code>.</code> and <code>..</code> as if they were <code>.</code> and+<code>..</code>, and rejects them with <code>Invalid path</code>. This+means that files with <code>/</code> or <code>\</code> in their names,+or which are called <code>.</code> or <code>..</code>, can't be stored+on Files.com.</p><h3 id="hashes-2">Hashes</h3><p>In December 2024 <code>files.com</code> started <ahref="https://www.files.com/blog/2024/11/01/new-modern-checksum-options-now-available-with-opt">supporting@@ -43761,7 +46549,7 @@ selecting more checksums will not affect rclone's operations.</p>can do an end to end integrity check.</p><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/filescom/filescom.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-19">Standard options</h3>+<h3 id="standard-options-20">Standard options</h3><p>Here are the Standard options specific to filescom (Files.com).</p><h4 id="filescom-site">--filescom-site</h4><p>Your site subdomain (e.g. mysite) or custom domain (e.g.@@ -43794,7 +46582,7 @@ obscure</a>.</p><li>Type: string</li><li>Required: false</li></ul>-<h3 id="advanced-options-19">Advanced options</h3>+<h3 id="advanced-options-20">Advanced options</h3><p>Here are the Advanced options specific to filescom (Files.com).</p><h4 id="filescom-api-key">--filescom-api-key</h4><p>The API key used to authenticate with Files.com.</p>@@ -43837,7 +46625,7 @@ package.</p>begin with a <code>/</code> it is relative to the home directory of theuser. An empty path <code>remote:</code> refers to the user's homedirectory.</p>-<h2 id="configuration-20">Configuration</h2>+<h2 id="configuration-22">Configuration</h2><p>To create an FTP configuration named <code>remote</code>, run</p><pre class="console"><code>rclone config</code></pre><p>Rclone config guides you through an interactive setup process. A@@ -43915,7 +46703,8 @@ use the special "anonymous" username. Traditionally, this user accountaccepts any string as a password, although it is common to use eitherthe password "anonymous" or "guest". Some servers require the use of avalid e-mail address as password.</p>-<p>Using <a href="#backend-path-to-dir">on-the-fly</a> or <a+<p>Using <a+href="https://rclone.org/docs/#backend-path-to-dir">on-the-fly</a> or <ahref="https://rclone.org/docs/#connection-strings">connection string</a>remotes makes it easy to access such servers, without requiring anyconfiguration in advance. The following are examples of that:</p>@@ -43953,7 +46742,7 @@ FTP backend then use the <code>override</code> syntax, eg</p><pre class="text"><code>override.ca_cert = XXXoverride.client_cert = XXXoverride.client_key = XXX</code></pre>-<h3 id="restricted-filename-characters-9">Restricted filename+<h3 id="restricted-filename-characters-11">Restricted filenamecharacters</h3><p>In addition to the <ahref="https://rclone.org/overview/#restricted-characters">default@@ -44002,7 +46791,7 @@ of sensible encoding settings for major FTP servers: ProFTPd, PureFTPd,VsFTPd. Just hit a selection number when prompted.</p><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/ftp/ftp.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-20">Standard options</h3>+<h3 id="standard-options-21">Standard options</h3><p>Here are the Standard options specific to ftp (FTP).</p><h4 id="ftp-host">--ftp-host</h4><p>FTP host to connect to.</p>@@ -44069,7 +46858,7 @@ an encrypted one. Cannot be used in combination with implicit FTPS.</p><li>Type: bool</li><li>Default: false</li></ul>-<h3 id="advanced-options-20">Advanced options</h3>+<h3 id="advanced-options-21">Advanced options</h3><p>Here are the Advanced options specific to ftp (FTP).</p><h4 id="ftp-concurrency">--ftp-concurrency</h4><p>Maximum number of FTP simultaneous connections, 0 for unlimited.</p>@@ -44303,7 +47092,7 @@ section in the overview</a> for more info.</p><li>Required: false</li></ul><!-- autogenerated options stop -->-<h2 id="limitations-14">Limitations</h2>+<h2 id="limitations-15">Limitations</h2><p>FTP servers acting as rclone remotes must support<code>passive</code> mode. The mode cannot be configured as<code>passive</code> is the only supported one. Rclone's FTP@@ -44331,7 +47120,7 @@ is).</p>present.</p><p>The <code>ftp_proxy</code> environment variable is not currentlysupported.</p>-<h3 id="modification-times-5">Modification times</h3>+<h3 id="modification-times-6">Modification times</h3><p>File modification time (timestamps) is supported to 1 secondresolution for major FTP servers: ProFTPd, PureFTPd, VsFTPd, andFileZilla FTP server. The <code>VsFTPd</code> server has non-standard@@ -44361,7 +47150,7 @@ and going to the "My Profile" section. Copy the "Account API token" foruse in the config file.</p><p>Note that if you wish to connect rclone to Gofile you will need apremium account.</p>-<h2 id="configuration-21">Configuration</h2>+<h2 id="configuration-23">Configuration</h2><p>Here is an example of how to make a remote called<code>remote</code>. First run:</p><pre class="console"><code>rclone config</code></pre>@@ -44408,12 +47197,12 @@ y/e/d> y</code></pre><pre class="console"><code>rclone lsf remote:</code></pre><p>To copy a local directory to an Gofile directory called backup</p><pre class="console"><code>rclone copy /home/source remote:backup</code></pre>-<h3 id="modification-times-and-hashes-10">Modification times and+<h3 id="modification-times-and-hashes-11">Modification times andhashes</h3><p>Gofile supports modification times with a resolution of 1 second.</p><p>Gofile supports MD5 hashes, so you can use the<code>--checksum</code> flag.</p>-<h3 id="restricted-filename-characters-10">Restricted filename+<h3 id="restricted-filename-characters-12">Restricted filenamecharacters</h3><p>In addition to the <ahref="https://rclone.org/overview/#restricted-characters">default@@ -44497,7 +47286,7 @@ name:</p><p>Invalid UTF-8 bytes will also be <ahref="https://rclone.org/overview/#invalid-utf8">replaced</a>, as theycan't be used in JSON strings.</p>-<h3 id="public-links">Public Links</h3>+<h3 id="public-links-1">Public Links</h3><p>Gofile supports <code>rclone link</code> to make public links tofiles or directories. If you specify a directory it will download as a<code>zip</code> file. You can use the <code>--expire</code> flag to@@ -44525,7 +47314,7 @@ set</p><p>To restrict rclone to the <code>Files</code> directory.</p><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/gofile/gofile.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-21">Standard options</h3>+<h3 id="standard-options-22">Standard options</h3><p>Here are the Standard options specific to gofile (Gofile).</p><h4 id="gofile-access-token">--gofile-access-token</h4><p>API Access token</p>@@ -44537,7 +47326,7 @@ set</p><li>Type: string</li><li>Required: false</li></ul>-<h3 id="advanced-options-21">Advanced options</h3>+<h3 id="advanced-options-22">Advanced options</h3><p>Here are the Advanced options specific to gofile (Gofile).</p><h4 id="gofile-root-folder-id">--gofile-root-folder-id</h4><p>ID of the root folder</p>@@ -44592,7 +47381,7 @@ Slash,LtGt,DoubleQuote,Colon,Question,Asterisk,Pipe,BackSlash,Del,Ctl,LeftPeriod<li>Required: false</li></ul><!-- autogenerated options stop -->-<h2 id="limitations-15">Limitations</h2>+<h2 id="limitations-16">Limitations</h2><p>Gofile only supports filenames up to 255 characters in length, wherea character is a unicode character.</p><p>Directories should not be cached for more than 24h otherwise files in@@ -44617,7 +47406,7 @@ messages in the log about duplicates.</p><p>Paths are specified as <code>remote:bucket</code> (or<code>remote:</code> for the <code>lsd</code> command.) You may putsubdirectories in too, e.g. <code>remote:bucket/path/to/dir</code>.</p>-<h2 id="configuration-22">Configuration</h2>+<h2 id="configuration-24">Configuration</h2><p>The initial setup for google cloud storage involves getting a tokenfrom Google Cloud Storage which you need to do in your browser.<code>rclone config</code> walks you through it.</p>@@ -44891,7 +47680,7 @@ with metadata documentation</a></p><p>Eg <code>--header-upload "Content-Type text/potato"</code></p><p>Note that the last of these is for setting custom metadata in theform <code>--header-upload "x-goog-meta-key: value"</code></p>-<h3 id="modification-times-6">Modification times</h3>+<h3 id="modification-times-7">Modification times</h3><p>Google Cloud Storage stores md5sum natively. Google's <ahref="https://cloud.google.com/storage/docs/gsutil">gsutil</a> toolstores modification time with one-second precision as@@ -44908,7 +47697,7 @@ gsutil only contain timestamps with one-second precision. If you userclone to sync files previously uploaded by gsutil, rclone will attemptto update modification time for all these files. To avoid these possiblyunnecessary updates, use <code>--modify-window 1s</code>.</p>-<h3 id="restricted-filename-characters-11">Restricted filename+<h3 id="restricted-filename-characters-13">Restricted filenamecharacters</h3><table><thead>@@ -44946,7 +47735,7 @@ href="https://rclone.org/overview/#invalid-utf8">replaced</a>, as theycan't be used in JSON strings.</p><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/googlecloudstorage/googlecloudstorage.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-22">Standard options</h3>+<h3 id="standard-options-23">Standard options</h3><p>Here are the Standard options specific to google cloud storage(Google Cloud Storage (this is not Google Drive)).</p><h4 id="gcs-client-id">--gcs-client-id</h4>@@ -45334,7 +48123,7 @@ is blank.</p></ul></li></ul></li></ul>-<h3 id="advanced-options-22">Advanced options</h3>+<h3 id="advanced-options-23">Advanced options</h3><p>Here are the Advanced options specific to google cloud storage(Google Cloud Storage (this is not Google Drive)).</p><h4 id="gcs-token">--gcs-token</h4>@@ -45481,7 +48270,7 @@ section in the overview</a> for more info.</p><li>Required: false</li></ul><!-- autogenerated options stop -->-<h2 id="limitations-16">Limitations</h2>+<h2 id="limitations-17">Limitations</h2><p><code>rclone about</code> is not supported by the Google CloudStorage backend. Backends without this capability cannot determine freespace for an rclone mount or use policy <code>mfs</code> (most free@@ -45493,7 +48282,7 @@ href="https://rclone.org/commands/rclone_about/">rclone about</a>.</p><p>Paths are specified as <code>drive:path</code></p><p>Drive paths may be as deep as required, e.g.<code>drive:directory/subdirectory</code>.</p>-<h2 id="configuration-23">Configuration</h2>+<h2 id="configuration-25">Configuration</h2><p>The initial setup for drive involves getting a token from Googledrive which you need to do in your browser. <code>rclone config</code>walks you through it.</p>@@ -45837,13 +48626,13 @@ gives:</p><li>without <code>--fast-list</code>: 22:05 min</li><li>with <code>--fast-list</code>: 58s</li></ul>-<h3 id="modification-times-and-hashes-11">Modification times and+<h3 id="modification-times-and-hashes-12">Modification times andhashes</h3><p>Google drive stores modification times accurate to 1 ms.</p><p>Hash algorithms MD5, SHA1 and SHA256 are supported. Note, however,that a small fraction of files uploaded may not have SHA1 or SHA256hashes especially if they were uploaded before 2018.</p>-<h3 id="restricted-filename-characters-12">Restricted filename+<h3 id="restricted-filename-characters-14">Restricted filenamecharacters</h3><p>Only Invalid UTF-8 bytes will be <ahref="https://rclone.org/overview/#invalid-utf8">replaced</a>, as they@@ -46236,7 +49025,7 @@ available Google Documents.</p></table><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/drive/drive.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-23">Standard options</h3>+<h3 id="standard-options-24">Standard options</h3><p>Here are the Standard options specific to drive (Google Drive).</p><h4 id="drive-client-id">--drive-client-id</h4><p>Google Application Client Id Leave blank to use rclone's shared@@ -46314,7 +49103,7 @@ environment variables such as <code>${RCLONE_CONFIG_DIR}</code>.</p><li>Type: string</li><li>Required: false</li></ul>-<h3 id="advanced-options-23">Advanced options</h3>+<h3 id="advanced-options-24">Advanced options</h3><p>Here are the Advanced options specific to drive (Google Drive).</p><h4 id="drive-token">--drive-token</h4><p>OAuth Access Token as a JSON blob.</p>@@ -46656,7 +49445,7 @@ consequences.</p>recommended usage is using the flag form --drive-size-as-quota whendoing rclone ls/lsl/lsf/lsjson/etc only.</p><p>If you do use this flag for syncing (not recommended) then you will-need to use --ignore size also.</p>+need to use --ignore-size also.</p><p>Properties:</p><ul><li>Config: size_as_quota</li>@@ -47184,34 +49973,34 @@ account.</p><p>Usage example:</p><pre class="console"><code>rclone backend [-o config] drives drive:</code></pre><p>This will return a JSON list of objects like this:</p>-<div class="sourceCode" id="cb1396"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb1396-1"><a href="#cb1396-1" aria-hidden="true" tabindex="-1"></a><span class="ot">[</span></span>-<span id="cb1396-2"><a href="#cb1396-2" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>-<span id="cb1396-3"><a href="#cb1396-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"id"</span><span class="fu">:</span> <span class="st">"0ABCDEF-01234567890"</span><span class="fu">,</span></span>-<span id="cb1396-4"><a href="#cb1396-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"kind"</span><span class="fu">:</span> <span class="st">"drive#teamDrive"</span><span class="fu">,</span></span>-<span id="cb1396-5"><a href="#cb1396-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"name"</span><span class="fu">:</span> <span class="st">"My Drive"</span></span>-<span id="cb1396-6"><a href="#cb1396-6" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span><span class="ot">,</span></span>-<span id="cb1396-7"><a href="#cb1396-7" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>-<span id="cb1396-8"><a href="#cb1396-8" aria-hidden="true" tabindex="-1"></a> <span class="dt">"id"</span><span class="fu">:</span> <span class="st">"0ABCDEFabcdefghijkl"</span><span class="fu">,</span></span>-<span id="cb1396-9"><a href="#cb1396-9" aria-hidden="true" tabindex="-1"></a> <span class="dt">"kind"</span><span class="fu">:</span> <span class="st">"drive#teamDrive"</span><span class="fu">,</span></span>-<span id="cb1396-10"><a href="#cb1396-10" aria-hidden="true" tabindex="-1"></a> <span class="dt">"name"</span><span class="fu">:</span> <span class="st">"Test Drive"</span></span>-<span id="cb1396-11"><a href="#cb1396-11" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>-<span id="cb1396-12"><a href="#cb1396-12" aria-hidden="true" tabindex="-1"></a><span class="ot">]</span></span></code></pre></div>+<div class="sourceCode" id="cb1455"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb1455-1"><a href="#cb1455-1" aria-hidden="true" tabindex="-1"></a><span class="ot">[</span></span>+<span id="cb1455-2"><a href="#cb1455-2" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>+<span id="cb1455-3"><a href="#cb1455-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"id"</span><span class="fu">:</span> <span class="st">"0ABCDEF-01234567890"</span><span class="fu">,</span></span>+<span id="cb1455-4"><a href="#cb1455-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"kind"</span><span class="fu">:</span> <span class="st">"drive#teamDrive"</span><span class="fu">,</span></span>+<span id="cb1455-5"><a href="#cb1455-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"name"</span><span class="fu">:</span> <span class="st">"My Drive"</span></span>+<span id="cb1455-6"><a href="#cb1455-6" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span><span class="ot">,</span></span>+<span id="cb1455-7"><a href="#cb1455-7" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>+<span id="cb1455-8"><a href="#cb1455-8" aria-hidden="true" tabindex="-1"></a> <span class="dt">"id"</span><span class="fu">:</span> <span class="st">"0ABCDEFabcdefghijkl"</span><span class="fu">,</span></span>+<span id="cb1455-9"><a href="#cb1455-9" aria-hidden="true" tabindex="-1"></a> <span class="dt">"kind"</span><span class="fu">:</span> <span class="st">"drive#teamDrive"</span><span class="fu">,</span></span>+<span id="cb1455-10"><a href="#cb1455-10" aria-hidden="true" tabindex="-1"></a> <span class="dt">"name"</span><span class="fu">:</span> <span class="st">"Test Drive"</span></span>+<span id="cb1455-11"><a href="#cb1455-11" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>+<span id="cb1455-12"><a href="#cb1455-12" aria-hidden="true" tabindex="-1"></a><span class="ot">]</span></span></code></pre></div><p>With the -o config parameter it will output the list in a formatsuitable for adding to a config file to make aliases for all the drivesfound and a combined drive.</p>-<div class="sourceCode" id="cb1397"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1397-1"><a href="#cb1397-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[My Drive]</span></span>-<span id="cb1397-2"><a href="#cb1397-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> alias</span></span>-<span id="cb1397-3"><a href="#cb1397-3" aria-hidden="true" tabindex="-1"></a><span class="dt">remote </span><span class="ot">=</span><span class="st"> drive,team_drive=0ABCDEF-01234567890,root_folder_id=:</span></span>-<span id="cb1397-4"><a href="#cb1397-4" aria-hidden="true" tabindex="-1"></a></span>-<span id="cb1397-5"><a href="#cb1397-5" aria-hidden="true" tabindex="-1"></a><span class="kw">[Test Drive]</span></span>-<span id="cb1397-6"><a href="#cb1397-6" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> alias</span></span>-<span id="cb1397-7"><a href="#cb1397-7" aria-hidden="true" tabindex="-1"></a><span class="dt">remote </span><span class="ot">=</span><span class="st"> drive,team_drive=0ABCDEFabcdefghijkl,root_folder_id=:</span></span>-<span id="cb1397-8"><a href="#cb1397-8" aria-hidden="true" tabindex="-1"></a></span>-<span id="cb1397-9"><a href="#cb1397-9" aria-hidden="true" tabindex="-1"></a><span class="kw">[AllDrives]</span></span>-<span id="cb1397-10"><a href="#cb1397-10" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> combine</span></span>-<span id="cb1397-11"><a href="#cb1397-11" aria-hidden="true" tabindex="-1"></a><span class="dt">upstreams </span><span class="ot">=</span><span class="st"> "My Drive=My Drive:" "Test Drive=Test Drive:"</span></span></code></pre></div>+<div class="sourceCode" id="cb1456"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1456-1"><a href="#cb1456-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[My Drive]</span></span>+<span id="cb1456-2"><a href="#cb1456-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> alias</span></span>+<span id="cb1456-3"><a href="#cb1456-3" aria-hidden="true" tabindex="-1"></a><span class="dt">remote </span><span class="ot">=</span><span class="st"> drive,team_drive=0ABCDEF-01234567890,root_folder_id=:</span></span>+<span id="cb1456-4"><a href="#cb1456-4" aria-hidden="true" tabindex="-1"></a></span>+<span id="cb1456-5"><a href="#cb1456-5" aria-hidden="true" tabindex="-1"></a><span class="kw">[Test Drive]</span></span>+<span id="cb1456-6"><a href="#cb1456-6" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> alias</span></span>+<span id="cb1456-7"><a href="#cb1456-7" aria-hidden="true" tabindex="-1"></a><span class="dt">remote </span><span class="ot">=</span><span class="st"> drive,team_drive=0ABCDEFabcdefghijkl,root_folder_id=:</span></span>+<span id="cb1456-8"><a href="#cb1456-8" aria-hidden="true" tabindex="-1"></a></span>+<span id="cb1456-9"><a href="#cb1456-9" aria-hidden="true" tabindex="-1"></a><span class="kw">[AllDrives]</span></span>+<span id="cb1456-10"><a href="#cb1456-10" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> combine</span></span>+<span id="cb1456-11"><a href="#cb1456-11" aria-hidden="true" tabindex="-1"></a><span class="dt">upstreams </span><span class="ot">=</span><span class="st"> "My Drive=My Drive:" "Test Drive=Test Drive:"</span></span></code></pre></div><p>Adding this to the rclone config file will cause those team drives tobe accessible with the aliases shown. Any illegal characters will besubstituted with "_" and duplicate names will have numbers suffixed. It@@ -47230,11 +50019,11 @@ use via the API.</p><p>Use the --interactive/-i or --dry-run flag to see what would berestored before restoring it.</p><p>Result:</p>-<div class="sourceCode" id="cb1400"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb1400-1"><a href="#cb1400-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>-<span id="cb1400-2"><a href="#cb1400-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Untrashed"</span><span class="fu">:</span> <span class="dv">17</span><span class="fu">,</span></span>-<span id="cb1400-3"><a href="#cb1400-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Errors"</span><span class="fu">:</span> <span class="dv">0</span></span>-<span id="cb1400-4"><a href="#cb1400-4" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<div class="sourceCode" id="cb1459"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb1459-1"><a href="#cb1459-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb1459-2"><a href="#cb1459-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Untrashed"</span><span class="fu">:</span> <span class="dv">17</span><span class="fu">,</span></span>+<span id="cb1459-3"><a href="#cb1459-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Errors"</span><span class="fu">:</span> <span class="dv">0</span></span>+<span id="cb1459-4"><a href="#cb1459-4" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div><h3 id="copyid">copyid</h3><p>Copy files by ID.</p><pre class="console"><code>rclone backend copyid remote: [options] [<arguments>+]</code></pre>@@ -47290,32 +50079,32 @@ escaped with characters. "'" becomes "'" and "" becomes "\", forexample to match a file named "foo ' .txt":</p><pre class="console"><code>rclone backend query drive: "name = 'foo \' \\\.txt'"</code></pre><p>The result is a JSON array of matches, for example:</p>-<div class="sourceCode" id="cb1411"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb1411-1"><a href="#cb1411-1" aria-hidden="true" tabindex="-1"></a><span class="ot">[</span></span>-<span id="cb1411-2"><a href="#cb1411-2" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>-<span id="cb1411-3"><a href="#cb1411-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"createdTime"</span><span class="fu">:</span> <span class="st">"2017-06-29T19:58:28.537Z"</span><span class="fu">,</span></span>-<span id="cb1411-4"><a href="#cb1411-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"id"</span><span class="fu">:</span> <span class="st">"0AxBe_CDEF4zkGHI4d0FjYko2QkD"</span><span class="fu">,</span></span>-<span id="cb1411-5"><a href="#cb1411-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"md5Checksum"</span><span class="fu">:</span> <span class="st">"68518d16be0c6fbfab918be61d658032"</span><span class="fu">,</span></span>-<span id="cb1411-6"><a href="#cb1411-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">"mimeType"</span><span class="fu">:</span> <span class="st">"text/plain"</span><span class="fu">,</span></span>-<span id="cb1411-7"><a href="#cb1411-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"modifiedTime"</span><span class="fu">:</span> <span class="st">"2024-02-02T10:40:02.874Z"</span><span class="fu">,</span></span>-<span id="cb1411-8"><a href="#cb1411-8" aria-hidden="true" tabindex="-1"></a> <span class="dt">"name"</span><span class="fu">:</span> <span class="st">"foo ' </span><span class="ch">\\</span><span class="st">.txt"</span><span class="fu">,</span></span>-<span id="cb1411-9"><a href="#cb1411-9" aria-hidden="true" tabindex="-1"></a> <span class="dt">"parents"</span><span class="fu">:</span> <span class="ot">[</span></span>-<span id="cb1411-10"><a href="#cb1411-10" aria-hidden="true" tabindex="-1"></a> <span class="st">"0BxAe_BCDE4zkFGZpcWJGek0xbzC"</span></span>-<span id="cb1411-11"><a href="#cb1411-11" aria-hidden="true" tabindex="-1"></a> <span class="ot">]</span><span class="fu">,</span></span>-<span id="cb1411-12"><a href="#cb1411-12" aria-hidden="true" tabindex="-1"></a> <span class="dt">"resourceKey"</span><span class="fu">:</span> <span class="st">"0-ABCDEFGHIXJQpIGqBJq3MC"</span><span class="fu">,</span></span>-<span id="cb1411-13"><a href="#cb1411-13" aria-hidden="true" tabindex="-1"></a> <span class="dt">"sha1Checksum"</span><span class="fu">:</span> <span class="st">"8f284fa768bfb4e45d076a579ab3905ab6bfa893"</span><span class="fu">,</span></span>-<span id="cb1411-14"><a href="#cb1411-14" aria-hidden="true" tabindex="-1"></a> <span class="dt">"size"</span><span class="fu">:</span> <span class="st">"311"</span><span class="fu">,</span></span>-<span id="cb1411-15"><a href="#cb1411-15" aria-hidden="true" tabindex="-1"></a> <span class="dt">"webViewLink"</span><span class="fu">:</span> <span class="st">"https://drive.google.com/file/d/0AxBe_CDEF4zkGHI4d0FjYko2QkD/view?usp=drivesdk</span><span class="ch">\u0026</span><span class="st">resourcekey=0-ABCDEFGHIXJQpIGqBJq3MC"</span></span>-<span id="cb1411-16"><a href="#cb1411-16" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>-<span id="cb1411-17"><a href="#cb1411-17" aria-hidden="true" tabindex="-1"></a><span class="ot">]</span></span>-<span id="cb1411-18"><a href="#cb1411-18" aria-hidden="true" tabindex="-1"></a><span class="er">```console</span></span>-<span id="cb1411-19"><a href="#cb1411-19" aria-hidden="true" tabindex="-1"></a></span>-<span id="cb1411-20"><a href="#cb1411-20" aria-hidden="true" tabindex="-1"></a><span class="er">###</span> <span class="er">rescue</span></span>-<span id="cb1411-21"><a href="#cb1411-21" aria-hidden="true" tabindex="-1"></a></span>-<span id="cb1411-22"><a href="#cb1411-22" aria-hidden="true" tabindex="-1"></a><span class="er">Rescue</span> <span class="er">or</span> <span class="er">delete</span> <span class="er">any</span> <span class="er">orphaned</span> <span class="er">files.</span></span>-<span id="cb1411-23"><a href="#cb1411-23" aria-hidden="true" tabindex="-1"></a></span>-<span id="cb1411-24"><a href="#cb1411-24" aria-hidden="true" tabindex="-1"></a><span class="er">```console</span></span>-<span id="cb1411-25"><a href="#cb1411-25" aria-hidden="true" tabindex="-1"></a><span class="er">rclone</span> <span class="er">backend</span> <span class="er">rescue</span> <span class="er">remote:</span> <span class="ot">[</span><span class="er">options</span><span class="ot">]</span> <span class="ot">[</span><span class="er"><arguments>+</span><span class="ot">]</span></span></code></pre></div>+<div class="sourceCode" id="cb1470"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb1470-1"><a href="#cb1470-1" aria-hidden="true" tabindex="-1"></a><span class="ot">[</span></span>+<span id="cb1470-2"><a href="#cb1470-2" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>+<span id="cb1470-3"><a href="#cb1470-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"createdTime"</span><span class="fu">:</span> <span class="st">"2017-06-29T19:58:28.537Z"</span><span class="fu">,</span></span>+<span id="cb1470-4"><a href="#cb1470-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"id"</span><span class="fu">:</span> <span class="st">"0AxBe_CDEF4zkGHI4d0FjYko2QkD"</span><span class="fu">,</span></span>+<span id="cb1470-5"><a href="#cb1470-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"md5Checksum"</span><span class="fu">:</span> <span class="st">"68518d16be0c6fbfab918be61d658032"</span><span class="fu">,</span></span>+<span id="cb1470-6"><a href="#cb1470-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">"mimeType"</span><span class="fu">:</span> <span class="st">"text/plain"</span><span class="fu">,</span></span>+<span id="cb1470-7"><a href="#cb1470-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"modifiedTime"</span><span class="fu">:</span> <span class="st">"2024-02-02T10:40:02.874Z"</span><span class="fu">,</span></span>+<span id="cb1470-8"><a href="#cb1470-8" aria-hidden="true" tabindex="-1"></a> <span class="dt">"name"</span><span class="fu">:</span> <span class="st">"foo ' </span><span class="ch">\\</span><span class="st">.txt"</span><span class="fu">,</span></span>+<span id="cb1470-9"><a href="#cb1470-9" aria-hidden="true" tabindex="-1"></a> <span class="dt">"parents"</span><span class="fu">:</span> <span class="ot">[</span></span>+<span id="cb1470-10"><a href="#cb1470-10" aria-hidden="true" tabindex="-1"></a> <span class="st">"0BxAe_BCDE4zkFGZpcWJGek0xbzC"</span></span>+<span id="cb1470-11"><a href="#cb1470-11" aria-hidden="true" tabindex="-1"></a> <span class="ot">]</span><span class="fu">,</span></span>+<span id="cb1470-12"><a href="#cb1470-12" aria-hidden="true" tabindex="-1"></a> <span class="dt">"resourceKey"</span><span class="fu">:</span> <span class="st">"0-ABCDEFGHIXJQpIGqBJq3MC"</span><span class="fu">,</span></span>+<span id="cb1470-13"><a href="#cb1470-13" aria-hidden="true" tabindex="-1"></a> <span class="dt">"sha1Checksum"</span><span class="fu">:</span> <span class="st">"8f284fa768bfb4e45d076a579ab3905ab6bfa893"</span><span class="fu">,</span></span>+<span id="cb1470-14"><a href="#cb1470-14" aria-hidden="true" tabindex="-1"></a> <span class="dt">"size"</span><span class="fu">:</span> <span class="st">"311"</span><span class="fu">,</span></span>+<span id="cb1470-15"><a href="#cb1470-15" aria-hidden="true" tabindex="-1"></a> <span class="dt">"webViewLink"</span><span class="fu">:</span> <span class="st">"https://drive.google.com/file/d/0AxBe_CDEF4zkGHI4d0FjYko2QkD/view?usp=drivesdk</span><span class="ch">\u0026</span><span class="st">resourcekey=0-ABCDEFGHIXJQpIGqBJq3MC"</span></span>+<span id="cb1470-16"><a href="#cb1470-16" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>+<span id="cb1470-17"><a href="#cb1470-17" aria-hidden="true" tabindex="-1"></a><span class="ot">]</span></span>+<span id="cb1470-18"><a href="#cb1470-18" aria-hidden="true" tabindex="-1"></a><span class="er">```console</span></span>+<span id="cb1470-19"><a href="#cb1470-19" aria-hidden="true" tabindex="-1"></a></span>+<span id="cb1470-20"><a href="#cb1470-20" aria-hidden="true" tabindex="-1"></a><span class="er">###</span> <span class="er">rescue</span></span>+<span id="cb1470-21"><a href="#cb1470-21" aria-hidden="true" tabindex="-1"></a></span>+<span id="cb1470-22"><a href="#cb1470-22" aria-hidden="true" tabindex="-1"></a><span class="er">Rescue</span> <span class="er">or</span> <span class="er">delete</span> <span class="er">any</span> <span class="er">orphaned</span> <span class="er">files.</span></span>+<span id="cb1470-23"><a href="#cb1470-23" aria-hidden="true" tabindex="-1"></a></span>+<span id="cb1470-24"><a href="#cb1470-24" aria-hidden="true" tabindex="-1"></a><span class="er">```console</span></span>+<span id="cb1470-25"><a href="#cb1470-25" aria-hidden="true" tabindex="-1"></a><span class="er">rclone</span> <span class="er">backend</span> <span class="er">rescue</span> <span class="er">remote:</span> <span class="ot">[</span><span class="er">options</span><span class="ot">]</span> <span class="ot">[</span><span class="er"><arguments>+</span><span class="ot">]</span></span></code></pre></div><p>This command rescues or deletes any orphaned files ordirectories.</p><p>Sometimes files can get orphaned in Google Drive. This means that@@ -47333,7 +50122,7 @@ level:</p><p>Third delete all orphaned files to the trash:</p><pre class="console"><code>rclone backend rescue drive: -o delete</code></pre><!-- autogenerated options stop -->-<h2 id="limitations-17">Limitations</h2>+<h2 id="limitations-18">Limitations</h2><p>Drive has quite a lot of rate limiting. This causes rclone to belimited to transferring about 2 files per second only. Individual filesmay be transferred much faster at 100s of MiB/s but lots of small files@@ -47441,7 +50230,7 @@ remove scopes" and select the three above and press update or go to thepress add to table then update.</p><p>You should now see the three scopes on your Data access page. Nowpress save at the bottom!</p></li>-<li><p>After adding scopes, click Audience Scroll down and click "+ Add+<li><p>After adding scopes, click Audience. Scroll down and click "+ Addusers". Add yourself as a test user and press save.</p></li><li><p>Go to Overview on the left panel, click "Create OAuth client".Choose an application type of "Desktop app" and click "Create". (the@@ -47452,7 +50241,15 @@ chose "Internal" you don't need to publish and can skip straight to Step10 but your destination drive must be part of the same GoogleWorkspace.)</p></li><li><p>Go to "Audience" and then click "PUBLISH APP" button and confirm.-Add yourself as a test user if you haven't already.</p></li>+Add yourself as a test user if you haven't already.</p>+<p>(If the "PUBLISH APP" button is greyed out, Google now requires a+homepage URL and a privacy policy URL before publishing, even for a+personal single-user app. Go to "Branding" in the left panel, fill in+"Application home page" and "Application privacy policy link" - a free+GitHub Pages site is fine for these if you don't have a domain of your+own - then "Authorized domains" with the domain you used, and click+Save. Return to "Audience" and the "PUBLISH APP" button should now be+available.)</p></li><li><p>Provide the noted client ID and client secret to rclone.</p></li><li><p>Run the web-based authorization flow from within<code>rclone config</code>, by answering "Y" when it asks "Token already@@ -47510,7 +50307,7 @@ section</a> carefully to make sure it is suitable for your use.</p>photos it uploaded. This limitation is due to policy changes at Google.You may need to run <code>rclone config reconnect remote:</code> to makerclone work again after upgrading to rclone v1.70.</p>-<h2 id="configuration-24">Configuration</h2>+<h2 id="configuration-26">Configuration</h2><p>The initial setup for google cloud storage involves getting a tokenfrom Google Photos which you need to do in your browser.<code>rclone config</code> walks you through it.</p>@@ -47718,7 +50515,7 @@ or by you. This is similar to the Sharing tab in the Google Photos webinterface.</p><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/googlephotos/googlephotos.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-24">Standard options</h3>+<h3 id="standard-options-25">Standard options</h3><p>Here are the Standard options specific to google photos (GooglePhotos).</p><h4 id="gphotos-client-id">--gphotos-client-id</h4>@@ -47757,7 +50554,7 @@ access to your photos, otherwise rclone will request full access.</p><li>Type: bool</li><li>Default: false</li></ul>-<h3 id="advanced-options-24">Advanced options</h3>+<h3 id="advanced-options-25">Advanced options</h3><p>Here are the Advanced options specific to google photos (GooglePhotos).</p><h4 id="gphotos-token">--gphotos-token</h4>@@ -47949,7 +50746,7 @@ default based on the batch_mode in use.</p><li>Required: false</li></ul><!-- autogenerated options stop -->-<h2 id="limitations-18">Limitations</h2>+<h2 id="limitations-19">Limitations</h2><p>Only images and videos can be uploaded. If you attempt to upload nonvideos or images or formats that Google Photos doesn't understand,rclone will upload the file, then Google Photos will give an error when@@ -48007,7 +50804,7 @@ uploaded an image to <code>upload</code> then uploaded the same image to<code>album/my_album</code> will be what it was uploaded with initially,not what you uploaded it with to <code>album</code>. In practise thisshouldn't cause too many problems.</p>-<h3 id="modification-times-7">Modification times</h3>+<h3 id="modification-times-8">Modification times</h3><p>The date shown of media in Google Photos is the creation date asdetermined by the EXIF information, or the upload date if that is notknown.</p>@@ -48130,18 +50927,18 @@ y/e/d> y</code></pre>config file, usually <code>YOURHOME/.config/rclone/rclone.conf</code>.Open it in your favorite text editor, find section for the base remoteand create new section for hasher like in the following examples:</p>-<div class="sourceCode" id="cb1429"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1429-1"><a href="#cb1429-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[Hasher1]</span></span>-<span id="cb1429-2"><a href="#cb1429-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> hasher</span></span>-<span id="cb1429-3"><a href="#cb1429-3" aria-hidden="true" tabindex="-1"></a><span class="dt">remote </span><span class="ot">=</span><span class="st"> myRemote:path</span></span>-<span id="cb1429-4"><a href="#cb1429-4" aria-hidden="true" tabindex="-1"></a><span class="dt">hashes </span><span class="ot">=</span><span class="st"> md5</span></span>-<span id="cb1429-5"><a href="#cb1429-5" aria-hidden="true" tabindex="-1"></a><span class="dt">max_age </span><span class="ot">=</span><span class="st"> </span><span class="kw">off</span></span>-<span id="cb1429-6"><a href="#cb1429-6" aria-hidden="true" tabindex="-1"></a></span>-<span id="cb1429-7"><a href="#cb1429-7" aria-hidden="true" tabindex="-1"></a><span class="kw">[Hasher2]</span></span>-<span id="cb1429-8"><a href="#cb1429-8" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> hasher</span></span>-<span id="cb1429-9"><a href="#cb1429-9" aria-hidden="true" tabindex="-1"></a><span class="dt">remote </span><span class="ot">=</span><span class="st"> /local/path</span></span>-<span id="cb1429-10"><a href="#cb1429-10" aria-hidden="true" tabindex="-1"></a><span class="dt">hashes </span><span class="ot">=</span><span class="st"> dropbox,sha1</span></span>-<span id="cb1429-11"><a href="#cb1429-11" aria-hidden="true" tabindex="-1"></a><span class="dt">max_age </span><span class="ot">=</span><span class="st"> 24h</span></span></code></pre></div>+<div class="sourceCode" id="cb1488"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1488-1"><a href="#cb1488-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[Hasher1]</span></span>+<span id="cb1488-2"><a href="#cb1488-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> hasher</span></span>+<span id="cb1488-3"><a href="#cb1488-3" aria-hidden="true" tabindex="-1"></a><span class="dt">remote </span><span class="ot">=</span><span class="st"> myRemote:path</span></span>+<span id="cb1488-4"><a href="#cb1488-4" aria-hidden="true" tabindex="-1"></a><span class="dt">hashes </span><span class="ot">=</span><span class="st"> md5</span></span>+<span id="cb1488-5"><a href="#cb1488-5" aria-hidden="true" tabindex="-1"></a><span class="dt">max_age </span><span class="ot">=</span><span class="st"> </span><span class="kw">off</span></span>+<span id="cb1488-6"><a href="#cb1488-6" aria-hidden="true" tabindex="-1"></a></span>+<span id="cb1488-7"><a href="#cb1488-7" aria-hidden="true" tabindex="-1"></a><span class="kw">[Hasher2]</span></span>+<span id="cb1488-8"><a href="#cb1488-8" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> hasher</span></span>+<span id="cb1488-9"><a href="#cb1488-9" aria-hidden="true" tabindex="-1"></a><span class="dt">remote </span><span class="ot">=</span><span class="st"> /local/path</span></span>+<span id="cb1488-10"><a href="#cb1488-10" aria-hidden="true" tabindex="-1"></a><span class="dt">hashes </span><span class="ot">=</span><span class="st"> dropbox,sha1</span></span>+<span id="cb1488-11"><a href="#cb1488-11" aria-hidden="true" tabindex="-1"></a><span class="dt">max_age </span><span class="ot">=</span><span class="st"> 24h</span></span></code></pre></div><p>Hasher takes basically the following parameters:</p><ul><li><code>remote</code> is required</li>@@ -48211,7 +51008,7 @@ full re-read/re-write of the files.</p><h2 id="configuration-reference">Configuration reference</h2><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/hasher/hasher.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-25">Standard options</h3>+<h3 id="standard-options-26">Standard options</h3><p>Here are the Standard options specific to hasher (Better checksumsfor other remotes).</p><h4 id="hasher-remote">--hasher-remote</h4>@@ -48242,7 +51039,7 @@ forever).</p><li>Type: Duration</li><li>Default: off</li></ul>-<h3 id="advanced-options-25">Advanced options</h3>+<h3 id="advanced-options-26">Advanced options</h3><p>Here are the Advanced options specific to hasher (Better checksumsfor other remotes).</p><h4 id="hasher-auto-size">--hasher-auto-size</h4>@@ -48363,14 +51160,14 @@ shared between multiple rclone processes.</p><p>Paths are specified as <code>huaweidrive:path</code></p><p>Huawei Drive paths may be as deep as required, e.g.<code>huaweidrive:directory/subdirectory</code>.</p>-<h2 id="configuration-25">Configuration</h2>+<h2 id="configuration-27">Configuration</h2><p>The initial setup for Huawei Drive involves getting a token fromHuawei Drive which you need to do in your browser.<code>rclone config</code> walks you through it.</p><p>Here is an example of how to make a remote called<code>remote</code>. First run:</p>-<div class="sourceCode" id="cb1444"><pre-class="sourceCode sh"><code class="sourceCode bash"><span id="cb1444-1"><a href="#cb1444-1" aria-hidden="true" tabindex="-1"></a><span class="ex">rclone</span> config</span></code></pre></div>+<div class="sourceCode" id="cb1503"><pre+class="sourceCode sh"><code class="sourceCode bash"><span id="cb1503-1"><a href="#cb1503-1" aria-hidden="true" tabindex="-1"></a><span class="ex">rclone</span> config</span></code></pre></div><p>This will guide you through an interactive setup process:</p><pre class="text"><code>No remotes found, make a new one?n) New remote@@ -48426,14 +51223,14 @@ it temporarily if you are running a host firewall, or use manualmode.</p><p>You can then use it like this,</p><p>List directories in top level of your drive</p>-<div class="sourceCode" id="cb1446"><pre-class="sourceCode sh"><code class="sourceCode bash"><span id="cb1446-1"><a href="#cb1446-1" aria-hidden="true" tabindex="-1"></a><span class="ex">rclone</span> lsd remote:</span></code></pre></div>+<div class="sourceCode" id="cb1505"><pre+class="sourceCode sh"><code class="sourceCode bash"><span id="cb1505-1"><a href="#cb1505-1" aria-hidden="true" tabindex="-1"></a><span class="ex">rclone</span> lsd remote:</span></code></pre></div><p>List all the files in your drive</p>-<div class="sourceCode" id="cb1447"><pre-class="sourceCode sh"><code class="sourceCode bash"><span id="cb1447-1"><a href="#cb1447-1" aria-hidden="true" tabindex="-1"></a><span class="ex">rclone</span> ls remote:</span></code></pre></div>+<div class="sourceCode" id="cb1506"><pre+class="sourceCode sh"><code class="sourceCode bash"><span id="cb1506-1"><a href="#cb1506-1" aria-hidden="true" tabindex="-1"></a><span class="ex">rclone</span> ls remote:</span></code></pre></div><p>To copy a local directory to a drive directory called backup</p>-<div class="sourceCode" id="cb1448"><pre-class="sourceCode sh"><code class="sourceCode bash"><span id="cb1448-1"><a href="#cb1448-1" aria-hidden="true" tabindex="-1"></a><span class="ex">rclone</span> copy /home/source remote:backup</span></code></pre></div>+<div class="sourceCode" id="cb1507"><pre+class="sourceCode sh"><code class="sourceCode bash"><span id="cb1507-1"><a href="#cb1507-1" aria-hidden="true" tabindex="-1"></a><span class="ex">rclone</span> copy /home/source remote:backup</span></code></pre></div><h3 id="getting-your-own-client-id-and-secret">Getting your own ClientID and Secret</h3><p>When you use rclone with Huawei Drive in its default configuration@@ -48466,14 +51263,19 @@ Development Guide".</p><h4 id="enable-drive-application-switch">Enable Drive ApplicationSwitch</h4><ol type="1">-<li>Log in to the AppGallery Connect website and click "My-Projects".</li>-<li>Find your project in the project list, select the application that-needs to access Drive in the project, and click "API Management".</li>-<li>In the "API Management" area of the "Project Settings" page, click-the switch on the right side of Drive Kit to ensure the switch is in the-open state.</li>+<li>Log in to AppGallery Connect with your Huawei account.</li>+<li>Create a project or select an existing project.</li>+<li>Add a web application to the project.</li>+<li>Open "Project settings" for the application and copy the OAuth 2.0+Client ID and Client Secret from the "General" tab.</li>+<li>Open "API Management" for the project and enable "Cloud Drive".</li>+<li>Add <code>http://127.0.0.1:53682/</code> as a redirect URI for the+OAuth 2.0 Client ID.</li></ol>+<p>The older Huawei Developer Console may show some of the same+credential information, but AppGallery Connect is the place to create+the project, application, API permission, and redirect URI used by+rclone.</p><h4 id="sign-huawei-cloud-drive-privacy-notice-and-user-agreement">SignHuawei Cloud Drive Privacy Notice and User Agreement</h4><p>Guide users to visit <a href="https://cloud.huawei.com/"@@ -48488,20 +51290,15 @@ Application Access Development Guide section in the "Huawei AccountService Development Guide".</p><p>Key steps summary:</p><ol type="1">-<li><p>Log into the <a-href="https://developer.huawei.com/consumer/en/console">Huawei Developer-Console</a> with your Huawei account.</p></li>-<li><p>Create a new project or select an existing project.</p></li>-<li><p>Go to "Manage APIs" and enable the "Drive Kit" API.</p></li>-<li><p>Click on "Credentials" in the left side panel.</p></li>-<li><p>Click "Create Credentials" and choose "OAuth client ID".</p></li>-<li><p>Choose "Web application" as the application type.</p></li>-<li><p>Add <code>http://localhost:53682/</code> to the "Authorized-redirect URIs".</p></li>-<li><p>Click "Create" and note down the Client ID and Client-Secret.</p></li>-<li><p>Provide the noted client ID and client secret to rclone during-configuration.</p></li>+<li>Create or select a project in AppGallery Connect.</li>+<li>Add a web application to the project.</li>+<li>Enable Cloud Drive in API Management.</li>+<li>Configure <code>http://127.0.0.1:53682/</code> as the OAuth redirect+URI.</li>+<li>Copy the OAuth 2.0 Client ID and Client Secret from the+application.</li>+<li>Provide the client ID and client secret to rclone during+configuration.</li></ol><h3 id="scopes-1">Scopes</h3><p>Rclone uses the following OAuth2 scopes when accessing Huawei@@ -48529,7 +51326,7 @@ Read-only access to file metadata only</li></ul><p>The default scopes provide read/write access to all files in yourHuawei Drive.</p>-<h3 id="modification-times-and-hashes-12">Modification times and+<h3 id="modification-times-and-hashes-13">Modification times andhashes</h3><p>Huawei Drive does not support setting modification times. The APIalways uses server-side timestamps, so file modification times from the@@ -48537,7 +51334,7 @@ source are not preserved during upload. Modification times are availableas read-only metadata (<code>mtime</code>, <code>btime</code>).</p><p>Hash algorithm SHA256 is supported for file integrityverification.</p>-<h3 id="restricted-filename-characters-13">Restricted filename+<h3 id="restricted-filename-characters-15">Restricted filenamecharacters</h3><p>Huawei Drive has strict restrictions on file and directory names:</p><ul>@@ -48569,7 +51366,7 @@ for recovery from network interruptions.</p>and quota information.</p><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/huaweidrive/huaweidrive.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-26">Standard options</h3>+<h3 id="standard-options-27">Standard options</h3><p>Here are the Standard options specific to huaweidrive (HuaweiDrive).</p><h4 id="huaweidrive-client-id">--huaweidrive-client-id</h4>@@ -48592,7 +51389,7 @@ Drive).</p><li>Type: string</li><li>Required: false</li></ul>-<h3 id="advanced-options-26">Advanced options</h3>+<h3 id="advanced-options-27">Advanced options</h3><p>Here are the Advanced options specific to huaweidrive (HuaweiDrive).</p><h4 id="huaweidrive-token">--huaweidrive-token</h4>@@ -48800,7 +51597,7 @@ backend.</p><p>See the <a href="https://rclone.org/docs/#metadata">metadata</a> docsfor more info.</p><!-- autogenerated options stop -->-<h2 id="limitations-19">Limitations</h2>+<h2 id="limitations-20">Limitations</h2><p>Huawei Drive has some rate limiting which may cause rclone to slowdown transfers to avoid hitting the limits.</p><h3 id="file-name-restrictions">File name restrictions</h3>@@ -48921,7 +51718,7 @@ is a distributed file-system, part of the <ahref="https://hadoop.apache.org/">Apache Hadoop</a> framework.</p><p>Paths are specified as <code>remote:</code> or<code>remote:path/to/dir</code>.</p>-<h2 id="configuration-26">Configuration</h2>+<h2 id="configuration-28">Configuration</h2><p>Here is an example of how to make a remote called<code>remote</code>. First run:</p><pre class="console"><code>rclone config</code></pre>@@ -49006,22 +51803,22 @@ docker build --rm -t rclone/test-hdfs .</code></pre><p><strong>NB</strong> it need few seconds to startup.</p><p>For this docker image the remote needs to be configured likethis:</p>-<div class="sourceCode" id="cb1456"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1456-1"><a href="#cb1456-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[remote]</span></span>-<span id="cb1456-2"><a href="#cb1456-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> hdfs</span></span>-<span id="cb1456-3"><a href="#cb1456-3" aria-hidden="true" tabindex="-1"></a><span class="dt">namenode </span><span class="ot">=</span><span class="st"> 127.0.0.1:8020</span></span>-<span id="cb1456-4"><a href="#cb1456-4" aria-hidden="true" tabindex="-1"></a><span class="dt">username </span><span class="ot">=</span><span class="st"> root</span></span></code></pre></div>+<div class="sourceCode" id="cb1515"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1515-1"><a href="#cb1515-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[remote]</span></span>+<span id="cb1515-2"><a href="#cb1515-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> hdfs</span></span>+<span id="cb1515-3"><a href="#cb1515-3" aria-hidden="true" tabindex="-1"></a><span class="dt">namenode </span><span class="ot">=</span><span class="st"> 127.0.0.1:8020</span></span>+<span id="cb1515-4"><a href="#cb1515-4" aria-hidden="true" tabindex="-1"></a><span class="dt">username </span><span class="ot">=</span><span class="st"> root</span></span></code></pre></div><p>You can stop this image with <code>docker kill rclone-hdfs</code>(<strong>NB</strong> it does not use volumes, so all data uploaded willbe lost.)</p>-<h3 id="modification-times-8">Modification times</h3>+<h3 id="modification-times-9">Modification times</h3><p>Time accurate to 1 second is stored.</p><h3 id="checksum">Checksum</h3><p>No checksums are implemented.</p><h3 id="usage-information">Usage information</h3><p>You can use the <code>rclone about remote:</code> command which willdisplay filesystem size and current usage.</p>-<h3 id="restricted-filename-characters-14">Restricted filename+<h3 id="restricted-filename-characters-16">Restricted filenamecharacters</h3><p>In addition to the <ahref="https://rclone.org/overview/#restricted-characters">default@@ -49047,7 +51844,7 @@ replaced:</p>href="https://rclone.org/overview/#invalid-utf8">replaced</a>.</p><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/hdfs/hdfs.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-27">Standard options</h3>+<h3 id="standard-options-28">Standard options</h3><p>Here are the Standard options specific to hdfs (Hadoop distributedfile system).</p><h4 id="hdfs-namenode">--hdfs-namenode</h4>@@ -49077,7 +51874,7 @@ namenodes at port 8020.</p></ul></li></ul></li></ul>-<h3 id="advanced-options-27">Advanced options</h3>+<h3 id="advanced-options-28">Advanced options</h3><p>Here are the Advanced options specific to hdfs (Hadoop distributedfile system).</p><h4 id="hdfs-service-principal-name">--hdfs-service-principal-name</h4>@@ -49136,7 +51933,7 @@ section in the overview</a> for more info.</p><li>Required: false</li></ul><!-- autogenerated options stop -->-<h2 id="limitations-20">Limitations</h2>+<h2 id="limitations-21">Limitations</h2><ul><li>Erasure coding not supported, see <ahref="https://github.com/rclone/rclone/issues/8808">issue #8808</a></li>@@ -49150,7 +51947,7 @@ href="https://github.com/rclone/rclone/issues/8808">issue #8808</a></li><p>The initial setup for hidrive involves getting a token from HiDrivewhich you need to do in your browser. <code>rclone config</code> walksyou through it.</p>-<h2 id="configuration-27">Configuration</h2>+<h2 id="configuration-29">Configuration</h2><p>Here is an example of how to make a remote called<code>remote</code>. First run:</p><pre class="console"><code>rclone config</code></pre>@@ -49244,14 +52041,14 @@ account again.</p><pre class="console"><code>rclone config reconnect remote:</code></pre><p>the process is very similar to the process of initial setupexemplified before.</p>-<h3 id="modification-times-and-hashes-13">Modification times and+<h3 id="modification-times-and-hashes-14">Modification times andhashes</h3><p>HiDrive allows modification times to be set on objects accurate to 1second.</p><p>HiDrive supports <a href="https://static.hidrive.com/dev/0001">itsown hash type</a> which is used to verify the integrity of file contentsafter successful transfers.</p>-<h3 id="restricted-filename-characters-15">Restricted filename+<h3 id="restricted-filename-characters-17">Restricted filenamecharacters</h3><p>HiDrive cannot store files or folders that include <code>/</code>(0x2F) or null-bytes (0x00) in their name. Any other characters can be@@ -49310,7 +52107,7 @@ information is not explicitly needed. For this, thedetails.</p><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/hidrive/hidrive.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-28">Standard options</h3>+<h3 id="standard-options-29">Standard options</h3><p>Here are the Standard options specific to hidrive (HiDrive).</p><h4 id="hidrive-client-id">--hidrive-client-id</h4><p>OAuth Client Id.</p>@@ -49353,7 +52150,7 @@ HiDrive.</p></ul></li></ul></li></ul>-<h3 id="advanced-options-28">Advanced options</h3>+<h3 id="advanced-options-29">Advanced options</h3><p>Here are the Advanced options specific to hidrive (HiDrive).</p><h4 id="hidrive-token">--hidrive-token</h4><p>OAuth Access Token as a JSON blob.</p>@@ -49545,7 +52342,7 @@ section in the overview</a> for more info.</p><li>Required: false</li></ul><!-- autogenerated options stop -->-<h2 id="limitations-21">Limitations</h2>+<h2 id="limitations-22">Limitations</h2><h3 id="symbolic-links">Symbolic links</h3><p>HiDrive is able to store symbolic links (<em>symlinks</em>) bydesign, for example, when unpacked from a zip archive.</p>@@ -49580,12 +52377,14 @@ remote url <code>https://beta.rclone.org/branch</code> and path<code>https://beta.rclone.org/branch/fix</code>, while with path<code>/fix</code> the resolved URL will be<code>https://beta.rclone.org/fix</code> as the absolute path is-resolved from the root of the domain.</p>+resolved from the root of the domain. The path can't change the scheme,+host or user of the configured url, so a path such as+<code>//other.example.com/fix</code> is rejected.</p><p>If the path following the <code>remote:</code> ends with<code>/</code> it will be assumed to point to a directory. If the pathdoes not end with <code>/</code>, then a HEAD request is sent and the-response used to decide if it it is treated as a file or a directory-(run with <code>-vv</code> to see details). When <a+response used to decide if it is treated as a file or a directory (run+with <code>-vv</code> to see details). When <ahref="#http-no-head">--http-no-head</a> is specified, a path withoutending <code>/</code> is always assumed to be a file. If rcloneincorrectly assumes the path is a file, the solution is to specify the@@ -49594,7 +52393,7 @@ ending it with <code>/</code> is always better as it avoids the initialHEAD request.</p><p>To just download a single file it is easier to use <ahref="https://rclone.org/commands/rclone_copyurl/">copyurl</a>.</p>-<h2 id="configuration-28">Configuration</h2>+<h2 id="configuration-30">Configuration</h2><p>Here is an example of how to make a remote called<code>remote</code>. First run:</p><pre class="console"><code>rclone config</code></pre>@@ -49671,7 +52470,7 @@ an unparseable response for <code>https://example.com/path/</code>:</p><p>If you just want to download a file or multiple files by URL thenusing <a href="https://rclone.org/commands/rclone_copyurl/">copyurl</a>is more efficient.</p>-<h3 id="modification-times-9">Modification times</h3>+<h3 id="modification-times-10">Modification times</h3><p>Most HTTP servers store time accurate to 1 second.</p><h3 id="checksum-1">Checksum</h3><p>No checksums are stored.</p>@@ -49683,7 +52482,7 @@ without a config file:</p><pre class="console"><code>rclone lsd :http,url='https://beta.rclone.org':</code></pre><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/http/http.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-29">Standard options</h3>+<h3 id="standard-options-30">Standard options</h3><p>Here are the Standard options specific to http (HTTP).</p><h4 id="http-url">--http-url</h4><p>URL of HTTP host to connect to.</p>@@ -49705,7 +52504,7 @@ a username and password.</p><li>Type: bool</li><li>Default: false</li></ul>-<h3 id="advanced-options-29">Advanced options</h3>+<h3 id="advanced-options-30">Advanced options</h3><p>Here are the Advanced options specific to http (HTTP).</p><h4 id="http-headers">--http-headers</h4><p>Set HTTP headers for all transactions.</p>@@ -49717,6 +52516,12 @@ used.</p>'"Cookie","name=value"'.</p><p>You can set multiple headers, e.g.'"Cookie","name=value","Authorization","xxx"'.</p>+<p>The headers are only sent to the host in the configured URL. If the+server redirects to another host (including a subdomain or a different+port) the headers are not sent to it, or to any further hop in that+redirect chain. When headers are set, a redirect from https to http is+refused as it would send them in cleartext, even with+--allow-https-downgrade.</p><p>Properties:</p><ul><li>Config: headers</li>@@ -49763,6 +52568,37 @@ may be in the listing.</p><li>Type: bool</li><li>Default: false</li></ul>+<h4 id="http-no-head-size">--http-no-head-size</h4>+<p>Set the reported file size for --http-no-head.</p>+<p>When HEAD requests are disabled, rclone normally reports an unknown+file size. Set this option to a non-negative value to report that size+for every file, including 0. Leave it unset (the default, shown as+"off") to keep reporting an unknown size.</p>+<p>The configured size is authoritative. If it does not match the actual+file size, transfers may fail their size check and retry unless+--ignore-size is used. It can also make multi-thread transfers request+incorrect ranges and make mounted files unreadable past the reported+size.</p>+<p>Properties:</p>+<ul>+<li>Config: no_head_size</li>+<li>Env Var: RCLONE_HTTP_NO_HEAD_SIZE</li>+<li>Type: SizeSuffix</li>+<li>Default: off</li>+</ul>+<h4 id="http-no-head-time">--http-no-head-time</h4>+<p>Set a default modification time for --http-no-head.</p>+<p>When HEAD requests are disabled, rclone normally reports an unset+modification time. Set this option to report the specified time for+every file instead. Changing the configured time later makes every file+appear modified to a sync.</p>+<p>Properties:</p>+<ul>+<li>Config: no_head_time</li>+<li>Env Var: RCLONE_HTTP_NO_HEAD_TIME</li>+<li>Type: Time</li>+<li>Default: off</li>+</ul><h4 id="http-description">--http-description</h4><p>Description of the remote.</p><p>Properties:</p>@@ -49865,7 +52701,7 @@ with the new parameters. Only new parameters need be passed as thevalues will default to those currently in use.</p><p>It doesn't return anything.</p><!-- autogenerated options stop -->-<h2 id="limitations-22">Limitations</h2>+<h2 id="limitations-23">Limitations</h2><p><code>rclone about</code> is not supported by the HTTP backend.Backends without this capability cannot determine free space for anrclone mount or use policy <code>mfs</code> (most free space) as a@@ -49885,7 +52721,7 @@ href="https://imagekit.io/registration/">create an account</a> onImageKit. Start with a free plan with generous usage limits. Then, asyour requirements grow, upgrade to a plan that best fits your needs. See<a href="https://imagekit.io/plans">the pricing details</a>.</p>-<h2 id="configuration-29">Configuration</h2>+<h2 id="configuration-31">Configuration</h2><p>Here is an example of making an imagekit configuration.</p><p>Firstly create a <a href="https://imagekit.io/">ImageKit.io</a>account and choose a plan.</p>@@ -49956,7 +52792,7 @@ y/e/d> y</code></pre><p>No checksums are supported.</p><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/imagekit/imagekit.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-30">Standard options</h3>+<h3 id="standard-options-31">Standard options</h3><p>Here are the Standard options specific to imagekit (ImageKit.io).</p><h4 id="imagekit-endpoint">--imagekit-endpoint</h4><p>You can find your ImageKit.io URL endpoint in your <a@@ -49988,7 +52824,7 @@ href="https://imagekit.io/dashboard/developer/api-keys">dashboard</a></p><li>Type: string</li><li>Required: true</li></ul>-<h3 id="advanced-options-30">Advanced options</h3>+<h3 id="advanced-options-31">Advanced options</h3><p>Here are the Advanced options specific to imagekit (ImageKit.io).</p><h4 id="imagekit-only-signed">--imagekit-only-signed</h4><p>If you have configured <code>Restrict unsigned image URLs</code> in@@ -50153,7 +52989,7 @@ image</td>for more info.</p><!-- autogenerated options stop --><h1 id="icloud-drive-and-photos">iCloud Drive and Photos</h1>-<h2 id="configuration-30">Configuration</h2>+<h2 id="configuration-32">Configuration</h2><p>The initial setup for an iCloud backend involves getting a trusttoken/session. This uses your regular Apple ID password plus 2FA, eitherfrom a trusted device prompt or an SMS code sent to a trusted phone@@ -50331,7 +53167,7 @@ videos to local disk for fast repeated access</li><p>The first listing of a very large album (e.g. 75,000 items in "AllPhotos") can take several minutes due to API pagination limits. Thishappens once — subsequent listings use the disk cache.</p>-<h3 id="limitations-23">Limitations</h3>+<h3 id="limitations-24">Limitations</h3><p>iCloud Photos is read-only. Upload, delete, rename, and moveoperations are not supported.</p><h2 id="advanced-data-protection">Advanced Data Protection</h2>@@ -50354,7 +53190,7 @@ prompt on your trusted device.</p>or delete and recreate the remote.</p><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/iclouddrive/iclouddrive.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-31">Standard options</h3>+<h3 id="standard-options-32">Standard options</h3><p>Here are the Standard options specific to iclouddrive (iCloud Driveand Photos).</p><h4 id="iclouddrive-service">--iclouddrive-service</h4>@@ -50398,7 +53234,7 @@ obscure</a>.</p><li>Type: string</li><li>Required: true</li></ul>-<h3 id="advanced-options-31">Advanced options</h3>+<h3 id="advanced-options-32">Advanced options</h3><p>Here are the Advanced options specific to iclouddrive (iCloud Driveand Photos).</p><h4 id="iclouddrive-client-id">--iclouddrive-client-id</h4>@@ -50572,7 +53408,7 @@ filtering</a>.</p><p>Which excludes from the sync any files which have the<code>source=metadata</code> or <code>format=Metadata</code> flags whichare added to Internet Archive auto-created files.</p>-<h2 id="configuration-31">Configuration</h2>+<h2 id="configuration-33">Configuration</h2><p>Here is an example of making an internetarchive configuration. Mostapplies to the other providers as well, any differences are described <ahref="#providers">below</a>.</p>@@ -50645,7 +53481,7 @@ d) Delete this remotey/e/d> y</code></pre><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/internetarchive/internetarchive.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-32">Standard options</h3>+<h3 id="standard-options-33">Standard options</h3><p>Here are the Standard options specific to internetarchive (InternetArchive).</p><h4@@ -50685,7 +53521,7 @@ infrastructure.</p><li>Type: bool</li><li>Default: true</li></ul>-<h3 id="advanced-options-32">Advanced options</h3>+<h3 id="advanced-options-33">Advanced options</h3><p>Here are the Advanced options specific to internetarchive (InternetArchive).</p><h4 id="internetarchive-endpoint">--internetarchive-endpoint</h4>@@ -50905,13 +53741,13 @@ encrypted cloud storage service.</p><p>Paths are specified as <code>remote:path</code></p><p>Paths may be as deep as required, e.g.<code>remote:directory/subdirectory</code>.</p>-<h2 id="limitations-24">Limitations</h2>+<h2 id="limitations-25">Limitations</h2><p><strong>Note:</strong> The Internxt backend may not work with allaccount types. Please refer to <ahref="https://internxt.com/pricing">Internxt plan details</a> or contact<a href="https://help.internxt.com">Internxt support</a> to verifyrclone compatibility with your subscription.</p>-<h2 id="configuration-32">Configuration</h2>+<h2 id="configuration-34">Configuration</h2><p>Here is an example of how to make a remote called<code>internxt</code>. Run <code>rclone config</code> and follow theprompts:</p>@@ -50979,20 +53815,20 @@ rclone mount internxt: /path/to/mountpoint# Check storage usagerclone about internxt:</code></pre>-<h3 id="modification-times-and-hashes-14">Modification times and+<h3 id="modification-times-and-hashes-15">Modification times andhashes</h3><p>The Internxt backend does not support hashes.</p><p>Modification times are read from the server but cannot be set. Thebackend reports <code>ModTimeNotSupported</code> precision, somodification times will not be used for sync comparisons.</p>-<h3 id="restricted-filename-characters-16">Restricted filename+<h3 id="restricted-filename-characters-18">Restricted filenamecharacters</h3><p>The Internxt backend replaces the <ahref="https://rclone.org/overview/#restricted-characters">defaultrestricted characters set</a>.</p><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/internxt/internxt.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-33">Standard options</h3>+<h3 id="standard-options-34">Standard options</h3><p>Here are the Standard options specific to internxt (InternxtDrive).</p><h4 id="internxt-email">--internxt-email</h4>@@ -51016,7 +53852,7 @@ obscure</a>.</p><li>Type: string</li><li>Required: true</li></ul>-<h3 id="advanced-options-33">Advanced options</h3>+<h3 id="advanced-options-34">Advanced options</h3><p>Here are the Advanced options specific to internxt (InternxtDrive).</p><h4@@ -51259,7 +54095,7 @@ migrated to the newer methods, some white-label versions (those fromElkjøp) still used this legacy method for a long time. Currently thereare no known uses of this, it is still supported by rclone, but thesupport will be removed in a future version.</p>-<h2 id="configuration-33">Configuration</h2>+<h2 id="configuration-35">Configuration</h2><p>Here is an example of how to make a remote called <code>remote</code>with the default setup. First run:</p><pre class="console"><code>rclone config</code></pre>@@ -51398,7 +54234,7 @@ to long wait time before the first results are shown.</p><a href="https://rclone.org/overview/#mime-type">MIME types</a> andmetadata item <a href="#metadata">utime</a> are not available when using<code>--fast-list</code>.</p>-<h3 id="modification-times-and-hashes-15">Modification times and+<h3 id="modification-times-and-hashes-16">Modification times andhashes</h3><p>Jottacloud allows modification times to be set on objects accurate to1 second. These will be used to detect whether objects need syncing or@@ -51416,7 +54252,7 @@ available, so this does not apply. Starting with rclone version 1.52 thesame is true for encrypted remotes (in older versions the crypt backendwould not calculate hashes for uploads from local disk, so theJottacloud backend had to do it as described above).</p>-<h3 id="restricted-filename-characters-17">Restricted filename+<h3 id="restricted-filename-characters-19">Restricted filenamecharacters</h3><p>In addition to the <ahref="https://rclone.org/overview/#restricted-characters">default@@ -51500,7 +54336,7 @@ available in the remote.</p>limit (unless it is unlimited) and the current usage.</p><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/jottacloud/jottacloud.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-34">Standard options</h3>+<h3 id="standard-options-35">Standard options</h3><p>Here are the Standard options specific to jottacloud(Jottacloud).</p><h4 id="jottacloud-client-id">--jottacloud-client-id</h4>@@ -51523,7 +54359,7 @@ limit (unless it is unlimited) and the current usage.</p><li>Type: string</li><li>Required: false</li></ul>-<h3 id="advanced-options-34">Advanced options</h3>+<h3 id="advanced-options-35">Advanced options</h3><p>Here are the Advanced options specific to jottacloud(Jottacloud).</p><h4 id="jottacloud-token">--jottacloud-token</h4>@@ -51696,7 +54532,7 @@ backend</td><p>See the <a href="https://rclone.org/docs/#metadata">metadata</a> docsfor more info.</p><!-- autogenerated options stop -->-<h2 id="limitations-25">Limitations</h2>+<h2 id="limitations-26">Limitations</h2><p>Note that Jottacloud is case insensitive so you can't have a filecalled "Hello.doc" and one called "hello.doc".</p><p>There are quite a few characters that can't be in Jottacloud file@@ -51710,11 +54546,297 @@ length.</p>files and folders which may cause Copy, Move and DirMove operations topreviously deleted paths to fail. Emptying the trash should help in suchcases.</p>+<h1 id="kdrive">kDrive</h1>+<p>Paths are specified as <code>remote:path</code></p>+<p>Paths may be as deep as required, e.g.+<code>remote:directory/subdirectory</code>.</p>+<h2 id="configuration-36">Configuration</h2>+<p>The initial setup for kDrive involves getting an OAuth2 token which+you need to do in your browser. <code>rclone config</code> walks you+through it.</p>+<p>kDrive uses the PKCE OAuth2 flow (RFC 7636), which means no+<code>client_secret</code> is required. The <code>client_id</code> is+optional and defaults to rclone's built-in application credentials.</p>+<p>Here is an example of how to make a remote called+<code>remote</code>. First run:</p>+<div class="sourceCode" id="cb1561"><pre+class="sourceCode sh"><code class="sourceCode bash"><span id="cb1561-1"><a href="#cb1561-1" aria-hidden="true" tabindex="-1"></a><span class="ex">rclone</span> config</span></code></pre></div>+<p>This will guide you through an interactive setup process:</p>+<pre class="text"><code>No remotes found, make a new one?+n) New remote+s) Set configuration password+q) Quit config+n/s/q> n+name> remote+Type of storage to configure.+Choose a number from below, or type in your own value+[snip]+XX / Infomaniak kDrive+ \ "kdrive"+[snip]+Storage> kdrive++Option client_id.+OAuth Client Id.+Leave blank normally.+Enter a value. Press Enter to leave empty.+client_id>++Option drive_id.+ID of the kDrive to use.+When showing a folder on kdrive, you can find the drive_id here:+https://ksuite.infomaniak.com/{account_id}/kdrive/app/drive/{drive_id}/files/...+Enter a value. Press Enter to leave empty.+drive_id> 0654321++Use web browser to automatically authenticate rclone with remote?+ * Say Y if the machine running rclone has a web browser you can use+ * Say N if running rclone on a (remote) machine without web browser access+If not sure try Y. If Y failed, try N.+y/n> y++Waiting for code...+Got code++Configuration complete.+Options:+- type: kdrive+- drive_id: 0654321++Keep this "remote" remote?+y) Yes this is OK+e) Edit this remote+d) Delete this remote+y/e/d> y</code></pre>+<p>Once configured you can then use <code>rclone</code> like this+(replace <code>remote</code> with the name you gave your remote):</p>+<p>List directories in top level of your rclone</p>+<div class="sourceCode" id="cb1563"><pre+class="sourceCode sh"><code class="sourceCode bash"><span id="cb1563-1"><a href="#cb1563-1" aria-hidden="true" tabindex="-1"></a><span class="ex">rclone</span> lsd remote:</span></code></pre></div>+<p><strong>Note</strong> that by default the top level directory of a+private kDrive storage is your "Private" folder. When used with a Pro+account, a second folder "Common documents" is also available, which can+be used as the root by setting <code>root_folder_id</code> to+<code>common</code>. <em>These top-level folders cannot be+modified</em>.</p>+<p>List all the files in your kDrive</p>+<div class="sourceCode" id="cb1564"><pre+class="sourceCode sh"><code class="sourceCode bash"><span id="cb1564-1"><a href="#cb1564-1" aria-hidden="true" tabindex="-1"></a><span class="ex">rclone</span> ls remote:</span></code></pre></div>+<p>To copy a local directory to a kDrive directory called backup</p>+<div class="sourceCode" id="cb1565"><pre+class="sourceCode sh"><code class="sourceCode bash"><span id="cb1565-1"><a href="#cb1565-1" aria-hidden="true" tabindex="-1"></a><span class="ex">rclone</span> copy /home/source remote:backup</span></code></pre></div>+<h3 id="modification-times-and-hashes-17">Modification times and+hashes</h3>+<p>kDrive allows modification times to be set on objects accurate to 1+second. These will be used to detect whether objects need syncing or+not.</p>+<p>kDrive supports the XXH3 checksum hash, so you can use the+<code>--checksum</code> flag. XXH3 is only available for files uploaded+as a single chunk: files uploaded in chunks (by rclone when larger than+<code>--kdrive-chunk-size</code>, or by the kDrive apps which chunk+files at their own sizes) have a "nested" hash which can't be compared+with a plain XXH3, and rclone reports no hash for those files.</p>+<h3 id="restricted-filename-characters-20">Restricted filename+characters</h3>+<p>In addition to the <a+href="https://rclone.org/overview/#restricted-characters">default+restricted characters set</a> the following characters are also+replaced:</p>+<table>+<thead>+<tr class="header">+<th>Character</th>+<th style="text-align: center;">Value</th>+<th style="text-align: center;">Replacement</th>+</tr>+</thead>+<tbody>+<tr class="odd">+<td>LeftSpace</td>+<td style="text-align: center;">0x20</td>+<td style="text-align: center;">␠</td>+</tr>+<tr class="even">+<td>RightSpace</td>+<td style="text-align: center;">0x20</td>+<td style="text-align: center;">␠</td>+</tr>+</tbody>+</table>+<p>Invalid UTF-8 bytes will also be <a+href="https://rclone.org/overview/#invalid-utf8">replaced</a>, as they+can't be used in JSON strings.</p>+<h3 id="overwriting-files">Overwriting files</h3>+<p>When rclone overwrites an existing file, kDrive keeps the replaced+content as a new version of the file. Old versions count against the+drive quota, so repeatedly overwriting large files grows the storage+usage. Manage old versions from the kDrive web app.</p>+<h3 id="uploading-files-of-unknown-size">Uploading files of unknown+size</h3>+<p>kDrive can't upload files of unknown size, so+<code>rclone rcat</code> and streaming uploads are not supported.</p>+<h3 id="deleting-files-4">Deleting files</h3>+<p>Deleted files will be moved to the trash. Your subscription level+will determine how long items stay in the trash.+<code>rclone cleanup</code> can be used to empty the trash.</p>+<h3 id="root-folder-id-5">Root folder ID</h3>+<p>You can set the <code>root_folder_id</code> for rclone. This is the+directory (identified by its numeric <code>Folder ID</code>) that rclone+considers to be the root of your kDrive drive.</p>+<p>Normally you will leave this blank and rclone will use your private+directory. You can also use one of the convenience shortcuts:</p>+<ul>+<li><code>private</code>: your user private directory (default)</li>+<li><code>common</code>: the kDrive common directory, shared among+users</li>+</ul>+<p>Or the numeric <code>Folder ID</code> of any directory in your drive.+The root of the drive only contains the <code>Private</code> and+<code>Common documents</code> folders, so a numeric ID is needed to use+any other directory as root. In order to find the+<code>Folder ID</code>, list the parent directory of the folder you wish+rclone to use with <code>rclone lsf</code>; the first column of the+output is the folder ID:</p>+<p>Example:</p>+<pre><code>$ rclone lsf --dirs-only -Fip --csv remote:Documents+456,Sub folder/+789,Photos/</code></pre>+<p>So if the folder you want rclone to use is "Photos/", then use the+returned id from <code>rclone lsf</code> command (ex. <code>789</code>)+as the <code>root_folder_id</code> variable value in the config+file.</p>+<!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/kdrive/kdrive.go and run make backenddocs to verify -->+<!-- markdownlint-disable-line line-length -->+<h3 id="standard-options-36">Standard options</h3>+<p>Here are the Standard options specific to kdrive (Infomaniak+kDrive).</p>+<h4 id="kdrive-client-id">--kdrive-client-id</h4>+<p>OAuth Client Id.</p>+<p>Leave blank normally.</p>+<p>Properties:</p>+<ul>+<li>Config: client_id</li>+<li>Env Var: RCLONE_KDRIVE_CLIENT_ID</li>+<li>Type: string</li>+<li>Required: false</li>+</ul>+<h4 id="kdrive-drive-id">--kdrive-drive-id</h4>+<p>ID of the kDrive to use.</p>+<p>When showing a folder on kdrive, you can find the drive_id here:+https://ksuite.infomaniak.com/{account_id}/kdrive/app/drive/{drive_id}/files/...</p>+<p>Properties:</p>+<ul>+<li>Config: drive_id</li>+<li>Env Var: RCLONE_KDRIVE_DRIVE_ID</li>+<li>Type: string</li>+<li>Required: false</li>+</ul>+<h3 id="advanced-options-36">Advanced options</h3>+<p>Here are the Advanced options specific to kdrive (Infomaniak+kDrive).</p>+<h4 id="kdrive-token">--kdrive-token</h4>+<p>OAuth Access Token as a JSON blob.</p>+<p>Properties:</p>+<ul>+<li>Config: token</li>+<li>Env Var: RCLONE_KDRIVE_TOKEN</li>+<li>Type: string</li>+<li>Required: false</li>+</ul>+<h4 id="kdrive-auth-url">--kdrive-auth-url</h4>+<p>Auth server URL.</p>+<p>Leave blank to use the provider defaults.</p>+<p>Properties:</p>+<ul>+<li>Config: auth_url</li>+<li>Env Var: RCLONE_KDRIVE_AUTH_URL</li>+<li>Type: string</li>+<li>Required: false</li>+</ul>+<h4 id="kdrive-token-url">--kdrive-token-url</h4>+<p>Token server url.</p>+<p>Leave blank to use the provider defaults.</p>+<p>Properties:</p>+<ul>+<li>Config: token_url</li>+<li>Env Var: RCLONE_KDRIVE_TOKEN_URL</li>+<li>Type: string</li>+<li>Required: false</li>+</ul>+<h4 id="kdrive-encoding">--kdrive-encoding</h4>+<p>The encoding for the backend.</p>+<p>See the <a href="https://rclone.org/overview/#encoding">encoding+section in the overview</a> for more info.</p>+<p>Properties:</p>+<ul>+<li>Config: encoding</li>+<li>Env Var: RCLONE_KDRIVE_ENCODING</li>+<li>Type: Encoding</li>+<li>Default: Slash,Del,Ctl,LeftSpace,RightSpace,InvalidUtf8,Dot</li>+</ul>+<h4 id="kdrive-root-folder-id">--kdrive-root-folder-id</h4>+<p>The folder to use as root.</p>+<p>You may use either one of the convenience shortcuts [private or+common] or an explicit folder ID.</p>+<ul>+<li>private: your user private directory</li>+<li>common: the kDrive common directory, shared among users</li>+</ul>+<p>Properties:</p>+<ul>+<li>Config: root_folder_id</li>+<li>Env Var: RCLONE_KDRIVE_ROOT_FOLDER_ID</li>+<li>Type: string</li>+<li>Required: false</li>+</ul>+<h4 id="kdrive-chunk-size">--kdrive-chunk-size</h4>+<p>Files above this size are uploaded in chunks of this size.</p>+<p>The kDrive API accepts chunks between 1 MB and 1 GB, up to 10,000+chunks per file and a total file size of 1 TiB. The chunk size is raised+automatically for very large files to stay within the 10,000 chunks+limit.</p>+<p>Properties:</p>+<ul>+<li>Config: chunk_size</li>+<li>Env Var: RCLONE_KDRIVE_CHUNK_SIZE</li>+<li>Type: SizeSuffix</li>+<li>Default: 20Mi</li>+</ul>+<h4 id="kdrive-upload-concurrency">--kdrive-upload-concurrency</h4>+<p>Concurrency for chunked uploads.</p>+<p>Properties:</p>+<ul>+<li>Config: upload_concurrency</li>+<li>Env Var: RCLONE_KDRIVE_UPLOAD_CONCURRENCY</li>+<li>Type: int</li>+<li>Default: 4</li>+</ul>+<h4 id="kdrive-endpoint">--kdrive-endpoint</h4>+<p>The API endpoint to use.</p>+<p>Leave blank normally. There is no reason to change the endpoint+except for internal use or beta-testing.</p>+<p>Properties:</p>+<ul>+<li>Config: endpoint</li>+<li>Env Var: RCLONE_KDRIVE_ENDPOINT</li>+<li>Type: string</li>+<li>Required: false</li>+</ul>+<h4 id="kdrive-description">--kdrive-description</h4>+<p>Description of the remote.</p>+<p>Properties:</p>+<ul>+<li>Config: description</li>+<li>Env Var: RCLONE_KDRIVE_DESCRIPTION</li>+<li>Type: string</li>+<li>Required: false</li>+</ul>+<!-- autogenerated options stop --><h1 id="koofr">Koofr</h1><p>Paths are specified as <code>remote:path</code></p><p>Paths may be as deep as required, e.g.<code>remote:directory/subdirectory</code>.</p>-<h2 id="configuration-34">Configuration</h2>+<h2 id="configuration-37">Configuration</h2><p>The initial setup for Koofr involves creating an application passwordfor rclone. You can do that by opening the Koofr <ahref="https://app.koofr.net/app/admin/preferences/password">web@@ -51789,7 +54911,7 @@ choose an alternative mount instead of your primary storage.</p><pre class="console"><code>rclone ls koofr:</code></pre><p>To copy a local directory to an Koofr directory called backup</p><pre class="console"><code>rclone copy /home/source koofr:backup</code></pre>-<h3 id="restricted-filename-characters-18">Restricted filename+<h3 id="restricted-filename-characters-21">Restricted filenamecharacters</h3><p>In addition to the <ahref="https://rclone.org/overview/#restricted-characters">default@@ -51816,7 +54938,7 @@ href="https://rclone.org/overview/#invalid-utf8">replaced</a>, as theycan't be used in XML strings.</p><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/koofr/koofr.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-35">Standard options</h3>+<h3 id="standard-options-37">Standard options</h3><p>Here are the Standard options specific to koofr (Koofr, Digi Storageand other Koofr-compatible storage providers).</p><h4 id="koofr-provider">--koofr-provider</h4>@@ -51876,7 +54998,7 @@ obscure</a>.</p><li>Type: string</li><li>Required: true</li></ul>-<h3 id="advanced-options-35">Advanced options</h3>+<h3 id="advanced-options-37">Advanced options</h3><p>Here are the Advanced options specific to koofr (Koofr, Digi Storageand other Koofr-compatible storage providers).</p><h4 id="koofr-mountid">--koofr-mountid</h4>@@ -51921,7 +55043,7 @@ section in the overview</a> for more info.</p><li>Required: false</li></ul><!-- autogenerated options stop -->-<h2 id="limitations-26">Limitations</h2>+<h2 id="limitations-27">Limitations</h2><p>Note that Koofr is case insensitive so you can't have a file called"Hello.doc" and one called "hello.doc".</p><h2 id="providers-2">Providers</h2>@@ -52061,7 +55183,7 @@ y/e/d> y</code></pre><h1 id="linkbox">Linkbox</h1><p>Linkbox is <a href="https://linkbox.to/">a private clouddrive</a>.</p>-<h2 id="configuration-35">Configuration</h2>+<h2 id="configuration-38">Configuration</h2><p>Here is an example of making a remote for Linkbox.</p><p>You will need: - An API token fromhttps://www.linkbox.to/admin/account - The email and password for your@@ -52120,7 +55242,7 @@ y/e/d> y</code></pre><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/linkbox/linkbox.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-36">Standard options</h3>+<h3 id="standard-options-38">Standard options</h3><p>Here are the Standard options specific to linkbox (Linkbox).</p><h4 id="linkbox-token">--linkbox-token</h4><p>Token from https://www.linkbox.to/admin/account</p>@@ -52152,7 +55274,7 @@ obscure</a>.</p><li>Type: string</li><li>Required: true</li></ul>-<h3 id="advanced-options-36">Advanced options</h3>+<h3 id="advanced-options-38">Advanced options</h3><p>Here are the Advanced options specific to linkbox (Linkbox).</p><h4 id="linkbox-description">--linkbox-description</h4><p>Description of the remote.</p>@@ -52164,10 +55286,18 @@ obscure</a>.</p><li>Required: false</li></ul><!-- autogenerated options stop -->-<h2 id="limitations-27">Limitations</h2>+<h2 id="limitations-28">Limitations</h2><p>Invalid UTF-8 bytes will also be <ahref="https://rclone.org/overview/#invalid-utf8">replaced</a>, as theycan't be used in JSON strings.</p>+<p>Linkbox converts file extensions to lower case on upload, so a file+uploaded as <code>file.TXT</code> is stored as+<code>file.txt</code>.</p>+<p>Linkbox renames an uploaded file to <code>name(1)</code> if its name+is the start of the name of a file already in the same directory, for+example when uploading <code>apple</code> to a directory containing+<code>applebanana</code>. rclone then can't find the file it uploaded+and the upload fails with <code>object not found</code>.</p><h1 id="mail.ru-cloud">Mail.ru Cloud</h1><p><a href="https://cloud.mail.ru/">Mail.ru Cloud</a> is a cloud storageprovided by a Russian internet company <a href="https://mail.ru">Mail.Ru@@ -52192,7 +55322,7 @@ deduplication, the hash algorithm is a modified SHA1</li>submit file hash instead of long file upload (this optimization issupported by rclone)</li></ul>-<h2 id="configuration-36">Configuration</h2>+<h2 id="configuration-39">Configuration</h2><p>Here is an example of making a mailru configuration.</p><p>First create a Mail.ru Cloud account and choose a tariff.</p><p>You will need to log in and create an app password for rclone. Rclone@@ -52283,7 +55413,7 @@ You can use the configured backend as shown below:</p><p>Sync <code>/home/local/directory</code> to the remote path, deletingany excess files in the path.</p><pre class="console"><code>rclone sync --interactive /home/local/directory remote:directory</code></pre>-<h3 id="modification-times-and-hashes-16">Modification times and+<h3 id="modification-times-and-hashes-18">Modification times andhashes</h3><p>Files support a modification time attribute with up to 1 secondprecision. Directories do not have a modification time, which is shown@@ -52304,7 +55434,7 @@ command does not take any path arguments.</p><p>To view your current quota you can use the<code>rclone about remote:</code> command which will display your usagelimit (quota) and the current usage.</p>-<h3 id="restricted-filename-characters-19">Restricted filename+<h3 id="restricted-filename-characters-22">Restricted filenamecharacters</h3><p>In addition to the <ahref="https://rclone.org/overview/#restricted-characters">default@@ -52366,7 +55496,7 @@ href="https://rclone.org/overview/#invalid-utf8">replaced</a>, as theycan't be used in JSON strings.</p><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/mailru/mailru.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-37">Standard options</h3>+<h3 id="standard-options-39">Standard options</h3><p>Here are the Standard options specific to mailru (Mail.ru Cloud).</p><h4 id="mailru-client-id">--mailru-client-id</h4><p>OAuth Client Id.</p>@@ -52441,7 +55571,7 @@ this optimization.</p></ul></li></ul></li></ul>-<h3 id="advanced-options-37">Advanced options</h3>+<h3 id="advanced-options-39">Advanced options</h3><p>Here are the Advanced options specific to mailru (Mail.ru Cloud).</p><h4 id="mailru-token">--mailru-token</h4><p>OAuth Access Token as a JSON blob.</p>@@ -52613,7 +55743,7 @@ Slash,LtGt,DoubleQuote,Colon,Question,Asterisk,Pipe,BackSlash,Del,Ctl,InvalidUtf<li>Required: false</li></ul><!-- autogenerated options stop -->-<h2 id="limitations-28">Limitations</h2>+<h2 id="limitations-29">Limitations</h2><p>File size limits depend on your account. A single file size islimited by 2G for a free account and unlimited for paid tariffs. Pleaserefer to the Mail.ru site for the total uploaded size limits.</p>@@ -52633,7 +55763,7 @@ recommended for new projects.</p><p>Paths are specified as <code>remote:path</code></p><p>Paths may be as deep as required, e.g.<code>remote:directory/subdirectory</code>.</p>-<h2 id="configuration-37">Configuration</h2>+<h2 id="configuration-40">Configuration</h2><p>Here is an example of how to make a remote called<code>remote</code>. First run:</p><pre class="console"><code>rclone config</code></pre>@@ -52684,10 +55814,10 @@ use the credentials in <code>rclone</code> will fail.</p><pre class="console"><code>rclone ls remote:</code></pre><p>To copy a local directory to an Mega directory called backup</p><pre class="console"><code>rclone copy /home/source remote:backup</code></pre>-<h3 id="modification-times-and-hashes-17">Modification times and+<h3 id="modification-times-and-hashes-19">Modification times andhashes</h3><p>Mega does not support modification times or hashes yet.</p>-<h3 id="restricted-filename-characters-20">Restricted filename+<h3 id="restricted-filename-characters-23">Restricted filenamecharacters</h3><table><thead>@@ -52787,7 +55917,7 @@ log-in and you are sure the user and the password are correct, likelyyou have got the remote blocked for a while.</p><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/mega/mega.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-38">Standard options</h3>+<h3 id="standard-options-40">Standard options</h3><p>Here are the Standard options specific to mega (Mega).</p><h4 id="mega-user">--mega-user</h4><p>User name.</p>@@ -52820,7 +55950,7 @@ one</p><li>Type: string</li><li>Required: false</li></ul>-<h3 id="advanced-options-38">Advanced options</h3>+<h3 id="advanced-options-40">Advanced options</h3><p>Here are the Advanced options specific to mega (Mega).</p><h4 id="mega-debug">--mega-debug</h4><p>Output more debug from Mega.</p>@@ -52888,7 +56018,7 @@ type has sufficient memory/CPU to execute the commands. Use the resourcemonitoring tools to inspect after sending the commands. Look <ahref="https://forum.rclone.org/t/rclone-with-mega-appears-to-work-only-in-some-accounts/40233/4">atthis issue</a>.</p>-<h2 id="limitations-29">Limitations</h2>+<h2 id="limitations-30">Limitations</h2><p>This backend uses the <ahref="https://github.com/t3rm1n4l/go-mega">go-mega go library</a> whichis an opensource go library implementing the Mega API. There doesn't@@ -52902,7 +56032,7 @@ there are likely quite a few errors still remaining in this library.</p><p>The memory backend behaves like a bucket-based remote (e.g. like s3).Because it has no parameters you can just use it with the<code>:memory:</code> remote name.</p>-<h2 id="configuration-38">Configuration</h2>+<h2 id="configuration-41">Configuration</h2><p>You can configure it as a remote like this with<code>rclone config</code> too if you want to:</p><pre class="text"><code>No remotes found, make a new one\?@@ -52936,18 +56066,18 @@ testing or with an rclone server or rclone mount, e.g.</p><pre class="console"><code>rclone mount :memory: /mnt/tmprclone serve webdav :memory:rclone serve sftp :memory:</code></pre>-<h3 id="modification-times-and-hashes-18">Modification times and+<h3 id="modification-times-and-hashes-20">Modification times andhashes</h3><p>The memory backend supports MD5 hashes and modification timesaccurate to 1 nS.</p>-<h3 id="restricted-filename-characters-21">Restricted filename+<h3 id="restricted-filename-characters-24">Restricted filenamecharacters</h3><p>The memory backend replaces the <ahref="https://rclone.org/overview/#restricted-characters">defaultrestricted characters set</a>.</p><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/memory/memory.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="advanced-options-39">Advanced options</h3>+<h3 id="advanced-options-41">Advanced options</h3><p>Here are the Advanced options specific to memory (In memory objectstorage system.).</p><h4 id="memory-discard">--memory-discard</h4>@@ -52996,7 +56126,7 @@ code:</p><p>The initial setup for Netstorage involves getting an account andsecret. Use <code>rclone config</code> to walk you through the setupprocess.</p>-<h2 id="configuration-39">Configuration</h2>+<h2 id="configuration-42">Configuration</h2><p>Here's an example of how to make a remote called<code>ns1</code>.</p><ol type="1">@@ -53166,7 +56296,7 @@ using quick-delete method will not delete the tree immediately andobjects targeted for quick-delete may still be accessible.</p><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/netstorage/netstorage.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-39">Standard options</h3>+<h3 id="standard-options-41">Standard options</h3><p>Here are the Standard options specific to netstorage (AkamaiNetStorage).</p><h4 id="netstorage-host">--netstorage-host</h4>@@ -53203,7 +56333,7 @@ obscure</a>.</p><li>Type: string</li><li>Required: true</li></ul>-<h3 id="advanced-options-40">Advanced options</h3>+<h3 id="advanced-options-42">Advanced options</h3><p>Here are the Advanced options specific to netstorage (AkamaiNetStorage).</p><h4 id="netstorage-protocol">--netstorage-protocol</h4>@@ -53268,7 +56398,7 @@ applicable.</p><code>remote:</code> for the <code>lsd</code> command.) You may putsubdirectories in too, e.g.<code>remote:container/path/to/dir</code>.</p>-<h2 id="configuration-40">Configuration</h2>+<h2 id="configuration-43">Configuration</h2><p>Here is an example of making a Microsoft Azure Blob Storageconfiguration. For a remote called <code>remote</code>. First run:</p><pre class="console"><code>rclone config</code></pre>@@ -53318,7 +56448,7 @@ deleting any excess files in the container.</p>fewer transactions in exchange for more memory. See the <ahref="https://rclone.org/docs/#fast-list">rclone docs</a> for moredetails.</p>-<h3 id="modification-times-and-hashes-19">Modification times and+<h3 id="modification-times-and-hashes-21">Modification times andhashes</h3><p>The modification time is stored as metadata on the object with the<code>mtime</code> key. It is stored using RFC3339 Format time with@@ -53360,14 +56490,29 @@ ignores the override.</li></ul><p>Notes: - Rclone ignores reserved <code>x-ms-*</code> keys (except<code>x-ms-tags</code>) for user metadata.</p>-<h3 id="performance-1">Performance</h3>+<h3 id="hierarchical-namespace">Hierarchical namespace</h3>+<p>Storage accounts with hierarchical namespace enabled (Azure Data Lake+Storage Gen2) have real directories. On these accounts rclone supports+empty directories, creating them with <code>rclone mkdir</code> and+removing them with <code>rclone rmdir</code>, whatever the+<code>--azureblob-directory-markers</code> setting.</p>+<p>Note that deleting the last file in a directory leaves the directory+behind. Use <code>rclone rmdirs</code> or+<code>--delete-empty-src-dirs</code> to remove empty directories.</p>+<p>Rclone reads whether the storage account has a hierarchical namespace+when it first uses the remote and saves it as the <code>hns</code>+option in the config file. Set <code>--azureblob-hns</code> to override+this.</p>+<p>Blob tags (<code>x-ms-tags</code> metadata) aren't supported on these+accounts.</p>+<h3 id="performance-2">Performance</h3><p>When uploading large files, increasing the value of<code>--azureblob-upload-concurrency</code> will increase performance atthe cost of using more memory. The default of 16 is set quiteconservatively to use less memory. It maybe be necessary raise it to 64or higher to fully utilize a 1 GBit/s link with a single filetransfer.</p>-<h3 id="restricted-filename-characters-22">Restricted filename+<h3 id="restricted-filename-characters-25">Restricted filenamecharacters</h3><p>In addition to the <ahref="https://rclone.org/overview/#restricted-characters">default@@ -53598,14 +56743,14 @@ must be explicitly specified using exactly one of the<code>msi_object_id</code>, <code>msi_client_id</code>, or<code>msi_mi_res_id</code> parameters.</p><p>If none of <code>msi_object_id</code>, <code>msi_client_id</code>, or-<code>msi_mi_res_id</code> is set, this is is equivalent to using+<code>msi_mi_res_id</code> is set, this is equivalent to using<code>env_auth</code>.</p><h4 id="federated-identity-credentials">Federated IdentityCredentials</h4><p>If these variables are set, rclone will authenticate with federatedidentity.</p><ul>-<li><code>tenant</code>: tenant ID of of the storage</li>+<li><code>tenant</code>: tenant ID of the storage</li><li><code>client_id</code>: client ID of the application the user willauthenticate to storage</li><li><code>msi_client_id</code>: managed identity client ID of the@@ -53629,7 +56774,7 @@ config:</p><pre class="console"><code>rclone lsf :azureblob,account=ACCOUNT:CONTAINER</code></pre><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/azureblob/azureblob.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-40">Standard options</h3>+<h3 id="standard-options-42">Standard options</h3><p>Here are the Standard options specific to azureblob (Microsoft AzureBlob Storage).</p><h4 id="azureblob-account">--azureblob-account</h4>@@ -53750,7 +56895,7 @@ obscure</a>.</p><li>Type: string</li><li>Required: false</li></ul>-<h3 id="advanced-options-41">Advanced options</h3>+<h3 id="advanced-options-43">Advanced options</h3><p>Here are the Advanced options specific to azureblob (Microsoft AzureBlob Storage).</p><h4@@ -54037,6 +57182,44 @@ avoid the time out.</p><li>Type: int</li><li>Default: 5000</li></ul>+<h4 id="azureblob-use-arrow-list">--azureblob-use-arrow-list</h4>+<p>Use the Apache Arrow listing format.</p>+<p>If set, directory listings are fetched using the ListBlobs Apache+Arrow response format instead of XML. Arrow responses are smaller and+much cheaper to parse, making listings of large containers several times+faster. Combine with "list_parallelism" for the biggest gains.</p>+<p>"Blob Listing with Apache Arrow" is in public preview at Microsoft+and is only supported on flat namespace accounts. On accounts with a+hierarchical namespace (ADLS Gen2), or where the feature is otherwise+unavailable, the server returns XML and the listing transparently falls+back to the normal XML path (logged at debug level).</p>+<p>Properties:</p>+<ul>+<li>Config: use_arrow_list</li>+<li>Env Var: RCLONE_AZUREBLOB_USE_ARROW_LIST</li>+<li>Type: bool</li>+<li>Default: false</li>+</ul>+<h4 id="azureblob-list-parallelism">--azureblob-list-parallelism</h4>+<p>Number of parallel shards to list a directory with.</p>+<p>If set greater than 1, the blob name keyspace of each directory is+split into ranges which are listed concurrently using the Arrow+startFrom/endBefore range parameters. This can dramatically speed up+listing containers with millions of objects, for both recursive (ListR)+and single directory listings. Speed keeps improving up to a parallelism+of around 30.</p>+<p>This has no effect unless "use_arrow_list" is also set, as Arrow is+the only listing path that supports server-side name ranges. If the+account does not support range listing (e.g. it has a hierarchical+namespace) the listing falls back to sequential. The default of 0 (or 1)+lists sequentially.</p>+<p>Properties:</p>+<ul>+<li>Config: list_parallelism</li>+<li>Env Var: RCLONE_AZUREBLOB_LIST_PARALLELISM</li>+<li>Type: int</li>+<li>Default: 0</li>+</ul><h4 id="azureblob-access-tier">--azureblob-access-tier</h4><p>Access tier of blob: hot, cool, cold or archive.</p><p>Archived blobs can be restored by setting access tier to hot, cool or@@ -54132,6 +57315,8 @@ created</p>creates an empty object ending with "/", to persist the folder.</p><p>This object also has the metadata "hdi_isfolder = true" to conform tothe Microsoft standard.</p>+<p>This is not needed on storage accounts with a hierarchical namespace+(see the hns option) as these always support empty directories.</p><p>Properties:</p><ul><li>Config: directory_markers</li>@@ -54139,6 +57324,19 @@ the Microsoft standard.</p><li>Type: bool</li><li>Default: false</li></ul>+<h4 id="azureblob-hns">--azureblob-hns</h4>+<p>Set if the storage account has a hierarchical namespace.</p>+<p>Storage accounts with hierarchical namespace enabled (Azure Data Lake+Storage Gen2) have real directories which can be empty.</p>+<p>If this is unset then rclone will read it from the storage account+and save it in the config file.</p>+<p>Properties:</p>+<ul>+<li>Config: hns</li>+<li>Env Var: RCLONE_AZUREBLOB_HNS</li>+<li>Type: Tristate</li>+<li>Default: unset</li>+</ul><h4id="azureblob-no-check-container">--azureblob-no-check-container</h4><p>If set, don't attempt to check the container exists or create it.</p>@@ -54300,7 +57498,7 @@ for more info.</p></ul><p>Eg <code>--header-upload "Content-Type: text/potato"</code> or<code>--header-upload "X-MS-Tags: foo=bar"</code>.</p>-<h2 id="limitations-30">Limitations</h2>+<h2 id="limitations-31">Limitations</h2><p>MD5 sums are only uploaded with chunked files if the source has anMD5 sum. This will always be the case for a local to azure copy.</p><p><code>rclone about</code> is not supported by the Microsoft Azure@@ -54330,7 +57528,7 @@ in the advanced settings, setting it toStorage</h1><p>Paths are specified as <code>remote:</code> You may putsubdirectories in too, e.g. <code>remote:path/to/dir</code>.</p>-<h2 id="configuration-41">Configuration</h2>+<h2 id="configuration-44">Configuration</h2><p>Here is an example of making a Microsoft Azure Files Storageconfiguration. For a remote called <code>remote</code>. First run:</p><pre class="console"><code>rclone config</code></pre>@@ -54411,14 +57609,14 @@ deleting any excess files in the directory.</p><h3 id="modified-time">Modified time</h3><p>The modified time is stored as the SMB <code>LastWriteTime</code>property on files. It has a precision of 100 ns.</p>-<h3 id="performance-2">Performance</h3>+<h3 id="performance-3">Performance</h3><p>When uploading large files, increasing the value of<code>--azurefiles-upload-concurrency</code> will increase performanceat the cost of using more memory. The default of 16 is set quiteconservatively to use less memory. It maybe be necessary raise it to 64or higher to fully utilize a 1 GBit/s link with a single filetransfer.</p>-<h3 id="restricted-filename-characters-23">Restricted filename+<h3 id="restricted-filename-characters-26">Restricted filenamecharacters</h3><p>In addition to the <ahref="https://rclone.org/overview/#restricted-characters">default@@ -54666,14 +57864,14 @@ must be explicitly specified using exactly one of the<code>msi_object_id</code>, <code>msi_client_id</code>, or<code>msi_mi_res_id</code> parameters.</p><p>If none of <code>msi_object_id</code>, <code>msi_client_id</code>, or-<code>msi_mi_res_id</code> is set, this is is equivalent to using+<code>msi_mi_res_id</code> is set, this is equivalent to using<code>env_auth</code>.</p><h4 id="federated-identity-credentials-1">Federated IdentityCredentials</h4><p>If these variables are set, rclone will authenticate with federatedidentity.</p><ul>-<li><code>tenant</code>: tenant ID of of the storage</li>+<li><code>tenant</code>: tenant ID of the storage</li><li><code>client_id</code>: client ID of the application the user willauthenticate to storage</li><li><code>msi_client_id</code>: managed identity client ID of the@@ -54692,7 +57890,7 @@ use.</p><p>Don't set <code>env_auth</code> at the same time.</p><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/azurefiles/azurefiles.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-41">Standard options</h3>+<h3 id="standard-options-43">Standard options</h3><p>Here are the Standard options specific to azurefiles (Microsoft AzureFiles).</p><h4 id="azurefiles-account">--azurefiles-account</h4>@@ -54824,7 +58022,7 @@ obscure</a>.</p><li>Type: string</li><li>Required: false</li></ul>-<h3 id="advanced-options-42">Advanced options</h3>+<h3 id="advanced-options-44">Advanced options</h3><p>Here are the Advanced options specific to azurefiles (Microsoft AzureFiles).</p><h4@@ -54996,7 +58194,7 @@ with a System Managed Identity that you do not want to use.</p><h4 id="azurefiles-chunk-size">--azurefiles-chunk-size</h4><p>Upload chunk size.</p><p>Note that this is stored in memory and there may be up to-"--transfers" * "--azurefile-upload-concurrency" chunks stored at once+"--transfers" * "--azurefiles-upload-concurrency" chunks stored at oncein memory.</p><p>Properties:</p><ul>@@ -55014,7 +58212,7 @@ concurrently.</p>links and these uploads do not fully utilize your bandwidth, thenincreasing this may help to speed up the transfers.</p><p>Note that chunks are stored in memory and there may be up to-"--transfers" * "--azurefile-upload-concurrency" chunks stored at once+"--transfers" * "--azurefiles-upload-concurrency" chunks stored at oncein memory.</p><p>Properties:</p><ul>@@ -55077,14 +58275,17 @@ Slash,LtGt,DoubleQuote,Colon,Question,Asterisk,Pipe,BackSlash,Del,Ctl,RightPerio<li>Content-Type</li></ul><p>Eg <code>--header-upload "Content-Type: text/potato"</code></p>-<h2 id="limitations-31">Limitations</h2>+<h2 id="limitations-32">Limitations</h2><p>MD5 sums are only uploaded with chunked files if the source has anMD5 sum. This will always be the case for a local to azure copy.</p><h1 id="microsoft-onedrive">Microsoft OneDrive</h1><p>Paths are specified as <code>remote:path</code></p><p>Paths may be as deep as required, e.g.<code>remote:directory/subdirectory</code>.</p>-<h2 id="configuration-42">Configuration</h2>+<p><strong>Note</strong>: OneDrive for Business and SharePoint users+need rclone v1.76 or later from April 1, 2027 - see <a+href="#pre-authenticated-urls">Pre-authenticated URLs</a>.</p>+<h2 id="configuration-45">Configuration</h2><p>The initial setup for OneDrive involves getting a token fromMicrosoft which you need to do in your browser.<code>rclone config</code> walks you through it.</p>@@ -55310,7 +58511,7 @@ Business OneDrive</h3><p>If you do not have admin access to your organization's OneDrive forBusiness, you can still connect by manually providing the SharePointtenant URL and drive ID. This works by overriding the base API URL from-the standard Microsoft Graph endpoint to the SharePoint v2.0+the standard Microsoft Graph endpoint to the SharePoint APIendpoint.</p><h4 id="steps-to-manually-obtain-credentials">Steps to manually obtaincredentials</h4>@@ -55322,22 +58523,32 @@ credentials</h4><li><p>Search for <code>driveAccessToken</code> in the networkrequests.</p></li><li><p>Extract the following information from the response:</p>-<div class="sourceCode" id="cb1579"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb1579-1"><a href="#cb1579-1" aria-hidden="true" tabindex="-1"></a><span class="er">".driveUrl":</span> <span class="er">"</span><span class="fu">{</span><span class="er">tenant_url</span><span class="fu">}</span><span class="er">/v2.0/drives/</span><span class="fu">{</span><span class="er">drive_id</span><span class="fu">}</span><span class="er">",</span></span>-<span id="cb1579-2"><a href="#cb1579-2" aria-hidden="true" tabindex="-1"></a><span class="er">".driveAccessToken":</span> <span class="er">"access_token=</span><span class="fu">{</span><span class="er">access_token</span><span class="fu">}</span><span class="er">"</span></span></code></pre></div></li>+<div class="sourceCode" id="cb1644"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb1644-1"><a href="#cb1644-1" aria-hidden="true" tabindex="-1"></a><span class="er">".driveUrl":</span> <span class="er">"</span><span class="fu">{</span><span class="er">tenant_url</span><span class="fu">}</span><span class="er">/v2.0/drives/</span><span class="fu">{</span><span class="er">drive_id</span><span class="fu">}</span><span class="er">",</span></span>+<span id="cb1644-2"><a href="#cb1644-2" aria-hidden="true" tabindex="-1"></a><span class="er">".driveAccessToken":</span> <span class="er">"access_token=</span><span class="fu">{</span><span class="er">access_token</span><span class="fu">}</span><span class="er">"</span></span></code></pre></div>+<p>In some guest access scenarios SharePoint may also expose a matching+v2.1 URL and token pair:</p>+<div class="sourceCode" id="cb1645"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb1645-1"><a href="#cb1645-1" aria-hidden="true" tabindex="-1"></a><span class="er">".driveUrlV21":</span> <span class="er">"</span><span class="fu">{</span><span class="er">tenant_url</span><span class="fu">}</span><span class="er">/v2.1/drives/</span><span class="fu">{</span><span class="er">drive_id</span><span class="fu">}</span><span class="er">",</span></span>+<span id="cb1645-2"><a href="#cb1645-2" aria-hidden="true" tabindex="-1"></a><span class="er">".driveAccessTokenV21":</span> <span class="er">"access_token=</span><span class="fu">{</span><span class="er">access_token</span><span class="fu">}</span><span class="er">"</span></span></code></pre></div></li></ol><h4 id="rclone-configuration">Rclone configuration</h4><p>Use the extracted values to configure your remote:</p>-<div class="sourceCode" id="cb1580"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1580-1"><a href="#cb1580-1" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> onedrive</span></span>-<span id="cb1580-2"><a href="#cb1580-2" aria-hidden="true" tabindex="-1"></a><span class="dt">token </span><span class="ot">=</span><span class="st"> {"access_token":"{access_token}","token_type":"Bearer","refresh_token":"","expiry":"2045-12-31T23:59:59Z"}</span></span>-<span id="cb1580-3"><a href="#cb1580-3" aria-hidden="true" tabindex="-1"></a><span class="dt">drive_id </span><span class="ot">=</span><span class="st"> {drive_id}</span></span>-<span id="cb1580-4"><a href="#cb1580-4" aria-hidden="true" tabindex="-1"></a><span class="dt">tenant_url </span><span class="ot">=</span><span class="st"> {tenant_url}</span></span>-<span id="cb1580-5"><a href="#cb1580-5" aria-hidden="true" tabindex="-1"></a><span class="dt">drive_type </span><span class="ot">=</span><span class="st"> business</span></span></code></pre></div>+<div class="sourceCode" id="cb1646"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1646-1"><a href="#cb1646-1" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> onedrive</span></span>+<span id="cb1646-2"><a href="#cb1646-2" aria-hidden="true" tabindex="-1"></a><span class="dt">token </span><span class="ot">=</span><span class="st"> {"access_token":"{access_token}","token_type":"Bearer","refresh_token":"","expiry":"2045-12-31T23:59:59Z"}</span></span>+<span id="cb1646-3"><a href="#cb1646-3" aria-hidden="true" tabindex="-1"></a><span class="dt">drive_id </span><span class="ot">=</span><span class="st"> {drive_id}</span></span>+<span id="cb1646-4"><a href="#cb1646-4" aria-hidden="true" tabindex="-1"></a><span class="dt">tenant_url </span><span class="ot">=</span><span class="st"> {tenant_url}</span></span>+<span id="cb1646-5"><a href="#cb1646-5" aria-hidden="true" tabindex="-1"></a><span class="dt">drive_type </span><span class="ot">=</span><span class="st"> business</span></span></code></pre></div>+<p>If you use a v2.1 token, set the matching SharePoint API version:</p>+<div class="sourceCode" id="cb1647"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1647-1"><a href="#cb1647-1" aria-hidden="true" tabindex="-1"></a><span class="dt">tenant_api_version </span><span class="ot">=</span><span class="st"> v2.1</span></span></code></pre></div>+<p>If <code>tenant_api_version</code> is omitted, rclone defaults to+<code>v2.0</code>.</p><p>Since the exact expiry time cannot be determined from web traffic,set the expiry to a future date. Note that the token will eventuallyexpire and you will need to repeat the process to obtain a new one.</p>-<h3 id="modification-times-and-hashes-20">Modification times and+<h3 id="modification-times-and-hashes-22">Modification times andhashes</h3><p>OneDrive allows modification times to be set on objects accurate to 1second. These will be used to detect whether objects need syncing or@@ -55372,15 +58583,15 @@ recommended way of reading all the file information from a drive.</p>href="https://rclone.org/rc/#vfs-refresh">rclone rc vfs/refreshrecursive=true</a>) to very quickly fill the mount with informationabout all the files.</p>-<p>The API used for the recursive listing (<code>ListR</code>) only-supports listing from the root of the drive. This will become-increasingly inefficient the further away you get from the root as-rclone will have to discard files outside of the directory you are-using.</p>+<p>Rclone asks for the delta listing of the directory being listed. Some+drives only support delta listings from the root of the drive, in which+case rclone lists the whole drive and discards the files outside of the+directory you are using, which is increasingly inefficient the further+away from the root you get.</p><p>Some commands (like <code>rclone lsf -R</code>) will use<code>ListR</code> by default - you can turn this off with<code>--disable ListR</code> if you need to.</p>-<h3 id="restricted-filename-characters-24">Restricted filename+<h3 id="restricted-filename-characters-27">Restricted filenamecharacters</h3><p>In addition to the <ahref="https://rclone.org/overview/#restricted-characters">default@@ -55486,14 +58697,14 @@ only get replaced if they are the first character in the name:</p><p>Invalid UTF-8 bytes will also be <ahref="https://rclone.org/overview/#invalid-utf8">replaced</a>, as theycan't be used in JSON strings.</p>-<h3 id="deleting-files-4">Deleting files</h3>+<h3 id="deleting-files-5">Deleting files</h3><p>Any files you delete with rclone will end up in the trash. Microsoftdoesn't provide an API to permanently delete files, nor to empty thetrash, so you will have to do that with one of Microsoft's apps or viathe OneDrive website.</p><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/onedrive/onedrive.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-42">Standard options</h3>+<h3 id="standard-options-44">Standard options</h3><p>Here are the Standard options specific to onedrive (MicrosoftOneDrive).</p><h4 id="onedrive-client-id">--onedrive-client-id</h4>@@ -55555,7 +58766,7 @@ ID.</p><li>Type: string</li><li>Required: false</li></ul>-<h3 id="advanced-options-43">Advanced options</h3>+<h3 id="advanced-options-45">Advanced options</h3><p>Here are the Advanced options specific to onedrive (MicrosoftOneDrive).</p><h4 id="onedrive-token">--onedrive-token</h4>@@ -55617,7 +58828,7 @@ new version.</p></ul><h4 id="onedrive-tenant-url">--onedrive-tenant-url</h4><p>The tenant URL for non-admin OneDrive access.</p>-<p>Set this to your SharePoint tenant URL to use the SharePoint v2.0 API+<p>Set this to your SharePoint tenant URL to use the SharePoint APIendpoint instead of the standard Microsoft Graph API. This allowsaccessing business OneDrive without admin consent.</p><p>The URL can be found in your browser's developer tools by searching@@ -55631,6 +58842,19 @@ field which contains the tenant URL and drive ID.</p><li>Type: string</li><li>Required: false</li></ul>+<h4 id="onedrive-tenant-api-version">--onedrive-tenant-api-version</h4>+<p>The SharePoint API version to use with tenant_url.</p>+<p>Set this to the SharePoint API version matching the browser-extracted+access token. For example, use v2.1 with a driveAccessTokenV21+token.</p>+<p>This only applies when tenant_url is set.</p>+<p>Properties:</p>+<ul>+<li>Config: tenant_api_version</li>+<li>Env Var: RCLONE_ONEDRIVE_TENANT_API_VERSION</li>+<li>Type: string</li>+<li>Default: "v2.0"</li>+</ul><h4 id="onedrive-chunk-size">--onedrive-chunk-size</h4><p>Chunk size to upload files with - must be multiple of 320k (327,680bytes).</p>@@ -55753,13 +58977,11 @@ will be slightly slower).</p></ul><h4 id="onedrive-no-versions">--onedrive-no-versions</h4><p>Remove all versions on modifying operations.</p>-<p>Onedrive for business creates versions when rclone uploads new files-overwriting an existing one and when it sets the modification time.</p>+<p>Onedrive creates versions when rclone uploads new files overwriting+an existing one and when it sets the modification time.</p><p>These versions take up space out of the quota.</p><p>This flag checks for versions after file upload and settingmodification time and removes all but the last version.</p>-<p><strong>NB</strong> Onedrive personal can't currently delete versions-so don't use this flag there.</p><p>Properties:</p><ul><li>Config: no_versions</li>@@ -55771,9 +58993,8 @@ so don't use this flag there.</p><p>Permanently delete files on removal.</p><p>Normally files will get sent to the recycle bin on deletion. Settingthis flag causes them to be permanently deleted. Use with care.</p>-<p>OneDrive personal accounts do not support the permanentDelete API, it-only applies to OneDrive for Business and SharePoint document-libraries.</p>+<p>This works with OneDrive for Business, SharePoint document libraries,+and OneDrive personal accounts, including free accounts.</p><p>Properties:</p><ul><li>Config: hard_delete</li>@@ -55826,13 +59047,16 @@ get access.</li><li>"embed"<ul><li>Creates an embeddable link to the item.</li>+<li>Only available in OneDrive personal.</li></ul></li></ul></li></ul><h4 id="onedrive-link-password">--onedrive-link-password</h4><p>Set the password for links created by the link command.</p>-<p>At the time of writing this only works with OneDrive personal paid-accounts.</p>+<p>At the time of writing this works with OneDrive for Business and+OneDrive personal paid accounts.</p>+<p>OneDrive personal free accounts can't set a password or an expiry+time (with --expire) on links.</p><p>Properties:</p><ul><li>Config: link_password</li>@@ -55922,20 +59146,11 @@ listings.</p><pre><code>rclone lsf -R onedrive:rclone size onedrive:rclone rc vfs/refresh recursive=true</code></pre>-<p><strong>However</strong> the delta listing API <strong>only</strong>-works at the root of the drive. If you use it not at the root then it-recurses from the root and discards all the data that is not under the-directory you asked for. So it will be correct but may not be very-efficient.</p>-<p>This is why this flag is not set as the default.</p>-<p>As a rule of thumb if nearly all of your data is under rclone's root-directory (the <code>root/directory</code> in-<code>onedrive:root/directory</code>) then using this flag will be a big-performance win. If your data is mostly not under the root then using-this flag will be a big performance loss.</p>-<p>It is recommended if you are mounting your onedrive at the root (or-near the root when using crypt) and using rclone-<code>rc vfs/refresh</code>.</p>+<p>Rclone asks for the delta listing of the directory being listed. If+the drive only supports delta listings at the root of the drive (as some+older OneDrive for Business and SharePoint drives do) then rclone lists+from the root and discards all the data that is not under the directory+you asked for. So it will be correct but may not be very efficient.</p><p>Properties:</p><ul><li>Config: delta</li>@@ -56020,69 +59235,69 @@ href="https://learn.microsoft.com/en-us/onedrive/developer/rest-api/resources/peAPI</a>, which differs slightly between OneDrive Personal andBusiness.</p><p>Example for OneDrive Personal:</p>-<div class="sourceCode" id="cb1583"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb1583-1"><a href="#cb1583-1" aria-hidden="true" tabindex="-1"></a><span class="ot">[</span></span>-<span id="cb1583-2"><a href="#cb1583-2" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>-<span id="cb1583-3"><a href="#cb1583-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"id"</span><span class="fu">:</span> <span class="st">"1234567890ABC!123"</span><span class="fu">,</span></span>-<span id="cb1583-4"><a href="#cb1583-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"grantedTo"</span><span class="fu">:</span> <span class="fu">{</span></span>-<span id="cb1583-5"><a href="#cb1583-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"user"</span><span class="fu">:</span> <span class="fu">{</span></span>-<span id="cb1583-6"><a href="#cb1583-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">"id"</span><span class="fu">:</span> <span class="st">"[email protected]"</span></span>-<span id="cb1583-7"><a href="#cb1583-7" aria-hidden="true" tabindex="-1"></a> <span class="fu">},</span></span>-<span id="cb1583-8"><a href="#cb1583-8" aria-hidden="true" tabindex="-1"></a> <span class="dt">"application"</span><span class="fu">:</span> <span class="fu">{},</span></span>-<span id="cb1583-9"><a href="#cb1583-9" aria-hidden="true" tabindex="-1"></a> <span class="dt">"device"</span><span class="fu">:</span> <span class="fu">{}</span></span>-<span id="cb1583-10"><a href="#cb1583-10" aria-hidden="true" tabindex="-1"></a> <span class="fu">},</span></span>-<span id="cb1583-11"><a href="#cb1583-11" aria-hidden="true" tabindex="-1"></a> <span class="dt">"invitation"</span><span class="fu">:</span> <span class="fu">{</span></span>-<span id="cb1583-12"><a href="#cb1583-12" aria-hidden="true" tabindex="-1"></a> <span class="dt">"email"</span><span class="fu">:</span> <span class="st">"[email protected]"</span></span>-<span id="cb1583-13"><a href="#cb1583-13" aria-hidden="true" tabindex="-1"></a> <span class="fu">},</span></span>-<span id="cb1583-14"><a href="#cb1583-14" aria-hidden="true" tabindex="-1"></a> <span class="dt">"link"</span><span class="fu">:</span> <span class="fu">{</span></span>-<span id="cb1583-15"><a href="#cb1583-15" aria-hidden="true" tabindex="-1"></a> <span class="dt">"webUrl"</span><span class="fu">:</span> <span class="st">"https://1drv.ms/t/s!1234567890ABC"</span></span>-<span id="cb1583-16"><a href="#cb1583-16" aria-hidden="true" tabindex="-1"></a> <span class="fu">},</span></span>-<span id="cb1583-17"><a href="#cb1583-17" aria-hidden="true" tabindex="-1"></a> <span class="dt">"roles"</span><span class="fu">:</span> <span class="ot">[</span></span>-<span id="cb1583-18"><a href="#cb1583-18" aria-hidden="true" tabindex="-1"></a> <span class="st">"read"</span></span>-<span id="cb1583-19"><a href="#cb1583-19" aria-hidden="true" tabindex="-1"></a> <span class="ot">]</span><span class="fu">,</span></span>-<span id="cb1583-20"><a href="#cb1583-20" aria-hidden="true" tabindex="-1"></a> <span class="dt">"shareId"</span><span class="fu">:</span> <span class="st">"s!1234567890ABC"</span></span>-<span id="cb1583-21"><a href="#cb1583-21" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>-<span id="cb1583-22"><a href="#cb1583-22" aria-hidden="true" tabindex="-1"></a><span class="ot">]</span></span></code></pre></div>+<div class="sourceCode" id="cb1650"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb1650-1"><a href="#cb1650-1" aria-hidden="true" tabindex="-1"></a><span class="ot">[</span></span>+<span id="cb1650-2"><a href="#cb1650-2" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>+<span id="cb1650-3"><a href="#cb1650-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"id"</span><span class="fu">:</span> <span class="st">"1234567890ABC!123"</span><span class="fu">,</span></span>+<span id="cb1650-4"><a href="#cb1650-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"grantedTo"</span><span class="fu">:</span> <span class="fu">{</span></span>+<span id="cb1650-5"><a href="#cb1650-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"user"</span><span class="fu">:</span> <span class="fu">{</span></span>+<span id="cb1650-6"><a href="#cb1650-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">"id"</span><span class="fu">:</span> <span class="st">"[email protected]"</span></span>+<span id="cb1650-7"><a href="#cb1650-7" aria-hidden="true" tabindex="-1"></a> <span class="fu">},</span></span>+<span id="cb1650-8"><a href="#cb1650-8" aria-hidden="true" tabindex="-1"></a> <span class="dt">"application"</span><span class="fu">:</span> <span class="fu">{},</span></span>+<span id="cb1650-9"><a href="#cb1650-9" aria-hidden="true" tabindex="-1"></a> <span class="dt">"device"</span><span class="fu">:</span> <span class="fu">{}</span></span>+<span id="cb1650-10"><a href="#cb1650-10" aria-hidden="true" tabindex="-1"></a> <span class="fu">},</span></span>+<span id="cb1650-11"><a href="#cb1650-11" aria-hidden="true" tabindex="-1"></a> <span class="dt">"invitation"</span><span class="fu">:</span> <span class="fu">{</span></span>+<span id="cb1650-12"><a href="#cb1650-12" aria-hidden="true" tabindex="-1"></a> <span class="dt">"email"</span><span class="fu">:</span> <span class="st">"[email protected]"</span></span>+<span id="cb1650-13"><a href="#cb1650-13" aria-hidden="true" tabindex="-1"></a> <span class="fu">},</span></span>+<span id="cb1650-14"><a href="#cb1650-14" aria-hidden="true" tabindex="-1"></a> <span class="dt">"link"</span><span class="fu">:</span> <span class="fu">{</span></span>+<span id="cb1650-15"><a href="#cb1650-15" aria-hidden="true" tabindex="-1"></a> <span class="dt">"webUrl"</span><span class="fu">:</span> <span class="st">"https://1drv.ms/t/s!1234567890ABC"</span></span>+<span id="cb1650-16"><a href="#cb1650-16" aria-hidden="true" tabindex="-1"></a> <span class="fu">},</span></span>+<span id="cb1650-17"><a href="#cb1650-17" aria-hidden="true" tabindex="-1"></a> <span class="dt">"roles"</span><span class="fu">:</span> <span class="ot">[</span></span>+<span id="cb1650-18"><a href="#cb1650-18" aria-hidden="true" tabindex="-1"></a> <span class="st">"read"</span></span>+<span id="cb1650-19"><a href="#cb1650-19" aria-hidden="true" tabindex="-1"></a> <span class="ot">]</span><span class="fu">,</span></span>+<span id="cb1650-20"><a href="#cb1650-20" aria-hidden="true" tabindex="-1"></a> <span class="dt">"shareId"</span><span class="fu">:</span> <span class="st">"s!1234567890ABC"</span></span>+<span id="cb1650-21"><a href="#cb1650-21" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>+<span id="cb1650-22"><a href="#cb1650-22" aria-hidden="true" tabindex="-1"></a><span class="ot">]</span></span></code></pre></div><p>Example for OneDrive Business:</p>-<div class="sourceCode" id="cb1584"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb1584-1"><a href="#cb1584-1" aria-hidden="true" tabindex="-1"></a><span class="ot">[</span></span>-<span id="cb1584-2"><a href="#cb1584-2" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>-<span id="cb1584-3"><a href="#cb1584-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"id"</span><span class="fu">:</span> <span class="st">"48d31887-5fad-4d73-a9f5-3c356e68a038"</span><span class="fu">,</span></span>-<span id="cb1584-4"><a href="#cb1584-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"grantedToIdentities"</span><span class="fu">:</span> <span class="ot">[</span></span>-<span id="cb1584-5"><a href="#cb1584-5" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>-<span id="cb1584-6"><a href="#cb1584-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">"user"</span><span class="fu">:</span> <span class="fu">{</span></span>-<span id="cb1584-7"><a href="#cb1584-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"displayName"</span><span class="fu">:</span> <span class="st">"[email protected]"</span></span>-<span id="cb1584-8"><a href="#cb1584-8" aria-hidden="true" tabindex="-1"></a> <span class="fu">},</span></span>-<span id="cb1584-9"><a href="#cb1584-9" aria-hidden="true" tabindex="-1"></a> <span class="dt">"application"</span><span class="fu">:</span> <span class="fu">{},</span></span>-<span id="cb1584-10"><a href="#cb1584-10" aria-hidden="true" tabindex="-1"></a> <span class="dt">"device"</span><span class="fu">:</span> <span class="fu">{}</span></span>-<span id="cb1584-11"><a href="#cb1584-11" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>-<span id="cb1584-12"><a href="#cb1584-12" aria-hidden="true" tabindex="-1"></a> <span class="ot">]</span><span class="fu">,</span></span>-<span id="cb1584-13"><a href="#cb1584-13" aria-hidden="true" tabindex="-1"></a> <span class="dt">"link"</span><span class="fu">:</span> <span class="fu">{</span></span>-<span id="cb1584-14"><a href="#cb1584-14" aria-hidden="true" tabindex="-1"></a> <span class="dt">"type"</span><span class="fu">:</span> <span class="st">"view"</span><span class="fu">,</span></span>-<span id="cb1584-15"><a href="#cb1584-15" aria-hidden="true" tabindex="-1"></a> <span class="dt">"scope"</span><span class="fu">:</span> <span class="st">"users"</span><span class="fu">,</span></span>-<span id="cb1584-16"><a href="#cb1584-16" aria-hidden="true" tabindex="-1"></a> <span class="dt">"webUrl"</span><span class="fu">:</span> <span class="st">"https://contoso.sharepoint.com/:w:/t/design/a577ghg9hgh737613bmbjf839026561fmzhsr85ng9f3hjck2t5s"</span></span>-<span id="cb1584-17"><a href="#cb1584-17" aria-hidden="true" tabindex="-1"></a> <span class="fu">},</span></span>-<span id="cb1584-18"><a href="#cb1584-18" aria-hidden="true" tabindex="-1"></a> <span class="dt">"roles"</span><span class="fu">:</span> <span class="ot">[</span></span>-<span id="cb1584-19"><a href="#cb1584-19" aria-hidden="true" tabindex="-1"></a> <span class="st">"read"</span></span>-<span id="cb1584-20"><a href="#cb1584-20" aria-hidden="true" tabindex="-1"></a> <span class="ot">]</span><span class="fu">,</span></span>-<span id="cb1584-21"><a href="#cb1584-21" aria-hidden="true" tabindex="-1"></a> <span class="dt">"shareId"</span><span class="fu">:</span> <span class="st">"u!LKj1lkdlals90j1nlkascl"</span></span>-<span id="cb1584-22"><a href="#cb1584-22" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span><span class="ot">,</span></span>-<span id="cb1584-23"><a href="#cb1584-23" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>-<span id="cb1584-24"><a href="#cb1584-24" aria-hidden="true" tabindex="-1"></a> <span class="dt">"id"</span><span class="fu">:</span> <span class="st">"5D33DD65C6932946"</span><span class="fu">,</span></span>-<span id="cb1584-25"><a href="#cb1584-25" aria-hidden="true" tabindex="-1"></a> <span class="dt">"grantedTo"</span><span class="fu">:</span> <span class="fu">{</span></span>-<span id="cb1584-26"><a href="#cb1584-26" aria-hidden="true" tabindex="-1"></a> <span class="dt">"user"</span><span class="fu">:</span> <span class="fu">{</span></span>-<span id="cb1584-27"><a href="#cb1584-27" aria-hidden="true" tabindex="-1"></a> <span class="dt">"displayName"</span><span class="fu">:</span> <span class="st">"John Doe"</span><span class="fu">,</span></span>-<span id="cb1584-28"><a href="#cb1584-28" aria-hidden="true" tabindex="-1"></a> <span class="dt">"id"</span><span class="fu">:</span> <span class="st">"efee1b77-fb3b-4f65-99d6-274c11914d12"</span></span>-<span id="cb1584-29"><a href="#cb1584-29" aria-hidden="true" tabindex="-1"></a> <span class="fu">},</span></span>-<span id="cb1584-30"><a href="#cb1584-30" aria-hidden="true" tabindex="-1"></a> <span class="dt">"application"</span><span class="fu">:</span> <span class="fu">{},</span></span>-<span id="cb1584-31"><a href="#cb1584-31" aria-hidden="true" tabindex="-1"></a> <span class="dt">"device"</span><span class="fu">:</span> <span class="fu">{}</span></span>-<span id="cb1584-32"><a href="#cb1584-32" aria-hidden="true" tabindex="-1"></a> <span class="fu">},</span></span>-<span id="cb1584-33"><a href="#cb1584-33" aria-hidden="true" tabindex="-1"></a> <span class="dt">"roles"</span><span class="fu">:</span> <span class="ot">[</span></span>-<span id="cb1584-34"><a href="#cb1584-34" aria-hidden="true" tabindex="-1"></a> <span class="st">"owner"</span></span>-<span id="cb1584-35"><a href="#cb1584-35" aria-hidden="true" tabindex="-1"></a> <span class="ot">]</span><span class="fu">,</span></span>-<span id="cb1584-36"><a href="#cb1584-36" aria-hidden="true" tabindex="-1"></a> <span class="dt">"shareId"</span><span class="fu">:</span> <span class="st">"FWxc1lasfdbEAGM5fI7B67aB5ZMPDMmQ11U"</span></span>-<span id="cb1584-37"><a href="#cb1584-37" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>-<span id="cb1584-38"><a href="#cb1584-38" aria-hidden="true" tabindex="-1"></a><span class="ot">]</span></span></code></pre></div>+<div class="sourceCode" id="cb1651"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb1651-1"><a href="#cb1651-1" aria-hidden="true" tabindex="-1"></a><span class="ot">[</span></span>+<span id="cb1651-2"><a href="#cb1651-2" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>+<span id="cb1651-3"><a href="#cb1651-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"id"</span><span class="fu">:</span> <span class="st">"48d31887-5fad-4d73-a9f5-3c356e68a038"</span><span class="fu">,</span></span>+<span id="cb1651-4"><a href="#cb1651-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"grantedToIdentities"</span><span class="fu">:</span> <span class="ot">[</span></span>+<span id="cb1651-5"><a href="#cb1651-5" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>+<span id="cb1651-6"><a href="#cb1651-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">"user"</span><span class="fu">:</span> <span class="fu">{</span></span>+<span id="cb1651-7"><a href="#cb1651-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"displayName"</span><span class="fu">:</span> <span class="st">"[email protected]"</span></span>+<span id="cb1651-8"><a href="#cb1651-8" aria-hidden="true" tabindex="-1"></a> <span class="fu">},</span></span>+<span id="cb1651-9"><a href="#cb1651-9" aria-hidden="true" tabindex="-1"></a> <span class="dt">"application"</span><span class="fu">:</span> <span class="fu">{},</span></span>+<span id="cb1651-10"><a href="#cb1651-10" aria-hidden="true" tabindex="-1"></a> <span class="dt">"device"</span><span class="fu">:</span> <span class="fu">{}</span></span>+<span id="cb1651-11"><a href="#cb1651-11" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>+<span id="cb1651-12"><a href="#cb1651-12" aria-hidden="true" tabindex="-1"></a> <span class="ot">]</span><span class="fu">,</span></span>+<span id="cb1651-13"><a href="#cb1651-13" aria-hidden="true" tabindex="-1"></a> <span class="dt">"link"</span><span class="fu">:</span> <span class="fu">{</span></span>+<span id="cb1651-14"><a href="#cb1651-14" aria-hidden="true" tabindex="-1"></a> <span class="dt">"type"</span><span class="fu">:</span> <span class="st">"view"</span><span class="fu">,</span></span>+<span id="cb1651-15"><a href="#cb1651-15" aria-hidden="true" tabindex="-1"></a> <span class="dt">"scope"</span><span class="fu">:</span> <span class="st">"users"</span><span class="fu">,</span></span>+<span id="cb1651-16"><a href="#cb1651-16" aria-hidden="true" tabindex="-1"></a> <span class="dt">"webUrl"</span><span class="fu">:</span> <span class="st">"https://contoso.sharepoint.com/:w:/t/design/a577ghg9hgh737613bmbjf839026561fmzhsr85ng9f3hjck2t5s"</span></span>+<span id="cb1651-17"><a href="#cb1651-17" aria-hidden="true" tabindex="-1"></a> <span class="fu">},</span></span>+<span id="cb1651-18"><a href="#cb1651-18" aria-hidden="true" tabindex="-1"></a> <span class="dt">"roles"</span><span class="fu">:</span> <span class="ot">[</span></span>+<span id="cb1651-19"><a href="#cb1651-19" aria-hidden="true" tabindex="-1"></a> <span class="st">"read"</span></span>+<span id="cb1651-20"><a href="#cb1651-20" aria-hidden="true" tabindex="-1"></a> <span class="ot">]</span><span class="fu">,</span></span>+<span id="cb1651-21"><a href="#cb1651-21" aria-hidden="true" tabindex="-1"></a> <span class="dt">"shareId"</span><span class="fu">:</span> <span class="st">"u!LKj1lkdlals90j1nlkascl"</span></span>+<span id="cb1651-22"><a href="#cb1651-22" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span><span class="ot">,</span></span>+<span id="cb1651-23"><a href="#cb1651-23" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>+<span id="cb1651-24"><a href="#cb1651-24" aria-hidden="true" tabindex="-1"></a> <span class="dt">"id"</span><span class="fu">:</span> <span class="st">"5D33DD65C6932946"</span><span class="fu">,</span></span>+<span id="cb1651-25"><a href="#cb1651-25" aria-hidden="true" tabindex="-1"></a> <span class="dt">"grantedTo"</span><span class="fu">:</span> <span class="fu">{</span></span>+<span id="cb1651-26"><a href="#cb1651-26" aria-hidden="true" tabindex="-1"></a> <span class="dt">"user"</span><span class="fu">:</span> <span class="fu">{</span></span>+<span id="cb1651-27"><a href="#cb1651-27" aria-hidden="true" tabindex="-1"></a> <span class="dt">"displayName"</span><span class="fu">:</span> <span class="st">"John Doe"</span><span class="fu">,</span></span>+<span id="cb1651-28"><a href="#cb1651-28" aria-hidden="true" tabindex="-1"></a> <span class="dt">"id"</span><span class="fu">:</span> <span class="st">"efee1b77-fb3b-4f65-99d6-274c11914d12"</span></span>+<span id="cb1651-29"><a href="#cb1651-29" aria-hidden="true" tabindex="-1"></a> <span class="fu">},</span></span>+<span id="cb1651-30"><a href="#cb1651-30" aria-hidden="true" tabindex="-1"></a> <span class="dt">"application"</span><span class="fu">:</span> <span class="fu">{},</span></span>+<span id="cb1651-31"><a href="#cb1651-31" aria-hidden="true" tabindex="-1"></a> <span class="dt">"device"</span><span class="fu">:</span> <span class="fu">{}</span></span>+<span id="cb1651-32"><a href="#cb1651-32" aria-hidden="true" tabindex="-1"></a> <span class="fu">},</span></span>+<span id="cb1651-33"><a href="#cb1651-33" aria-hidden="true" tabindex="-1"></a> <span class="dt">"roles"</span><span class="fu">:</span> <span class="ot">[</span></span>+<span id="cb1651-34"><a href="#cb1651-34" aria-hidden="true" tabindex="-1"></a> <span class="st">"owner"</span></span>+<span id="cb1651-35"><a href="#cb1651-35" aria-hidden="true" tabindex="-1"></a> <span class="ot">]</span><span class="fu">,</span></span>+<span id="cb1651-36"><a href="#cb1651-36" aria-hidden="true" tabindex="-1"></a> <span class="dt">"shareId"</span><span class="fu">:</span> <span class="st">"FWxc1lasfdbEAGM5fI7B67aB5ZMPDMmQ11U"</span></span>+<span id="cb1651-37"><a href="#cb1651-37" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>+<span id="cb1651-38"><a href="#cb1651-38" aria-hidden="true" tabindex="-1"></a><span class="ot">]</span></span></code></pre></div><p>To write permissions, pass in a "permissions" metadata key using thissame format. The <ahref="https://rclone.org/docs/#metadata-mapper"><code>--metadata-mapper</code></a>@@ -56096,12 +59311,12 @@ for a user. Creating a Public Link is also supported, if<code>Link.Scope</code> is set to <code>"anonymous"</code>.</p><p>Example request to add a "read" permission with<code>--metadata-mapper</code>:</p>-<div class="sourceCode" id="cb1585"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb1585-1"><a href="#cb1585-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>-<span id="cb1585-2"><a href="#cb1585-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Metadata"</span><span class="fu">:</span> <span class="fu">{</span></span>-<span id="cb1585-3"><a href="#cb1585-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"permissions"</span><span class="fu">:</span> <span class="st">"[{</span><span class="ch">\"</span><span class="st">grantedToIdentities</span><span class="ch">\"</span><span class="st">:[{</span><span class="ch">\"</span><span class="st">user</span><span class="ch">\"</span><span class="st">:{</span><span class="ch">\"</span><span class="st">id</span><span class="ch">\"</span><span class="st">:</span><span class="ch">\"</span><span class="st">[email protected]</span><span class="ch">\"</span><span class="st">}}],</span><span class="ch">\"</span><span class="st">roles</span><span class="ch">\"</span><span class="st">:[</span><span class="ch">\"</span><span class="st">read</span><span class="ch">\"</span><span class="st">]}]"</span></span>-<span id="cb1585-4"><a href="#cb1585-4" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>-<span id="cb1585-5"><a href="#cb1585-5" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<div class="sourceCode" id="cb1652"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb1652-1"><a href="#cb1652-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb1652-2"><a href="#cb1652-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Metadata"</span><span class="fu">:</span> <span class="fu">{</span></span>+<span id="cb1652-3"><a href="#cb1652-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"permissions"</span><span class="fu">:</span> <span class="st">"[{</span><span class="ch">\"</span><span class="st">grantedToIdentities</span><span class="ch">\"</span><span class="st">:[{</span><span class="ch">\"</span><span class="st">user</span><span class="ch">\"</span><span class="st">:{</span><span class="ch">\"</span><span class="st">id</span><span class="ch">\"</span><span class="st">:</span><span class="ch">\"</span><span class="st">[email protected]</span><span class="ch">\"</span><span class="st">}}],</span><span class="ch">\"</span><span class="st">roles</span><span class="ch">\"</span><span class="st">:[</span><span class="ch">\"</span><span class="st">read</span><span class="ch">\"</span><span class="st">]}]"</span></span>+<span id="cb1652-4"><a href="#cb1652-4" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>+<span id="cb1652-5"><a href="#cb1652-5" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div><p>Note that adding a permission can fail if a conflicting permissionalready exists for the file/folder.</p><p>To update an existing permission, include both the Permission ID and@@ -56145,8 +59360,7 @@ backend.</p><tbody><tr class="odd"><td>btime</td>-<td>Time of file birth (creation) with S accuracy (mS for OneDrive-Personal).</td>+<td>Time of file birth (creation) with S accuracy.</td><td>RFC 3339</td><td>2006-01-02T15:04:05Z</td><td>N</td>@@ -56210,8 +59424,7 @@ supported by Microsoft.</td></tr><tr class="even"><td>mtime</td>-<td>Time of last modification with S accuracy (mS for OneDrive-Personal).</td>+<td>Time of last modification with S accuracy.</td><td>RFC 3339</td><td>2006-01-02T15:04:05Z</td><td>N</td>@@ -56259,15 +59472,14 @@ organization, or users.</td></tr><tr class="even"><td>shared-time</td>-<td>Time when the item was shared, with S accuracy (mS for OneDrive-Personal).</td>+<td>Time when the item was shared, with S accuracy.</td><td>RFC 3339</td><td>2006-01-02T15:04:05Z</td><td><strong>Y</strong></td></tr><tr class="odd"><td>utime</td>-<td>Time of upload with S accuracy (mS for OneDrive Personal).</td>+<td>Time of upload with S accuracy.</td><td>RFC 3339</td><td>2006-01-02T15:04:05Z</td><td><strong>Y</strong></td>@@ -56309,7 +59521,7 @@ hopefully give you a message ofURL of the format<code>https://{tenant}-my.sharepoint.com/personal/{user_name_domain_tld}/Documents</code></p></li></ol>-<h2 id="limitations-32">Limitations</h2>+<h2 id="limitations-33">Limitations</h2><p>If you don't use rclone for 90 days the refresh token will expire.This will result in authorization problems. This is easy to fix byrunning the <code>rclone config reconnect remote:</code> command to get@@ -56356,10 +59568,6 @@ remove all old versions.</p><code>true</code> and rclone will remove versions after operations whichcreate new versions. This takes extra transactions so only enable it ifyou need it.</p>-<p><strong>Note</strong> At the time of writing Onedrive Personal-creates versions (but not for setting the modification time) but the API-for removing them returns "API not found" so cleanup and-<code>no_versions</code> should not be used on Onedrive Personal.</p><h3 id="disabling-versioning">Disabling versioning</h3><p>Starting October 2018, users will no longer be able to disableversioning by default. This is because Microsoft has brought an <a@@ -56411,9 +59619,33 @@ then querying each file for versions it can be quite slow. Rclone doeswhich is a great way to see what it would do.</p><pre class="text"><code>rclone cleanup --interactive remote:path/subdir # interactively remove all old version for path/subdirrclone cleanup remote:path/subdir # unconditionally remove all old version for path/subdir</code></pre>-<p><strong>NB</strong> Onedrive personal can't currently delete-versions</p><h2 id="troubleshooting-4">Troubleshooting</h2>+<h3 id="pre-authenticated-urls">Pre-authenticated URLs</h3>+<p>Microsoft is <a+href="https://devblogs.microsoft.com/microsoft365dev/pre-authenticated-tempauth-urls-are-retiring-move-to-microsoft-entra-id-tokens-before-april-1-2027/">retiring+pre-authenticated (tempauth) URLs</a> in OneDrive for Business and+SharePoint from April 1, 2027.</p>+<p><strong>Upgrade to rclone v1.76 or later before April 1,+2027.</strong> Earlier versions rely on pre-authenticated URLs, so after+that date their downloads, chunked uploads and server-side copies will+fail with OneDrive for Business and SharePoint. Rclone v1.76 and later+keep working with the existing configuration, apart from the client+credentials flow with a client secret (see below).</p>+<p>Rclone uses these URLs when downloading files, when uploading files+bigger than <code>--onedrive-upload-cutoff</code> in chunks and when+doing server-side copies. When SharePoint returns them without+pre-authentication rclone authenticates them with a SharePoint token+which it gets using the existing login, so no reconfiguration is needed.+OneDrive Personal is not affected.</p>+<p>If rclone can't get a SharePoint token, or SharePoint refuses it, you+will see the error+<code>SharePoint refused to authenticate a request</code>. In this case+rclone uploads files smaller than 250 MiB in a single request instead+and copies files by downloading and uploading them, but bigger files+can't be uploaded and downloads may fail.</p>+<p>SharePoint does not accept tokens from the client credentials flow+which were got with a client secret, so this affects+<code>client_credentials</code> users.</p><h3 id="excessive-throttling-or-blocked-on-sharepoint">Excessivethrottling or blocked on SharePoint</h3><p>If you experience excessive throttling or is being blocked on@@ -56468,8 +59700,8 @@ your account. You can't do much about it, maybe write an email to youradmins.</p><p>However, there are other ways to interact with your OneDrive account.Have a look at the WebDAV backend: <a-href="https://rclone.org/webdav/#sharepoint"-class="uri">https://rclone.org/webdav/#sharepoint</a></p>+href="https://rclone.org/webdav/#sharepoint-online"+class="uri">https://rclone.org/webdav/#sharepoint-online</a></p><h3 id="invalid_grant-aadsts50076">invalid_grant (AADSTS50076)</h3><pre class="text"><code>Error: invalid_grantCode: AADSTS50076@@ -56511,6 +59743,22 @@ src="https://i.imgur.com/0S8H3li.png" title="Screenshot (My Files)"alt="in_my_files" /> <img src="https://i.imgur.com/2Iq66sW.png"title="Screenshot (rclone mount)" alt="rclone_mount" /></li></ol>+<p>Listing such a shortcut may still fail with</p>+<pre class="text"><code>invalidRequest: The provided drive id appears to be malformed, or does not represent a valid drive.</code></pre>+<p>as the drive it points to can't always be read with the drive ID the+API supplies. <code>Personal Vault</code> is a shortcut of the same kind+and fails to list in the same way, though it reports+<code>invalidResourceId: ObjectHandle is Invalid</code>.</p>+<p>A normal listing reports the error for that folder and carries on, so+the rest of the drive is still listed, and excluding the folder (for+example with <code>--exclude "/Personal Vault/**"</code>) avoids the+error.</p>+<p>A recursive listing with <code>--fast-list</code> stops at the first+error, so listing the root of a drive which has one of these shortcuts+in it will fail. Filters don't help here as the folder is listed by the+backend before the filters are applied - use+<code>--disable ListR</code> (or don't use <code>--fast-list</code>) on+such a drive.</p><h3id="live-photos-uploaded-from-ios-small-video-clips-in-.heic-files">LivePhotos uploaded from iOS (small video clips in .heic files)</h3>@@ -56552,7 +59800,7 @@ ERROR : 20230203_123826234_iOS.heic: vfs cache: failed to download: vfs reader:<p>Paths are specified as <code>remote:path</code></p><p>Paths may be as deep as required, e.g.<code>remote:directory/subdirectory</code>.</p>-<h2 id="configuration-43">Configuration</h2>+<h2 id="configuration-46">Configuration</h2><p>Here is an example of how to make a remote called<code>remote</code>. First run:</p><pre class="console"><code>rclone config</code></pre>@@ -56595,13 +59843,13 @@ y/e/d> y</code></pre><pre class="console"><code>rclone ls remote:</code></pre><p>To copy a local directory to an OpenDrive directory called backup</p><pre class="console"><code>rclone copy /home/source remote:backup</code></pre>-<h3 id="modification-times-and-hashes-21">Modification times and+<h3 id="modification-times-and-hashes-23">Modification times andhashes</h3><p>OpenDrive allows modification times to be set on objects accurate to1 second. These will be used to detect whether objects need syncing ornot.</p><p>The MD5 hash algorithm is supported.</p>-<h3 id="restricted-filename-characters-25">Restricted filename+<h3 id="restricted-filename-characters-28">Restricted filenamecharacters</h3><table><thead>@@ -56708,7 +59956,7 @@ href="https://rclone.org/overview/#invalid-utf8">replaced</a>, as theycan't be used in JSON strings.</p><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/opendrive/opendrive.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-43">Standard options</h3>+<h3 id="standard-options-45">Standard options</h3><p>Here are the Standard options specific to opendrive (OpenDrive).</p><h4 id="opendrive-username">--opendrive-username</h4><p>Username.</p>@@ -56731,7 +59979,7 @@ obscure</a>.</p><li>Type: string</li><li>Required: true</li></ul>-<h3 id="advanced-options-44">Advanced options</h3>+<h3 id="advanced-options-46">Advanced options</h3><p>Here are the Advanced options specific to opendrive (OpenDrive).</p><h4 id="opendrive-encoding">--opendrive-encoding</h4><p>The encoding for the backend.</p>@@ -56796,7 +60044,7 @@ access the contents</li><li>Required: false</li></ul><!-- autogenerated options stop -->-<h2 id="limitations-33">Limitations</h2>+<h2 id="limitations-34">Limitations</h2><p>Note that OpenDrive is case insensitive so you can't have a filecalled "Hello.doc" and one called "hello.doc".</p><p>There are quite a few characters that can't be in OpenDrive file@@ -56828,7 +60076,7 @@ subdirectories in too, e.g. <code>remote:bucket/path/to/dir</code>.</p><p>Sample command to transfer local artifacts to remote:bucket in oracleobject storage:</p><pre class="console"><code>rclone -vvv --progress --stats-one-line --max-stats-groups 10 --log-format date,time,UTC,longfile --fast-list --buffer-size 256Mi --oos-no-check-bucket --oos-upload-cutoff 10Mi --multi-thread-cutoff 16Mi --multi-thread-streams 3000 --transfers 3000 --checkers 64 --retries 2 --oos-chunk-size 10Mi --oos-upload-concurrency 10000 --oos-attempt-resume-upload --oos-leave-parts-on-error sync ./artifacts remote:bucket -vv</code></pre>-<h2 id="configuration-44">Configuration</h2>+<h2 id="configuration-47">Configuration</h2><p>Here is an example of making an oracle object storage configuration.<code>rclone config</code> walks you through it.</p><p>Here is an example of how to make a remote called@@ -56956,15 +60204,15 @@ No authentication</code></pre><h3 id="user-principal">User Principal</h3><p>Sample rclone config file for Authentication Provider UserPrincipal:</p>-<div class="sourceCode" id="cb1609"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1609-1"><a href="#cb1609-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[oos]</span></span>-<span id="cb1609-2"><a href="#cb1609-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> oracleobjectstorage</span></span>-<span id="cb1609-3"><a href="#cb1609-3" aria-hidden="true" tabindex="-1"></a><span class="dt">namespace </span><span class="ot">=</span><span class="st"> id<redacted>34</span></span>-<span id="cb1609-4"><a href="#cb1609-4" aria-hidden="true" tabindex="-1"></a><span class="dt">compartment </span><span class="ot">=</span><span class="st"> ocid1.compartment.oc1..aa<redacted>ba</span></span>-<span id="cb1609-5"><a href="#cb1609-5" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span><span class="st"> us-ashburn-1</span></span>-<span id="cb1609-6"><a href="#cb1609-6" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> user_principal_auth</span></span>-<span id="cb1609-7"><a href="#cb1609-7" aria-hidden="true" tabindex="-1"></a><span class="dt">config_file </span><span class="ot">=</span><span class="st"> /home/opc/.oci/config</span></span>-<span id="cb1609-8"><a href="#cb1609-8" aria-hidden="true" tabindex="-1"></a><span class="dt">config_profile </span><span class="ot">=</span><span class="st"> </span><span class="kw">Default</span></span></code></pre></div>+<div class="sourceCode" id="cb1677"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1677-1"><a href="#cb1677-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[oos]</span></span>+<span id="cb1677-2"><a href="#cb1677-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> oracleobjectstorage</span></span>+<span id="cb1677-3"><a href="#cb1677-3" aria-hidden="true" tabindex="-1"></a><span class="dt">namespace </span><span class="ot">=</span><span class="st"> id<redacted>34</span></span>+<span id="cb1677-4"><a href="#cb1677-4" aria-hidden="true" tabindex="-1"></a><span class="dt">compartment </span><span class="ot">=</span><span class="st"> ocid1.compartment.oc1..aa<redacted>ba</span></span>+<span id="cb1677-5"><a href="#cb1677-5" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span><span class="st"> us-ashburn-1</span></span>+<span id="cb1677-6"><a href="#cb1677-6" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> user_principal_auth</span></span>+<span id="cb1677-7"><a href="#cb1677-7" aria-hidden="true" tabindex="-1"></a><span class="dt">config_file </span><span class="ot">=</span><span class="st"> /home/opc/.oci/config</span></span>+<span id="cb1677-8"><a href="#cb1677-8" aria-hidden="true" tabindex="-1"></a><span class="dt">config_profile </span><span class="ot">=</span><span class="st"> </span><span class="kw">Default</span></span></code></pre></div><p>Advantages:</p><ul><li>One can use this method from any server within OCI or on-premises or@@ -57021,13 +60269,13 @@ export OCI_RESOURCE_PRINCIPAL_PRIVATE_PEM=/usr/share/model-server/key.pemexport OCI_RESOURCE_PRINCIPAL_RPST=/usr/share/model-server/security_token</code></pre><p>Sample rclone configuration file for Authentication Provider ResourcePrincipal:</p>-<div class="sourceCode" id="cb1612"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1612-1"><a href="#cb1612-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[oos]</span></span>-<span id="cb1612-2"><a href="#cb1612-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> oracleobjectstorage</span></span>-<span id="cb1612-3"><a href="#cb1612-3" aria-hidden="true" tabindex="-1"></a><span class="dt">namespace </span><span class="ot">=</span><span class="st"> id<redacted>34</span></span>-<span id="cb1612-4"><a href="#cb1612-4" aria-hidden="true" tabindex="-1"></a><span class="dt">compartment </span><span class="ot">=</span><span class="st"> ocid1.compartment.oc1..aa<redacted>ba</span></span>-<span id="cb1612-5"><a href="#cb1612-5" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span><span class="st"> us-ashburn-1</span></span>-<span id="cb1612-6"><a href="#cb1612-6" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> resource_principal_auth</span></span></code></pre></div>+<div class="sourceCode" id="cb1680"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1680-1"><a href="#cb1680-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[oos]</span></span>+<span id="cb1680-2"><a href="#cb1680-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> oracleobjectstorage</span></span>+<span id="cb1680-3"><a href="#cb1680-3" aria-hidden="true" tabindex="-1"></a><span class="dt">namespace </span><span class="ot">=</span><span class="st"> id<redacted>34</span></span>+<span id="cb1680-4"><a href="#cb1680-4" aria-hidden="true" tabindex="-1"></a><span class="dt">compartment </span><span class="ot">=</span><span class="st"> ocid1.compartment.oc1..aa<redacted>ba</span></span>+<span id="cb1680-5"><a href="#cb1680-5" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span><span class="st"> us-ashburn-1</span></span>+<span id="cb1680-6"><a href="#cb1680-6" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> resource_principal_auth</span></span></code></pre></div><h3 id="workload-identity">Workload Identity</h3><p>Workload Identity auth may be used when running Rclone fromKubernetes pod on a Container Engine for Kubernetes (OKE) cluster. For@@ -57041,14 +60289,14 @@ export OCI_RESOURCE_PRINCIPAL_REGION=us-ashburn-1</code></pre><h3 id="no-authentication">No authentication</h3><p>Public buckets do not require any authentication mechanism to readobjects. Sample rclone configuration file for No authentication:</p>-<div class="sourceCode" id="cb1614"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1614-1"><a href="#cb1614-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[oos]</span></span>-<span id="cb1614-2"><a href="#cb1614-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> oracleobjectstorage</span></span>-<span id="cb1614-3"><a href="#cb1614-3" aria-hidden="true" tabindex="-1"></a><span class="dt">namespace </span><span class="ot">=</span><span class="st"> id<redacted>34</span></span>-<span id="cb1614-4"><a href="#cb1614-4" aria-hidden="true" tabindex="-1"></a><span class="dt">compartment </span><span class="ot">=</span><span class="st"> ocid1.compartment.oc1..aa<redacted>ba</span></span>-<span id="cb1614-5"><a href="#cb1614-5" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span><span class="st"> us-ashburn-1</span></span>-<span id="cb1614-6"><a href="#cb1614-6" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> no_auth</span></span></code></pre></div>-<h3 id="modification-times-and-hashes-22">Modification times and+<div class="sourceCode" id="cb1682"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1682-1"><a href="#cb1682-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[oos]</span></span>+<span id="cb1682-2"><a href="#cb1682-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> oracleobjectstorage</span></span>+<span id="cb1682-3"><a href="#cb1682-3" aria-hidden="true" tabindex="-1"></a><span class="dt">namespace </span><span class="ot">=</span><span class="st"> id<redacted>34</span></span>+<span id="cb1682-4"><a href="#cb1682-4" aria-hidden="true" tabindex="-1"></a><span class="dt">compartment </span><span class="ot">=</span><span class="st"> ocid1.compartment.oc1..aa<redacted>ba</span></span>+<span id="cb1682-5"><a href="#cb1682-5" aria-hidden="true" tabindex="-1"></a><span class="dt">region </span><span class="ot">=</span><span class="st"> us-ashburn-1</span></span>+<span id="cb1682-6"><a href="#cb1682-6" aria-hidden="true" tabindex="-1"></a><span class="dt">provider </span><span class="ot">=</span><span class="st"> no_auth</span></span></code></pre></div>+<h3 id="modification-times-and-hashes-24">Modification times andhashes</h3><p>The modification time is stored as metadata on the object as<code>opc-meta-mtime</code> as floating point since the epoch, accurate@@ -57087,7 +60335,7 @@ values are high enough to gain most of the possible performance withoutusing too much memory.</p><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/oracleobjectstorage/oracleobjectstorage.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-44">Standard options</h3>+<h3 id="standard-options-46">Standard options</h3><p>Here are the Standard options specific to oracleobjectstorage (OracleCloud Infrastructure Object Storage).</p><h4 id="oos-provider">--oos-provider</h4>@@ -57211,7 +60459,7 @@ buckets</li></ul></li></ul></li></ul>-<h3 id="advanced-options-45">Advanced options</h3>+<h3 id="advanced-options-47">Advanced options</h3><p>Here are the Advanced options specific to oracleobjectstorage (OracleCloud Infrastructure Object Storage).</p><h4 id="oos-storage-tier">--oos-storage-tier</h4>@@ -57350,7 +60598,7 @@ section in the overview</a> for more info.</p><li>Config: encoding</li><li>Env Var: RCLONE_OOS_ENCODING</li><li>Type: Encoding</li>-<li>Default: Slash,InvalidUtf8,Dot</li>+<li>Default: Slash,CrLf,InvalidUtf8,Dot</li></ul><h4 id="oos-leave-parts-on-error">--oos-leave-parts-on-error</h4><p>If true avoid calling abort upload on a failure, leaving all@@ -57621,26 +60869,26 @@ format.</p>multipart uploads.</p><p>You can call it with no bucket in which case it lists all bucket,with a bucket or with a bucket and path.</p>-<div class="sourceCode" id="cb1620"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb1620-1"><a href="#cb1620-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>-<span id="cb1620-2"><a href="#cb1620-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"test-bucket"</span><span class="fu">:</span> <span class="ot">[</span></span>-<span id="cb1620-3"><a href="#cb1620-3" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>-<span id="cb1620-4"><a href="#cb1620-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"namespace"</span><span class="fu">:</span> <span class="st">"test-namespace"</span><span class="fu">,</span></span>-<span id="cb1620-5"><a href="#cb1620-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"bucket"</span><span class="fu">:</span> <span class="st">"test-bucket"</span><span class="fu">,</span></span>-<span id="cb1620-6"><a href="#cb1620-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">"object"</span><span class="fu">:</span> <span class="st">"600m.bin"</span><span class="fu">,</span></span>-<span id="cb1620-7"><a href="#cb1620-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"uploadId"</span><span class="fu">:</span> <span class="st">"51dd8114-52a4-b2f2-c42f-5291f05eb3c8"</span><span class="fu">,</span></span>-<span id="cb1620-8"><a href="#cb1620-8" aria-hidden="true" tabindex="-1"></a> <span class="dt">"timeCreated"</span><span class="fu">:</span> <span class="st">"2022-07-29T06:21:16.595Z"</span><span class="fu">,</span></span>-<span id="cb1620-9"><a href="#cb1620-9" aria-hidden="true" tabindex="-1"></a> <span class="dt">"storageTier"</span><span class="fu">:</span> <span class="st">"Standard"</span></span>-<span id="cb1620-10"><a href="#cb1620-10" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>-<span id="cb1620-11"><a href="#cb1620-11" aria-hidden="true" tabindex="-1"></a> <span class="ot">]</span></span>-<span id="cb1620-12"><a href="#cb1620-12" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span>-<span id="cb1620-13"><a href="#cb1620-13" aria-hidden="true" tabindex="-1"></a></span>-<span id="cb1620-14"><a href="#cb1620-14" aria-hidden="true" tabindex="-1"></a><span class="er">###</span> <span class="er">cleanup</span></span>-<span id="cb1620-15"><a href="#cb1620-15" aria-hidden="true" tabindex="-1"></a></span>-<span id="cb1620-16"><a href="#cb1620-16" aria-hidden="true" tabindex="-1"></a><span class="er">Remove</span> <span class="er">unfinished</span> <span class="er">multipart</span> <span class="er">uploads.</span></span>-<span id="cb1620-17"><a href="#cb1620-17" aria-hidden="true" tabindex="-1"></a></span>-<span id="cb1620-18"><a href="#cb1620-18" aria-hidden="true" tabindex="-1"></a><span class="er">```console</span></span>-<span id="cb1620-19"><a href="#cb1620-19" aria-hidden="true" tabindex="-1"></a><span class="er">rclone</span> <span class="er">backend</span> <span class="er">cleanup</span> <span class="er">remote:</span> <span class="ot">[</span><span class="er">options</span><span class="ot">]</span> <span class="ot">[</span><span class="er"><arguments>+</span><span class="ot">]</span></span></code></pre></div>+<div class="sourceCode" id="cb1688"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb1688-1"><a href="#cb1688-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb1688-2"><a href="#cb1688-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"test-bucket"</span><span class="fu">:</span> <span class="ot">[</span></span>+<span id="cb1688-3"><a href="#cb1688-3" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>+<span id="cb1688-4"><a href="#cb1688-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"namespace"</span><span class="fu">:</span> <span class="st">"test-namespace"</span><span class="fu">,</span></span>+<span id="cb1688-5"><a href="#cb1688-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"bucket"</span><span class="fu">:</span> <span class="st">"test-bucket"</span><span class="fu">,</span></span>+<span id="cb1688-6"><a href="#cb1688-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">"object"</span><span class="fu">:</span> <span class="st">"600m.bin"</span><span class="fu">,</span></span>+<span id="cb1688-7"><a href="#cb1688-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"uploadId"</span><span class="fu">:</span> <span class="st">"51dd8114-52a4-b2f2-c42f-5291f05eb3c8"</span><span class="fu">,</span></span>+<span id="cb1688-8"><a href="#cb1688-8" aria-hidden="true" tabindex="-1"></a> <span class="dt">"timeCreated"</span><span class="fu">:</span> <span class="st">"2022-07-29T06:21:16.595Z"</span><span class="fu">,</span></span>+<span id="cb1688-9"><a href="#cb1688-9" aria-hidden="true" tabindex="-1"></a> <span class="dt">"storageTier"</span><span class="fu">:</span> <span class="st">"Standard"</span></span>+<span id="cb1688-10"><a href="#cb1688-10" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>+<span id="cb1688-11"><a href="#cb1688-11" aria-hidden="true" tabindex="-1"></a> <span class="ot">]</span></span>+<span id="cb1688-12"><a href="#cb1688-12" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span>+<span id="cb1688-13"><a href="#cb1688-13" aria-hidden="true" tabindex="-1"></a></span>+<span id="cb1688-14"><a href="#cb1688-14" aria-hidden="true" tabindex="-1"></a><span class="er">###</span> <span class="er">cleanup</span></span>+<span id="cb1688-15"><a href="#cb1688-15" aria-hidden="true" tabindex="-1"></a></span>+<span id="cb1688-16"><a href="#cb1688-16" aria-hidden="true" tabindex="-1"></a><span class="er">Remove</span> <span class="er">unfinished</span> <span class="er">multipart</span> <span class="er">uploads.</span></span>+<span id="cb1688-17"><a href="#cb1688-17" aria-hidden="true" tabindex="-1"></a></span>+<span id="cb1688-18"><a href="#cb1688-18" aria-hidden="true" tabindex="-1"></a><span class="er">```console</span></span>+<span id="cb1688-19"><a href="#cb1688-19" aria-hidden="true" tabindex="-1"></a><span class="er">rclone</span> <span class="er">backend</span> <span class="er">cleanup</span> <span class="er">remote:</span> <span class="ot">[</span><span class="er">options</span><span class="ot">]</span> <span class="ot">[</span><span class="er"><arguments>+</span><span class="ot">]</span></span></code></pre></div><p>This command removes unfinished multipart uploads of age greater thanmax-age which defaults to 24 hours.</p><p>Note that you can use --interactive/-i or --dry-run with this command@@ -57669,17 +60917,17 @@ rclone backend restore oos:bucket -o hours=HOURS</code></pre><p>It returns a list of status dictionaries with Object Name and Statuskeys. The Status will be "RESTORED"" if it was successful or an errormessage if not.</p>-<div class="sourceCode" id="cb1626"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb1626-1"><a href="#cb1626-1" aria-hidden="true" tabindex="-1"></a><span class="ot">[</span></span>-<span id="cb1626-2"><a href="#cb1626-2" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>-<span id="cb1626-3"><a href="#cb1626-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Object"</span><span class="fu">:</span> <span class="st">"test.txt"</span></span>-<span id="cb1626-4"><a href="#cb1626-4" aria-hidden="true" tabindex="-1"></a> <span class="st">"Status"</span><span class="er">:</span> <span class="st">"RESTORED"</span><span class="fu">,</span></span>-<span id="cb1626-5"><a href="#cb1626-5" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span><span class="ot">,</span></span>-<span id="cb1626-6"><a href="#cb1626-6" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>-<span id="cb1626-7"><a href="#cb1626-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Object"</span><span class="fu">:</span> <span class="st">"test/file4.txt"</span></span>-<span id="cb1626-8"><a href="#cb1626-8" aria-hidden="true" tabindex="-1"></a> <span class="st">"Status"</span><span class="er">:</span> <span class="st">"RESTORED"</span><span class="fu">,</span></span>-<span id="cb1626-9"><a href="#cb1626-9" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>-<span id="cb1626-10"><a href="#cb1626-10" aria-hidden="true" tabindex="-1"></a><span class="ot">]</span></span></code></pre></div>+<div class="sourceCode" id="cb1694"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb1694-1"><a href="#cb1694-1" aria-hidden="true" tabindex="-1"></a><span class="ot">[</span></span>+<span id="cb1694-2"><a href="#cb1694-2" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>+<span id="cb1694-3"><a href="#cb1694-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Object"</span><span class="fu">:</span> <span class="st">"test.txt"</span></span>+<span id="cb1694-4"><a href="#cb1694-4" aria-hidden="true" tabindex="-1"></a> <span class="st">"Status"</span><span class="er">:</span> <span class="st">"RESTORED"</span><span class="fu">,</span></span>+<span id="cb1694-5"><a href="#cb1694-5" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span><span class="ot">,</span></span>+<span id="cb1694-6"><a href="#cb1694-6" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span>+<span id="cb1694-7"><a href="#cb1694-7" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Object"</span><span class="fu">:</span> <span class="st">"test/file4.txt"</span></span>+<span id="cb1694-8"><a href="#cb1694-8" aria-hidden="true" tabindex="-1"></a> <span class="st">"Status"</span><span class="er">:</span> <span class="st">"RESTORED"</span><span class="fu">,</span></span>+<span id="cb1694-9"><a href="#cb1694-9" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span>+<span id="cb1694-10"><a href="#cb1694-10" aria-hidden="true" tabindex="-1"></a><span class="ot">]</span></span></code></pre></div><p>Options:</p><ul><li>"hours": The number of hours for which this object will be restored.@@ -57694,7 +60942,7 @@ Buckets</a></h3><p>Paths are specified as <code>remote:bucket</code> (or<code>remote:</code> for the <code>lsd</code> command.) You may putsubdirectories in too, e.g. <code>remote:bucket/path/to/dir</code>.</p>-<h2 id="configuration-45">Configuration</h2>+<h2 id="configuration-48">Configuration</h2><p>Here is an example of making an QingStor configuration. First run</p><pre class="console"><code>rclone config</code></pre><p>This will guide you through an interactive setup process.</p>@@ -57809,7 +61057,7 @@ file</li></ul></li></ul></li></ul>-<h3 id="restricted-filename-characters-26">Restricted filename+<h3 id="restricted-filename-characters-29">Restricted filenamecharacters</h3><p>The control characters 0x00-0x1F and / are replaced as in the <ahref="https://rclone.org/overview/#restricted-characters">default@@ -57819,7 +61067,7 @@ href="https://rclone.org/overview/#invalid-utf8">replaced</a>, as theycan't be used in JSON strings.</p><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/qingstor/qingstor.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-45">Standard options</h3>+<h3 id="standard-options-47">Standard options</h3><p>Here are the Standard options specific to qingstor (QingCloud ObjectStorage).</p><h4 id="qingstor-env-auth">--qingstor-env-auth</h4>@@ -57903,7 +61151,7 @@ IAM).</li></ul></li></ul></li></ul>-<h3 id="advanced-options-46">Advanced options</h3>+<h3 id="advanced-options-48">Advanced options</h3><p>Here are the Advanced options specific to qingstor (QingCloud ObjectStorage).</p><h4 id="qingstor-connection-retries">--qingstor-connection-retries</h4>@@ -57979,7 +61227,7 @@ section in the overview</a> for more info.</p><li>Required: false</li></ul><!-- autogenerated options stop -->-<h2 id="limitations-34">Limitations</h2>+<h2 id="limitations-35">Limitations</h2><p><code>rclone about</code> is not supported by the qingstor backend.Backends without this capability cannot determine free space for anrclone mount or use policy <code>mfs</code> (most free space) as a@@ -58003,7 +61251,7 @@ class="uri">https://docs.maytech.net/quatrix/quatrix-api/api-explorer#/API-Key/p<p>See complete <ahref="https://docs.maytech.net/quatrix/quatrix-api/api-explorer">Swaggerdocumentation for Quatrix</a>.</p>-<h2 id="configuration-46">Configuration</h2>+<h2 id="configuration-49">Configuration</h2><p>Here is an example of how to make a remote called<code>remote</code>. First run:</p><pre class="console"><code>rclone config</code></pre>@@ -58093,14 +61341,14 @@ y) Yes this is OKe) Edit this remoted) Delete this remotey/e/d> y</code></pre>-<h3 id="modification-times-and-hashes-23">Modification times and+<h3 id="modification-times-and-hashes-25">Modification times andhashes</h3><p>Quatrix allows modification times to be set on objects accurate to 1microsecond. These will be used to detect whether objects need syncingor not.</p><p>Quatrix does not support hashes, so you cannot use the<code>--checksum</code> flag.</p>-<h3 id="restricted-filename-characters-27">Restricted filename+<h3 id="restricted-filename-characters-30">Restricted filenamecharacters</h3><p>File names in Quatrix are case sensitive and have limitations likethe maximum length of a filename is 255, and the minimum length is 1. A@@ -58123,14 +61371,14 @@ and <code>minimal_chunk_size</code> * <code>transfers</code>), the chunksize may increase in case of high upload speed. As well as it candecrease in case of upload speed problems. If no free memory isavailable, all chunks will equal <code>minimal_chunk_size</code>.</p>-<h3 id="deleting-files-5">Deleting files</h3>+<h3 id="deleting-files-6">Deleting files</h3><p>Files you delete with rclone will end up in Trash and be stored therefor 30 days. Quatrix also provides an API to permanently delete filesand an API to empty the Trash so that you can remove files permanentlyfrom your account.</p><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/quatrix/quatrix.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-46">Standard options</h3>+<h3 id="standard-options-48">Standard options</h3><p>Here are the Standard options specific to quatrix (Quatrix byMaytech).</p><h4 id="quatrix-api-key">--quatrix-api-key</h4>@@ -58151,7 +61399,7 @@ Maytech).</p><li>Type: string</li><li>Required: true</li></ul>-<h3 id="advanced-options-47">Advanced options</h3>+<h3 id="advanced-options-49">Advanced options</h3><p>Here are the Advanced options specific to quatrix (Quatrix byMaytech).</p><h4 id="quatrix-encoding">--quatrix-encoding</h4>@@ -58250,7 +61498,7 @@ locally on your computer or on local network (e.g. a NAS). Please followthe <a href="https://sia.tech/get-started">Get started</a> guide andinstall one.</p><p>rclone interacts with Sia network by talking to the Sia daemon via <a-href="https://sia.tech/docs/">HTTP API</a> which is usually available on+href="https://docs.sia.tech/">HTTP API</a> which is usually available onport <em>9980</em>. By default you will run the daemon locally on thesame computer so it's safe to leave the API password blank (the API URLwill be <code>http://127.0.0.1:9980</code> making external access@@ -58292,7 +61540,7 @@ it <strong>on localhost</strong> with command line argument<code>--authorize-api=false</code>, but this is insecure and<strong>strongly discouraged</strong>.</li></ol>-<h2 id="configuration-47">Configuration</h2>+<h2 id="configuration-50">Configuration</h2><p>Here is an example of how to make a <code>sia</code> remote called<code>mySia</code>. First, run:</p><pre class="console"><code>rclone config</code></pre>@@ -58352,7 +61600,7 @@ y/e/d> y</code></pre></ul><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/sia/sia.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-47">Standard options</h3>+<h3 id="standard-options-49">Standard options</h3><p>Here are the Standard options specific to sia (Sia DecentralizedCloud).</p><h4 id="sia-api-url">--sia-api-url</h4>@@ -58381,7 +61629,7 @@ obscure</a>.</p><li>Type: string</li><li>Required: false</li></ul>-<h3 id="advanced-options-48">Advanced options</h3>+<h3 id="advanced-options-50">Advanced options</h3><p>Here are the Advanced options specific to sia (Sia DecentralizedCloud).</p><h4 id="sia-user-agent">--sia-user-agent</h4>@@ -58416,7 +61664,7 @@ section in the overview</a> for more info.</p><li>Required: false</li></ul><!-- autogenerated options stop -->-<h2 id="limitations-35">Limitations</h2>+<h2 id="limitations-36">Limitations</h2><ul><li>Modification times not supported</li><li>Checksums not supported</li>@@ -58448,12 +61696,15 @@ Storage</a></li><li><ahref="https://console.bluemix.net/docs/infrastructure/objectstorage-swift/index.html">IBMBluemix Cloud ObjectStorage Swift</a></li>+<li><a+href="https://www.infomaniak.com/en/swiss-backup/cloud-storage">Infomaniak+Swiss Backup</a></li></ul><p>Paths are specified as <code>remote:container</code> (or<code>remote:</code> for the <code>lsd</code> command.) You may putsubdirectories in too, e.g.<code>remote:container/path/to/dir</code>.</p>-<h2 id="configuration-48">Configuration</h2>+<h2 id="configuration-51">Configuration</h2><p>Here is an example of making a swift configuration. First run</p><pre class="console"><code>rclone config</code></pre><p>This will guide you through an interactive setup process.</p>@@ -58560,29 +61811,29 @@ deleting any excess files in the container.</p><pre class="console"><code>rclone sync --interactive /home/local/directory remote:container</code></pre><h3 id="configuration-from-an-openstack-credentials-file">Configurationfrom an OpenStack credentials file</h3>-<p>An OpenStack credentials file typically looks something something-like this (without the comments)</p>-<div class="sourceCode" id="cb1650"><pre-class="sourceCode sh"><code class="sourceCode bash"><span id="cb1650-1"><a href="#cb1650-1" aria-hidden="true" tabindex="-1"></a><span class="bu">export</span> <span class="va">OS_AUTH_URL</span><span class="op">=</span>https://a.provider.net/v2.0</span>-<span id="cb1650-2"><a href="#cb1650-2" aria-hidden="true" tabindex="-1"></a><span class="bu">export</span> <span class="va">OS_TENANT_ID</span><span class="op">=</span>ffffffffffffffffffffffffffffffff</span>-<span id="cb1650-3"><a href="#cb1650-3" aria-hidden="true" tabindex="-1"></a><span class="bu">export</span> <span class="va">OS_TENANT_NAME</span><span class="op">=</span><span class="st">"1234567890123456"</span></span>-<span id="cb1650-4"><a href="#cb1650-4" aria-hidden="true" tabindex="-1"></a><span class="bu">export</span> <span class="va">OS_USERNAME</span><span class="op">=</span><span class="st">"123abc567xy"</span></span>-<span id="cb1650-5"><a href="#cb1650-5" aria-hidden="true" tabindex="-1"></a><span class="bu">echo</span> <span class="st">"Please enter your OpenStack Password: "</span></span>-<span id="cb1650-6"><a href="#cb1650-6" aria-hidden="true" tabindex="-1"></a><span class="bu">read</span> <span class="at">-sr</span> <span class="va">OS_PASSWORD_INPUT</span></span>-<span id="cb1650-7"><a href="#cb1650-7" aria-hidden="true" tabindex="-1"></a><span class="bu">export</span> <span class="va">OS_PASSWORD</span><span class="op">=</span><span class="va">$OS_PASSWORD_INPUT</span></span>-<span id="cb1650-8"><a href="#cb1650-8" aria-hidden="true" tabindex="-1"></a><span class="bu">export</span> <span class="va">OS_REGION_NAME</span><span class="op">=</span><span class="st">"SBG1"</span></span>-<span id="cb1650-9"><a href="#cb1650-9" aria-hidden="true" tabindex="-1"></a><span class="cf">if</span> <span class="bu">[</span> <span class="ot">-z</span> <span class="st">"</span><span class="va">$OS_REGION_NAME</span><span class="st">"</span> <span class="bu">]</span><span class="kw">;</span> <span class="cf">then</span> <span class="bu">unset</span> <span class="va">OS_REGION_NAME</span><span class="kw">;</span> <span class="cf">fi</span></span></code></pre></div>+<p>An OpenStack credentials file typically looks something like this+(without the comments)</p>+<div class="sourceCode" id="cb1718"><pre+class="sourceCode sh"><code class="sourceCode bash"><span id="cb1718-1"><a href="#cb1718-1" aria-hidden="true" tabindex="-1"></a><span class="bu">export</span> <span class="va">OS_AUTH_URL</span><span class="op">=</span>https://a.provider.net/v2.0</span>+<span id="cb1718-2"><a href="#cb1718-2" aria-hidden="true" tabindex="-1"></a><span class="bu">export</span> <span class="va">OS_TENANT_ID</span><span class="op">=</span>ffffffffffffffffffffffffffffffff</span>+<span id="cb1718-3"><a href="#cb1718-3" aria-hidden="true" tabindex="-1"></a><span class="bu">export</span> <span class="va">OS_TENANT_NAME</span><span class="op">=</span><span class="st">"1234567890123456"</span></span>+<span id="cb1718-4"><a href="#cb1718-4" aria-hidden="true" tabindex="-1"></a><span class="bu">export</span> <span class="va">OS_USERNAME</span><span class="op">=</span><span class="st">"123abc567xy"</span></span>+<span id="cb1718-5"><a href="#cb1718-5" aria-hidden="true" tabindex="-1"></a><span class="bu">echo</span> <span class="st">"Please enter your OpenStack Password: "</span></span>+<span id="cb1718-6"><a href="#cb1718-6" aria-hidden="true" tabindex="-1"></a><span class="bu">read</span> <span class="at">-sr</span> <span class="va">OS_PASSWORD_INPUT</span></span>+<span id="cb1718-7"><a href="#cb1718-7" aria-hidden="true" tabindex="-1"></a><span class="bu">export</span> <span class="va">OS_PASSWORD</span><span class="op">=</span><span class="va">$OS_PASSWORD_INPUT</span></span>+<span id="cb1718-8"><a href="#cb1718-8" aria-hidden="true" tabindex="-1"></a><span class="bu">export</span> <span class="va">OS_REGION_NAME</span><span class="op">=</span><span class="st">"SBG1"</span></span>+<span id="cb1718-9"><a href="#cb1718-9" aria-hidden="true" tabindex="-1"></a><span class="cf">if</span> <span class="bu">[</span> <span class="ot">-z</span> <span class="st">"</span><span class="va">$OS_REGION_NAME</span><span class="st">"</span> <span class="bu">]</span><span class="kw">;</span> <span class="cf">then</span> <span class="bu">unset</span> <span class="va">OS_REGION_NAME</span><span class="kw">;</span> <span class="cf">fi</span></span></code></pre></div><p>The config file needs to look something like this where<code>$OS_USERNAME</code> represents the value of the<code>OS_USERNAME</code> variable - <code>123abc567xy</code> in theexample above.</p>-<div class="sourceCode" id="cb1651"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1651-1"><a href="#cb1651-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[remote]</span></span>-<span id="cb1651-2"><a href="#cb1651-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> swift</span></span>-<span id="cb1651-3"><a href="#cb1651-3" aria-hidden="true" tabindex="-1"></a><span class="dt">user </span><span class="ot">=</span><span class="st"> $OS_USERNAME</span></span>-<span id="cb1651-4"><a href="#cb1651-4" aria-hidden="true" tabindex="-1"></a><span class="dt">key </span><span class="ot">=</span><span class="st"> $OS_PASSWORD</span></span>-<span id="cb1651-5"><a href="#cb1651-5" aria-hidden="true" tabindex="-1"></a><span class="dt">auth </span><span class="ot">=</span><span class="st"> $OS_AUTH_URL</span></span>-<span id="cb1651-6"><a href="#cb1651-6" aria-hidden="true" tabindex="-1"></a><span class="dt">tenant </span><span class="ot">=</span><span class="st"> $OS_TENANT_NAME</span></span></code></pre></div>+<div class="sourceCode" id="cb1719"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1719-1"><a href="#cb1719-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[remote]</span></span>+<span id="cb1719-2"><a href="#cb1719-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> swift</span></span>+<span id="cb1719-3"><a href="#cb1719-3" aria-hidden="true" tabindex="-1"></a><span class="dt">user </span><span class="ot">=</span><span class="st"> $OS_USERNAME</span></span>+<span id="cb1719-4"><a href="#cb1719-4" aria-hidden="true" tabindex="-1"></a><span class="dt">key </span><span class="ot">=</span><span class="st"> $OS_PASSWORD</span></span>+<span id="cb1719-5"><a href="#cb1719-5" aria-hidden="true" tabindex="-1"></a><span class="dt">auth </span><span class="ot">=</span><span class="st"> $OS_AUTH_URL</span></span>+<span id="cb1719-6"><a href="#cb1719-6" aria-hidden="true" tabindex="-1"></a><span class="dt">tenant </span><span class="ot">=</span><span class="st"> $OS_TENANT_NAME</span></span></code></pre></div><p>Note that you may (or may not) need to set <code>region</code> too -try without first.</p><h3 id="configuration-from-the-environment">Configuration from the@@ -58611,11 +61862,11 @@ OpenStack installation.</p>config file</h4><p>You can use rclone with swift without a config file, if desired, likethis:</p>-<div class="sourceCode" id="cb1652"><pre-class="sourceCode sh"><code class="sourceCode bash"><span id="cb1652-1"><a href="#cb1652-1" aria-hidden="true" tabindex="-1"></a><span class="bu">source</span> openstack-credentials-file</span>-<span id="cb1652-2"><a href="#cb1652-2" aria-hidden="true" tabindex="-1"></a><span class="bu">export</span> <span class="va">RCLONE_CONFIG_MYREMOTE_TYPE</span><span class="op">=</span>swift</span>-<span id="cb1652-3"><a href="#cb1652-3" aria-hidden="true" tabindex="-1"></a><span class="bu">export</span> <span class="va">RCLONE_CONFIG_MYREMOTE_ENV_AUTH</span><span class="op">=</span>true</span>-<span id="cb1652-4"><a href="#cb1652-4" aria-hidden="true" tabindex="-1"></a><span class="ex">rclone</span> lsd myremote:</span></code></pre></div>+<div class="sourceCode" id="cb1720"><pre+class="sourceCode sh"><code class="sourceCode bash"><span id="cb1720-1"><a href="#cb1720-1" aria-hidden="true" tabindex="-1"></a><span class="bu">source</span> openstack-credentials-file</span>+<span id="cb1720-2"><a href="#cb1720-2" aria-hidden="true" tabindex="-1"></a><span class="bu">export</span> <span class="va">RCLONE_CONFIG_MYREMOTE_TYPE</span><span class="op">=</span>swift</span>+<span id="cb1720-3"><a href="#cb1720-3" aria-hidden="true" tabindex="-1"></a><span class="bu">export</span> <span class="va">RCLONE_CONFIG_MYREMOTE_ENV_AUTH</span><span class="op">=</span>true</span>+<span id="cb1720-4"><a href="#cb1720-4" aria-hidden="true" tabindex="-1"></a><span class="ex">rclone</span> lsd myremote:</span></code></pre></div><h3 id="fast-list-8">--fast-list</h3><p>This remote supports <code>--fast-list</code> which allows you to usefewer transactions in exchange for more memory. See the <a@@ -58633,7 +61884,7 @@ remote is sufficient to determine if it is "dirty". By using<code>--update</code> along with <code>--use-server-modtime</code>, youcan avoid the extra API call and simply upload files whose local modtimeis newer than the time it was last uploaded.</p>-<h3 id="modification-times-and-hashes-24">Modification times and+<h3 id="modification-times-and-hashes-26">Modification times andhashes</h3><p>The modified time is stored as metadata on the object as<code>X-Object-Meta-Mtime</code> as floating point since the epoch@@ -58641,7 +61892,7 @@ accurate to 1 ns.</p><p>This is a de facto standard (used in the official python-swiftclientamongst others) for storing the modification time for an object.</p><p>The MD5 hash algorithm is supported.</p>-<h3 id="restricted-filename-characters-28">Restricted filename+<h3 id="restricted-filename-characters-31">Restricted filenamecharacters</h3><table><thead>@@ -58669,7 +61920,7 @@ href="https://rclone.org/overview/#invalid-utf8">replaced</a>, as theycan't be used in JSON strings.</p><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/swift/swift.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-48">Standard options</h3>+<h3 id="standard-options-50">Standard options</h3><p>Here are the Standard options specific to swift (OpenStack Swift(Rackspace Cloud Files, Blomp Cloud Storage, Memset Memstore, OVH)).</p><h4 id="swift-env-auth">--swift-env-auth</h4>@@ -58750,6 +62001,22 @@ OpenStack form.</p><ul><li>Blomp Cloud Storage</li></ul></li>+<li>"https://swiss-backup01.infomaniak.com/identity/v3"+<ul>+<li>Infomaniak Swiss Backup Location 01</li>+</ul></li>+<li>"https://swiss-backup02.infomaniak.com/identity/v3"+<ul>+<li>Infomaniak Swiss Backup Location 02</li>+</ul></li>+<li>"https://swiss-backup03.infomaniak.com/identity/v3"+<ul>+<li>Infomaniak Swiss Backup Location 03</li>+</ul></li>+<li>"https://swiss-backup04.infomaniak.com/identity/v3"+<ul>+<li>Infomaniak Swiss Backup Location 04</li>+</ul></li></ul></li></ul><h4 id="swift-user-id">--swift-user-id</h4>@@ -58921,7 +62188,7 @@ provider.</p></ul></li></ul></li></ul>-<h3 id="advanced-options-49">Advanced options</h3>+<h3 id="advanced-options-51">Advanced options</h3><p>Here are the Advanced options specific to swift (OpenStack Swift(Rackspace Cloud Files, Blomp Cloud Storage, Memset Memstore, OVH)).</p><h4 id="swift-leave-parts-on-error">--swift-leave-parts-on-error</h4>@@ -59081,7 +62348,7 @@ section in the overview</a> for more info.</p><li>Required: false</li></ul><!-- autogenerated options stop -->-<h2 id="limitations-36">Limitations</h2>+<h2 id="limitations-37">Limitations</h2><p>The Swift API doesn't return a correct MD5SUM for segmented files(Dynamic or Static Large Objects) so rclone won't check or use theMD5SUM for these.</p>@@ -59093,7 +62360,7 @@ gives Failed to create file system for "remote:": Bad Request</h3>Request" error rather than a more sensible error when the authenticationfails for Swift.</p><p>So this most likely means your username / password is wrong. You can-investigate further with the <code>--dump-bodies</code> flag.</p>+investigate further with the <code>--dump bodies</code> flag.</p><p>This may also be caused by specifying the region when you shouldn'thave (e.g. OVH).</p><h3@@ -59122,7 +62389,7 @@ the following:</p><p>Paths are specified as <code>remote:path</code></p><p>Paths may be as deep as required, e.g.<code>remote:directory/subdirectory</code>.</p>-<h2 id="configuration-49">Configuration</h2>+<h2 id="configuration-52">Configuration</h2><p>The initial setup for pCloud involves getting a token from pCloudwhich you need to do in your browser. <code>rclone config</code> walksyou through it.</p>@@ -59193,7 +62460,7 @@ you to unblock it temporarily if you are running a host firewall.</p><pre class="console"><code>rclone ls remote:</code></pre><p>To copy a local directory to a pCloud directory called backup</p><pre class="console"><code>rclone copy /home/source remote:backup</code></pre>-<h3 id="modification-times-and-hashes-25">Modification times and+<h3 id="modification-times-and-hashes-27">Modification times andhashes</h3><p>pCloud allows modification times to be set on objects accurate to 1second. These will be used to detect whether objects need syncing or@@ -59202,7 +62469,7 @@ re-uploaded.</p><p>pCloud supports MD5 and SHA1 hashes in the US region, and SHA1 andSHA256 hashes in the EU region, so you can use the<code>--checksum</code> flag.</p>-<h3 id="restricted-filename-characters-29">Restricted filename+<h3 id="restricted-filename-characters-32">Restricted filenamecharacters</h3><p>In addition to the <ahref="https://rclone.org/overview/#restricted-characters">default@@ -59227,7 +62494,7 @@ replaced:</p><p>Invalid UTF-8 bytes will also be <ahref="https://rclone.org/overview/#invalid-utf8">replaced</a>, as theycan't be used in JSON strings.</p>-<h3 id="deleting-files-6">Deleting files</h3>+<h3 id="deleting-files-7">Deleting files</h3><p>Deleted files will be moved to the trash. Your subscription levelwill determine how long items stay in the trash.<code>rclone cleanup</code> can be used to empty the trash.</p>@@ -59237,7 +62504,7 @@ will only work if you set your username and password in the advancedoptions for this backend. Since we generally want to avoid storing userpasswords in the rclone config file, we advise you to only set this upif you need the <code>rclone cleanup</code> command to work.</p>-<h3 id="root-folder-id-5">Root folder ID</h3>+<h3 id="root-folder-id-6">Root folder ID</h3><p>You can set the <code>root_folder_id</code> for rclone. This is thedirectory (identified by its <code>Folder ID</code>) that rcloneconsiders to be the root of your pCloud drive.</p>@@ -59280,12 +62547,12 @@ without manual refreshes.</li></ul><p>If you want to debug or verify notifications, you can use the helpercommand:</p>-<div class="sourceCode" id="cb1660"><pre-class="sourceCode bash"><code class="sourceCode bash"><span id="cb1660-1"><a href="#cb1660-1" aria-hidden="true" tabindex="-1"></a><span class="ex">rclone</span> test changenotify remote:</span></code></pre></div>+<div class="sourceCode" id="cb1728"><pre+class="sourceCode bash"><code class="sourceCode bash"><span id="cb1728-1"><a href="#cb1728-1" aria-hidden="true" tabindex="-1"></a><span class="ex">rclone</span> test changenotify remote:</span></code></pre></div><p>This will log incoming change notifications for the given remote.</p><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/pcloud/pcloud.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-49">Standard options</h3>+<h3 id="standard-options-51">Standard options</h3><p>Here are the Standard options specific to pcloud (Pcloud).</p><h4 id="pcloud-client-id">--pcloud-client-id</h4><p>OAuth Client Id.</p>@@ -59307,7 +62574,7 @@ class="sourceCode bash"><code class="sourceCode bash"><span id="cb1660-1"><a hre<li>Type: string</li><li>Required: false</li></ul>-<h3 id="advanced-options-50">Advanced options</h3>+<h3 id="advanced-options-52">Advanced options</h3><p>Here are the Advanced options specific to pcloud (Pcloud).</p><h4 id="pcloud-token">--pcloud-token</h4><p>OAuth Access Token as a JSON blob.</p>@@ -59434,7 +62701,7 @@ obscure</a>.</p>drive</a>.</p><p>Paths are specified as <code>remote:path</code>, and may be as deepas required, e.g. <code>remote:directory/subdirectory</code>.</p>-<h2 id="configuration-50">Configuration</h2>+<h2 id="configuration-53">Configuration</h2><p>Here is an example of making a remote for PikPak.</p><p>First run:</p><pre class="console"><code>rclone config</code></pre>@@ -59487,7 +62754,7 @@ y) Yes this is OK (default)e) Edit this remoted) Delete this remotey/e/d> y</code></pre>-<h3 id="modification-times-and-hashes-26">Modification times and+<h3 id="modification-times-and-hashes-28">Modification times andhashes</h3><p>PikPak keeps modification times on objects, and updates them whenuploading objects, but it does not support changing only the@@ -59495,7 +62762,7 @@ modification time</p><p>The MD5 hash algorithm is supported.</p><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/pikpak/pikpak.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-50">Standard options</h3>+<h3 id="standard-options-52">Standard options</h3><p>Here are the Standard options specific to pikpak (PikPak).</p><h4 id="pikpak-user">--pikpak-user</h4><p>Pikpak username.</p>@@ -59518,7 +62785,7 @@ obscure</a>.</p><li>Type: string</li><li>Required: true</li></ul>-<h3 id="advanced-options-51">Advanced options</h3>+<h3 id="advanced-options-53">Advanced options</h3><p>Here are the Advanced options specific to pikpak (PikPak).</p><h4 id="pikpak-device-id">--pikpak-device-id</h4><p>Device ID used for authorization.</p>@@ -59590,6 +62857,7 @@ download speeds.</p><h4 id="pikpak-hash-memory-limit">--pikpak-hash-memory-limit</h4><p>Files bigger than this will be cached on disk to calculate hash ifrequired.</p>+<p>Files of unknown size are always cached on disk.</p><p>Properties:</p><ul><li>Config: hash_memory_limit</li>@@ -59701,14 +62969,14 @@ in 'pikpak:dirpath'. You may want to pass '-o password=password' for apassword-protected files. Also, pass '-o delete-src-file' to deletesource files after decompression finished.</p><p>Result:</p>-<div class="sourceCode" id="cb1668"><pre-class="sourceCode json"><code class="sourceCode json"><span id="cb1668-1"><a href="#cb1668-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>-<span id="cb1668-2"><a href="#cb1668-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Decompressed"</span><span class="fu">:</span> <span class="dv">17</span><span class="fu">,</span></span>-<span id="cb1668-3"><a href="#cb1668-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"SourceDeleted"</span><span class="fu">:</span> <span class="dv">0</span><span class="fu">,</span></span>-<span id="cb1668-4"><a href="#cb1668-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Errors"</span><span class="fu">:</span> <span class="dv">0</span></span>-<span id="cb1668-5"><a href="#cb1668-5" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div>+<div class="sourceCode" id="cb1736"><pre+class="sourceCode json"><code class="sourceCode json"><span id="cb1736-1"><a href="#cb1736-1" aria-hidden="true" tabindex="-1"></a><span class="fu">{</span></span>+<span id="cb1736-2"><a href="#cb1736-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Decompressed"</span><span class="fu">:</span> <span class="dv">17</span><span class="fu">,</span></span>+<span id="cb1736-3"><a href="#cb1736-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"SourceDeleted"</span><span class="fu">:</span> <span class="dv">0</span><span class="fu">,</span></span>+<span id="cb1736-4"><a href="#cb1736-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"Errors"</span><span class="fu">:</span> <span class="dv">0</span></span>+<span id="cb1736-5"><a href="#cb1736-5" aria-hidden="true" tabindex="-1"></a><span class="fu">}</span></span></code></pre></div><!-- autogenerated options stop -->-<h2 id="limitations-37">Limitations</h2>+<h2 id="limitations-38">Limitations</h2><h3 id="hashes-may-be-empty">Hashes may be empty</h3><p>PikPak supports MD5 hash, but sometimes given empty especially foruser-uploaded files.</p>@@ -59813,7 +63081,7 @@ directory and their public IDs.</p>access the directory.</p><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/pixeldrain/pixeldrain.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-51">Standard options</h3>+<h3 id="standard-options-53">Standard options</h3><p>Here are the Standard options specific to pixeldrain (PixeldrainFilesystem).</p><h4 id="pixeldrain-api-key">--pixeldrain-api-key</h4>@@ -59837,7 +63105,7 @@ directory ID to use a shared directory.</p><li>Type: string</li><li>Default: "me"</li></ul>-<h3 id="advanced-options-52">Advanced options</h3>+<h3 id="advanced-options-54">Advanced options</h3><p>Here are the Advanced options specific to pixeldrain (PixeldrainFilesystem).</p><h4 id="pixeldrain-api-url">--pixeldrain-api-url</h4>@@ -59912,7 +63180,7 @@ for more info.</p><p>Paths are specified as <code>remote:path</code></p><p>Paths may be as deep as required, e.g.<code>remote:directory/subdirectory</code>.</p>-<h2 id="configuration-51">Configuration</h2>+<h2 id="configuration-54">Configuration</h2><p>The initial setup for <ahref="https://premiumize.me/">premiumize.me</a> involves getting a tokenfrom premiumize.me which you need to do in your browser.@@ -59975,12 +63243,12 @@ you to unblock it temporarily if you are running a host firewall.</p><p>To copy a local directory to an premiumize.me directory calledbackup</p><pre class="console"><code>rclone copy /home/source remote:backup</code></pre>-<h3 id="modification-times-and-hashes-27">Modification times and+<h3 id="modification-times-and-hashes-29">Modification times andhashes</h3><p>premiumize.me does not support modification times or hashes,therefore syncing will default to <code>--size-only</code> checking.Note that using <code>--update</code> will work.</p>-<h3 id="restricted-filename-characters-30">Restricted filename+<h3 id="restricted-filename-characters-33">Restricted filenamecharacters</h3><p>In addition to the <ahref="https://rclone.org/overview/#restricted-characters">default@@ -60012,7 +63280,7 @@ href="https://rclone.org/overview/#invalid-utf8">replaced</a>, as theycan't be used in JSON strings.</p><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/premiumizeme/premiumizeme.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-52">Standard options</h3>+<h3 id="standard-options-54">Standard options</h3><p>Here are the Standard options specific to premiumizeme(premiumize.me).</p><h4 id="premiumizeme-client-id">--premiumizeme-client-id</h4>@@ -60035,7 +63303,7 @@ can't be used in JSON strings.</p><li>Type: string</li><li>Required: false</li></ul>-<h3 id="advanced-options-53">Advanced options</h3>+<h3 id="advanced-options-55">Advanced options</h3><p>Here are the Advanced options specific to premiumizeme(premiumize.me).</p><h4 id="premiumizeme-token">--premiumizeme-token</h4>@@ -60101,7 +63369,7 @@ section in the overview</a> for more info.</p><li>Required: false</li></ul><!-- autogenerated options stop -->-<h2 id="limitations-38">Limitations</h2>+<h2 id="limitations-39">Limitations</h2><p>Note that premiumize.me is case insensitive so you can't have a filecalled "Hello.doc" and one called "hello.doc".</p><p>premiumize.me file names can't have the <code>\</code> or@@ -60183,12 +63451,12 @@ attempting to use the credentials in <code>rclone</code> will fail.</p><p>To copy a local directory to an Proton Drive directory calledbackup</p><pre class="console"><code>rclone copy /home/source remote:backup</code></pre>-<h3 id="modification-times-and-hashes-28">Modification times and+<h3 id="modification-times-and-hashes-30">Modification times andhashes</h3><p>Proton Drive Bridge does not support updating modification timesyet.</p><p>The SHA1 hash algorithm is supported.</p>-<h3 id="restricted-filename-characters-31">Restricted filename+<h3 id="restricted-filename-characters-34">Restricted filenamecharacters</h3><p>Invalid UTF-8 bytes will be <ahref="https://rclone.org/overview/#invalid-utf8">replaced</a>, also left@@ -60213,7 +63481,7 @@ clients accessing the same mount point, then we might have a problemwith caching the stale data.</p><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/protondrive/protondrive.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-53">Standard options</h3>+<h3 id="standard-options-55">Standard options</h3><p>Here are the Standard options specific to protondrive (ProtonDrive).</p><h4 id="protondrive-username">--protondrive-username</h4>@@ -60265,7 +63533,7 @@ obscure</a>.</p><li>Type: string</li><li>Required: false</li></ul>-<h3 id="advanced-options-54">Advanced options</h3>+<h3 id="advanced-options-56">Advanced options</h3><p>Here are the Advanced options specific to protondrive (ProtonDrive).</p><h4@@ -60378,7 +63646,7 @@ with caching the stale data.</p><li>Required: false</li></ul><!-- autogenerated options stop -->-<h2 id="limitations-39">Limitations</h2>+<h2 id="limitations-40">Limitations</h2><p>This backend uses the <ahref="https://github.com/henrybear327/Proton-API-Bridge">Proton-API-Bridge</a>,which is based on <a@@ -60407,7 +63675,7 @@ this library, as there isn't official documentation available.</p><p>Paths are specified as <code>remote:path</code></p><p>put.io paths may be as deep as required, e.g.<code>remote:directory/subdirectory</code>.</p>-<h2 id="configuration-52">Configuration</h2>+<h2 id="configuration-55">Configuration</h2><p>The initial setup for put.io involves getting a token from put.iowhich you need to do in your browser. <code>rclone config</code> walksyou through it.</p>@@ -60483,7 +63751,7 @@ mode.</p><pre class="console"><code>rclone ls remote:</code></pre><p>To copy a local directory to a put.io directory called backup</p><pre class="console"><code>rclone copy /home/source remote:backup</code></pre>-<h3 id="restricted-filename-characters-32">Restricted filename+<h3 id="restricted-filename-characters-35">Restricted filenamecharacters</h3><p>In addition to the <ahref="https://rclone.org/overview/#restricted-characters">default@@ -60510,7 +63778,7 @@ href="https://rclone.org/overview/#invalid-utf8">replaced</a>, as theycan't be used in JSON strings.</p><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/putio/putio.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-54">Standard options</h3>+<h3 id="standard-options-56">Standard options</h3><p>Here are the Standard options specific to putio (Put.io).</p><h4 id="putio-client-id">--putio-client-id</h4><p>OAuth Client Id.</p>@@ -60532,7 +63800,7 @@ can't be used in JSON strings.</p><li>Type: string</li><li>Required: false</li></ul>-<h3 id="advanced-options-55">Advanced options</h3>+<h3 id="advanced-options-57">Advanced options</h3><p>Here are the Advanced options specific to putio (Put.io).</p><h4 id="putio-token">--putio-token</h4><p>OAuth Access Token as a JSON blob.</p>@@ -60596,7 +63864,7 @@ section in the overview</a> for more info.</p><li>Required: false</li></ul><!-- autogenerated options stop -->-<h2 id="limitations-40">Limitations</h2>+<h2 id="limitations-41">Limitations</h2><p>put.io has rate limiting. When you hit a limit, rclone automaticallyretries after waiting the amount of time requested by the server.</p><p>If you want to avoid ever hitting these limits, you may use the@@ -60677,12 +63945,12 @@ attempting to use the credentials in <code>rclone</code> will fail.</p><p>To copy a local directory to an Proton Drive directory calledbackup</p><pre class="console"><code>rclone copy /home/source remote:backup</code></pre>-<h3 id="modification-times-and-hashes-29">Modification times and+<h3 id="modification-times-and-hashes-31">Modification times andhashes</h3><p>Proton Drive Bridge does not support updating modification timesyet.</p><p>The SHA1 hash algorithm is supported.</p>-<h3 id="restricted-filename-characters-33">Restricted filename+<h3 id="restricted-filename-characters-36">Restricted filenamecharacters</h3><p>Invalid UTF-8 bytes will be <ahref="https://rclone.org/overview/#invalid-utf8">replaced</a>, also left@@ -60707,7 +63975,7 @@ clients accessing the same mount point, then we might have a problemwith caching the stale data.</p><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/protondrive/protondrive.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-55">Standard options</h3>+<h3 id="standard-options-57">Standard options</h3><p>Here are the Standard options specific to protondrive (ProtonDrive).</p><h4 id="protondrive-username-1">--protondrive-username</h4>@@ -60759,7 +64027,7 @@ obscure</a>.</p><li>Type: string</li><li>Required: false</li></ul>-<h3 id="advanced-options-56">Advanced options</h3>+<h3 id="advanced-options-58">Advanced options</h3><p>Here are the Advanced options specific to protondrive (ProtonDrive).</p><h4@@ -60872,7 +64140,7 @@ with caching the stale data.</p><li>Required: false</li></ul><!-- autogenerated options stop -->-<h2 id="limitations-41">Limitations</h2>+<h2 id="limitations-42">Limitations</h2><p>This backend uses the <ahref="https://github.com/henrybear327/Proton-API-Bridge">Proton-API-Bridge</a>,which is based on <a@@ -60908,7 +64176,7 @@ edition.</li><li>It supports 2FA enabled users</li><li>Using a Library API Token is <strong>not</strong> supported</li></ul>-<h2 id="configuration-53">Configuration</h2>+<h2 id="configuration-56">Configuration</h2><p>There are two distinct modes you can setup your remote:</p><ul><li>you point your remote to the <strong>root of the server</strong>,@@ -61095,7 +64363,7 @@ to use fewer transactions in exchange for more memory. See the <ahref="https://rclone.org/docs/#fast-list">rclone docs</a> for moredetails. Please note this is not supported on seafile server version6.x</p>-<h3 id="restricted-filename-characters-34">Restricted filename+<h3 id="restricted-filename-characters-37">Restricted filenamecharacters</h3><p>In addition to the <ahref="https://rclone.org/overview/#restricted-characters">default@@ -61130,6 +64398,11 @@ replaced:</p><p>Invalid UTF-8 bytes will also be <ahref="https://rclone.org/overview/#invalid-utf8">replaced</a>, as theycan't be used in JSON strings.</p>+<p>Seafile stores file and directory names in Unicode NFC form whatever+form they are uploaded in, so rclone normalizes names to NFC before+sending them. Listings therefore always show NFC names, even for files+created with decomposed (NFD) names as macOS produces them. Library+names are not normalized as Seafile keeps them exactly as entered.</p><h3 id="seafile-and-rclone-link">Seafile and rclone link</h3><p>Rclone supports generating share links for non-encrypted librariesonly. They can either be for a file or a directory:</p>@@ -61142,7 +64415,7 @@ http://my.seafile.server/d/9ea2455f6f55478bbb0d/</code></pre><p>Please note a share link is unique for each file or directory. If yourun a link command on a file/dir that has already been shared, you willget the exact same link.</p>-<h3 id="compatibility">Compatibility</h3>+<h3 id="compatibility-1">Compatibility</h3><p>It has been actively developed using the <ahref="https://github.com/haiwen/seafile-docker">seafile docker image</a>of these versions:</p>@@ -61160,7 +64433,7 @@ href="https://hub.docker.com/r/seafileltd/seafile-mc/">latest dockerimage</a> of the seafile community server.</p><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/seafile/seafile.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-56">Standard options</h3>+<h3 id="standard-options-58">Standard options</h3><p>Here are the Standard options specific to seafile (seafile).</p><h4 id="seafile-url">--seafile-url</h4><p>URL of seafile host to connect to.</p>@@ -61232,7 +64505,7 @@ obscure</a>.</p><li>Type: string</li><li>Required: false</li></ul>-<h3 id="advanced-options-57">Advanced options</h3>+<h3 id="advanced-options-59">Advanced options</h3><p>Here are the Advanced options specific to seafile (seafile).</p><h4 id="seafile-create-library">--seafile-create-library</h4><p>Should rclone create a library if it doesn't exist.</p>@@ -61293,7 +64566,7 @@ users to OMIT the leading /.</p><p>Note that by default rclone will try to execute shell commands on theserver, see <a href="#shell-access-considerations">shell accessconsiderations</a>.</p>-<h2 id="configuration-54">Configuration</h2>+<h2 id="configuration-57">Configuration</h2><p>Here is an example of making an SFTP configuration. First run</p><pre class="console"><code>rclone config</code></pre><p>This will guide you through an interactive setup process.</p>@@ -61400,13 +64673,13 @@ provide the path to the user certificate public key file inkey, typically saved as <code>/home/$USER/.ssh/id_rsa.pub</code>.Setting this path in <code>pubkey_file</code> will not work.</p><p>Example:</p>-<div class="sourceCode" id="cb1715"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1715-1"><a href="#cb1715-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[remote]</span></span>-<span id="cb1715-2"><a href="#cb1715-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> sftp</span></span>-<span id="cb1715-3"><a href="#cb1715-3" aria-hidden="true" tabindex="-1"></a><span class="dt">host </span><span class="ot">=</span><span class="st"> example.com</span></span>-<span id="cb1715-4"><a href="#cb1715-4" aria-hidden="true" tabindex="-1"></a><span class="dt">user </span><span class="ot">=</span><span class="st"> sftpuser</span></span>-<span id="cb1715-5"><a href="#cb1715-5" aria-hidden="true" tabindex="-1"></a><span class="dt">key_file </span><span class="ot">=</span><span class="st"> ~/id_rsa</span></span>-<span id="cb1715-6"><a href="#cb1715-6" aria-hidden="true" tabindex="-1"></a><span class="dt">pubkey_file </span><span class="ot">=</span><span class="st"> ~/id_rsa-cert.pub</span></span></code></pre></div>+<div class="sourceCode" id="cb1783"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1783-1"><a href="#cb1783-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[remote]</span></span>+<span id="cb1783-2"><a href="#cb1783-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> sftp</span></span>+<span id="cb1783-3"><a href="#cb1783-3" aria-hidden="true" tabindex="-1"></a><span class="dt">host </span><span class="ot">=</span><span class="st"> example.com</span></span>+<span id="cb1783-4"><a href="#cb1783-4" aria-hidden="true" tabindex="-1"></a><span class="dt">user </span><span class="ot">=</span><span class="st"> sftpuser</span></span>+<span id="cb1783-5"><a href="#cb1783-5" aria-hidden="true" tabindex="-1"></a><span class="dt">key_file </span><span class="ot">=</span><span class="st"> ~/id_rsa</span></span>+<span id="cb1783-6"><a href="#cb1783-6" aria-hidden="true" tabindex="-1"></a><span class="dt">pubkey_file </span><span class="ot">=</span><span class="st"> ~/id_rsa-cert.pub</span></span></code></pre></div><p>If you concatenate a cert with a private key then you can specify themerged file in both places.</p><p>Note: the cert must come first in the file. e.g.</p>@@ -61431,13 +64704,13 @@ automatically with <code>--sftp-pin-host-key</code>.</p><h3 id="using-the-openssh-known_hosts-file">Using the OpenSSHknown_hosts file</h3><p>Using the OpenSSH <code>known_hosts</code> file looks like this:</p>-<div class="sourceCode" id="cb1717"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1717-1"><a href="#cb1717-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[remote]</span></span>-<span id="cb1717-2"><a href="#cb1717-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> sftp</span></span>-<span id="cb1717-3"><a href="#cb1717-3" aria-hidden="true" tabindex="-1"></a><span class="dt">host </span><span class="ot">=</span><span class="st"> example.com</span></span>-<span id="cb1717-4"><a href="#cb1717-4" aria-hidden="true" tabindex="-1"></a><span class="dt">user </span><span class="ot">=</span><span class="st"> sftpuser</span></span>-<span id="cb1717-5"><a href="#cb1717-5" aria-hidden="true" tabindex="-1"></a><span class="dt">pass </span><span class="ot">=</span><span class="st"> </span></span>-<span id="cb1717-6"><a href="#cb1717-6" aria-hidden="true" tabindex="-1"></a><span class="dt">known_hosts_file </span><span class="ot">=</span><span class="st"> ~/.ssh/known_hosts</span></span></code></pre></div>+<div class="sourceCode" id="cb1785"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1785-1"><a href="#cb1785-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[remote]</span></span>+<span id="cb1785-2"><a href="#cb1785-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> sftp</span></span>+<span id="cb1785-3"><a href="#cb1785-3" aria-hidden="true" tabindex="-1"></a><span class="dt">host </span><span class="ot">=</span><span class="st"> example.com</span></span>+<span id="cb1785-4"><a href="#cb1785-4" aria-hidden="true" tabindex="-1"></a><span class="dt">user </span><span class="ot">=</span><span class="st"> sftpuser</span></span>+<span id="cb1785-5"><a href="#cb1785-5" aria-hidden="true" tabindex="-1"></a><span class="dt">pass </span><span class="ot">=</span><span class="st"> </span></span>+<span id="cb1785-6"><a href="#cb1785-6" aria-hidden="true" tabindex="-1"></a><span class="dt">known_hosts_file </span><span class="ot">=</span><span class="st"> ~/.ssh/known_hosts</span></span></code></pre></div><p>Alternatively you can create your own known hosts file like this:</p><pre class="console"><code>ssh-keyscan -t dsa,rsa,ecdsa,ed25519 example.com >> known_hosts</code></pre><p>There are some limitations:</p>@@ -61489,13 +64762,13 @@ server operator.</p><code>--sftp-pin-host-key</code> is ignored.</p><p>After the first successful connection the config will contain a newline:</p>-<div class="sourceCode" id="cb1723"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1723-1"><a href="#cb1723-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[remote]</span></span>-<span id="cb1723-2"><a href="#cb1723-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> sftp</span></span>-<span id="cb1723-3"><a href="#cb1723-3" aria-hidden="true" tabindex="-1"></a><span class="dt">host </span><span class="ot">=</span><span class="st"> example.com</span></span>-<span id="cb1723-4"><a href="#cb1723-4" aria-hidden="true" tabindex="-1"></a><span class="dt">user </span><span class="ot">=</span><span class="st"> sftpuser</span></span>-<span id="cb1723-5"><a href="#cb1723-5" aria-hidden="true" tabindex="-1"></a><span class="dt">pass </span><span class="ot">=</span><span class="st"> </span></span>-<span id="cb1723-6"><a href="#cb1723-6" aria-hidden="true" tabindex="-1"></a><span class="dt">host_keys </span><span class="ot">=</span><span class="st"> ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAA...</span></span></code></pre></div>+<div class="sourceCode" id="cb1791"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1791-1"><a href="#cb1791-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[remote]</span></span>+<span id="cb1791-2"><a href="#cb1791-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> sftp</span></span>+<span id="cb1791-3"><a href="#cb1791-3" aria-hidden="true" tabindex="-1"></a><span class="dt">host </span><span class="ot">=</span><span class="st"> example.com</span></span>+<span id="cb1791-4"><a href="#cb1791-4" aria-hidden="true" tabindex="-1"></a><span class="dt">user </span><span class="ot">=</span><span class="st"> sftpuser</span></span>+<span id="cb1791-5"><a href="#cb1791-5" aria-hidden="true" tabindex="-1"></a><span class="dt">pass </span><span class="ot">=</span><span class="st"> </span></span>+<span id="cb1791-6"><a href="#cb1791-6" aria-hidden="true" tabindex="-1"></a><span class="dt">host_keys </span><span class="ot">=</span><span class="st"> ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAA...</span></span></code></pre></div><p>The <code>host_keys</code> field is always validated against theoffered host key when non-empty, so you can also pin a known key by handwithout ever running with the flag. Each entry is the complete public@@ -61518,7 +64791,7 @@ re-run once with the flag:</p>rclone --sftp-pin-host-key lsd remote:</code></pre><p>Alternatively edit <code>host_keys</code> directly to replace or addthe new entry. Multiple entries (separated by commas) are supported.</p>-<h4 id="limitations-42">Limitations</h4>+<h4 id="limitations-43">Limitations</h4><ul><li>If the server uses an <code>@cert-authority</code>-signed hostcertificate, host key pinning cannot validate it (the certificate is@@ -61600,12 +64873,11 @@ therefore always be correct, and support all features.</p><p>The shell type auto-detection logic, described above, means that bydefault rclone will try to run a shell command the first time a new sftpremote is accessed. If you configure a sftp remote without a config-file, e.g. an <a-href="https://rclone.org/docs/#backend-path-to-dir%5D">on the fly</a>-remote, rclone will have nowhere to store the result, and it will re-run-the command on every access. To avoid this you should explicitly set the-<code>shell_type</code> option to the correct value, or to-<code>none</code> if you want to prevent rclone from executing any+file, e.g. an <a href="https://rclone.org/docs/#backend-path-to-dir">on+the fly</a> remote, rclone will have nowhere to store the result, and it+will re-run the command on every access. To avoid this you should+explicitly set the <code>shell_type</code> option to the correct value,+or to <code>none</code> if you want to prevent rclone from executing anyremote shell commands.</p><p>It is also important to note that, since the shell type decides howquoting and escaping of file paths used as command-line arguments are@@ -61669,7 +64941,7 @@ option <code>hashes</code> to <code>none</code> or options<code>none</code> to not only disable checksumming, but also disable allother functionality that are based on remote shell commandexecution.</p>-<h3 id="modification-times-and-hashes-30">Modification times and+<h3 id="modification-times-and-hashes-32">Modification times andhashes</h3><p>Modified times are stored on the server to 1 second precision.</p><p>Modified times are used in syncing and are fully supported.</p>@@ -61678,6 +64950,38 @@ time after upload (for example, certain configurations of ProFTPd withmod_sftp). If you are using one of these servers, you can set the option<code>set_modtime = false</code> in your RClone backend configuration todisable this behaviour.</p>+<h3 id="restricted-filename-characters-38">Restricted filename+characters</h3>+<p>SFTP servers on Unix-like systems can store any character in a file+name except NUL and <code>/</code>, so by default only these are+replaced:</p>+<table>+<thead>+<tr class="header">+<th>Character</th>+<th style="text-align: center;">Value</th>+<th style="text-align: center;">Replacement</th>+</tr>+</thead>+<tbody>+<tr class="odd">+<td>NUL</td>+<td style="text-align: center;">0x00</td>+<td style="text-align: center;">␀</td>+</tr>+<tr class="even">+<td>/</td>+<td style="text-align: center;">0x2F</td>+<td style="text-align: center;">/</td>+</tr>+</tbody>+</table>+<p>If the server runs on Windows, which can't store many more+characters, add the Windows restricted characters with+<code>--sftp-encoding</code>, as in the <a+href="https://rclone.org/overview/#encoding-example-ftp">FTP+example</a>:</p>+<pre class="text"><code>--sftp-encoding "Slash,LtGt,DoubleQuote,Colon,Question,Asterisk,Pipe,BackSlash,Ctl,RightSpace,RightPeriod,InvalidUtf8,Dot"</code></pre><h3 id="about-command">About command</h3><p>The <code>about</code> command returns the total space, free space,and used space on the remote for the disk of the specified path on the@@ -61694,7 +64998,7 @@ built-in shell command (see <a href="#shell-access">shell access</a>).If none of the above is applicable, <code>about</code> will fail.</p><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/sftp/sftp.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-57">Standard options</h3>+<h3 id="standard-options-59">Standard options</h3><p>Here are the Standard options specific to sftp (SSH/SFTP).</p><h4 id="sftp-host">--sftp-host</h4><p>SSH host to connect to.</p>@@ -61889,7 +65193,7 @@ connection for every hash it calculates.</p><li>Type: SpaceSepList</li><li>Default:</li></ul>-<h3 id="advanced-options-58">Advanced options</h3>+<h3 id="advanced-options-60">Advanced options</h3><p>Here are the Advanced options specific to sftp (SSH/SFTP).</p><h4 id="sftp-known-hosts-file">--sftp-known-hosts-file</h4><p>Optional path to known_hosts file.</p>@@ -62002,7 +65306,7 @@ section in the overview</a> for more info.</p><li>Config: encoding</li><li>Env Var: RCLONE_SFTP_ENCODING</li><li>Type: Encoding</li>-<li>Default: Slash,Del,Ctl,Dot</li>+<li>Default: Slash,Dot</li></ul><h4 id="sftp-set-modtime">--sftp-set-modtime</h4><p>Set the modified time on the remote if set.</p>@@ -62210,6 +65514,28 @@ the performance greatly, especially for distant servers.</p><li>Type: bool</li><li>Default: false</li></ul>+<h4 id="sftp-multithread-upload">--sftp-multithread-upload</h4>+<p>Set this to use multi-thread (multi-connection) uploads.</p>+<p>Normally rclone uploads a single large file over one SFTP connection.+With this set rclone uploads large files over several SFTP connections+at once (one per chunk), which greatly improves throughput on high+latency links, the same way multi-thread downloads already work. Each+connection writes a different, non-overlapping byte range of the file,+so this requires a server that allows several handles to the same file+with writes at arbitrary offsets, which OpenSSH does.</p>+<p>This is off by default because many SFTP servers only accept+sequential writes (for example some object storage backed SFTP gateways)+and would fail large uploads with "truncate failed" or "invalid offset"+style errors.</p>+<p>This is ignored (uploads stay single-threaded) when+--sftp-disable-concurrent-writes or --sftp-connections is in use.</p>+<p>Properties:</p>+<ul>+<li>Config: multithread_upload</li>+<li>Env Var: RCLONE_SFTP_MULTITHREAD_UPLOAD</li>+<li>Type: bool</li>+<li>Default: false</li>+</ul><h4 id="sftp-idle-timeout">--sftp-idle-timeout</h4><p>Max time before closing idle connections.</p><p>If no connections have been returned to the connection pool in the@@ -62414,26 +65740,26 @@ as the source and the destination will be the same file.</p><li>Required: false</li></ul><!-- autogenerated options stop -->-<h2 id="limitations-43">Limitations</h2>+<h2 id="limitations-44">Limitations</h2><p>On some SFTP servers (e.g. Synology) the paths are different for SSHand SFTP so the hashes can't be calculated properly. You can either use-<a href="#--sftp-path-override"><code>--sftp-path-override</code></a> or+<a href="#sftp-path-override"><code>--sftp-path-override</code></a> or<a-href="#--sftp-disable-hashcheck"><code>disable_hashcheck</code></a>.</p>+href="#sftp-disable-hashcheck"><code>disable_hashcheck</code></a>.</p><p>The only ssh agent supported under Windows is Putty's pageant.</p><p>The Go SSH library disables the use of the aes128-cbc cipher bydefault, due to security concerns. This can be re-enabled on aper-connection basis by setting the <code>use_insecure_cipher</code>setting in the configuration file to <code>true</code>. Further detailson the insecurity of this cipher can be found <a-href="http://www.isg.rhul.ac.uk/~kp/SandPfinal.pdf">in this-paper</a>.</p>+href="https://conferences.computer.org/sp/pdfs/sp/2009/oakland2009-01.pdf">in+this paper</a>.</p><p>SFTP isn't supported under plan9 until <ahref="https://github.com/pkg/sftp/issues/156">this issue</a> isfixed.</p><p>Note that since SFTP isn't HTTP based the following flags don't work-with it: <code>--dump-headers</code>, <code>--dump-bodies</code>,-<code>--dump-auth</code>.</p>+with it: <code>--dump headers</code>, <code>--dump bodies</code>,+<code>--dump auth</code>.</p><p>Note that <code>--timeout</code> and <code>--contimeout</code> areboth supported.</p><h2 id="rsync-net">rsync.net</h2>@@ -62462,7 +65788,7 @@ can start with a free account and get 20GB of storage for free.</p><p>Paths are specified as <code>remote:path</code></p><p>Paths may be as deep as required, e.g.<code>remote:directory/subdirectory</code>.</p>-<h2 id="configuration-55">Configuration</h2>+<h2 id="configuration-58">Configuration</h2><p>Here is an example of making a Shade configuration.</p><p>First, create a <a href="https://app.shade.inc/">create a freeaccount</a> account and choose a plan.</p>@@ -62472,55 +65798,55 @@ account and created drive respectively.</p><p>Now run</p><p><code>rclone config</code></p><p>Follow this interactive process:</p>-<div class="sourceCode" id="cb1746"><pre-class="sourceCode sh"><code class="sourceCode bash"><span id="cb1746-1"><a href="#cb1746-1" aria-hidden="true" tabindex="-1"></a><span class="ex">$</span> rclone config</span>-<span id="cb1746-2"><a href="#cb1746-2" aria-hidden="true" tabindex="-1"></a><span class="ex">e</span><span class="er">)</span> <span class="ex">Edit</span> existing remote</span>-<span id="cb1746-3"><a href="#cb1746-3" aria-hidden="true" tabindex="-1"></a><span class="ex">n</span><span class="er">)</span> <span class="ex">New</span> remote</span>-<span id="cb1746-4"><a href="#cb1746-4" aria-hidden="true" tabindex="-1"></a><span class="ex">d</span><span class="er">)</span> <span class="ex">Delete</span> remote</span>-<span id="cb1746-5"><a href="#cb1746-5" aria-hidden="true" tabindex="-1"></a><span class="ex">r</span><span class="er">)</span> <span class="ex">Rename</span> remote</span>-<span id="cb1746-6"><a href="#cb1746-6" aria-hidden="true" tabindex="-1"></a><span class="ex">c</span><span class="er">)</span> <span class="ex">Copy</span> remote</span>-<span id="cb1746-7"><a href="#cb1746-7" aria-hidden="true" tabindex="-1"></a><span class="ex">s</span><span class="er">)</span> <span class="ex">Set</span> configuration password</span>-<span id="cb1746-8"><a href="#cb1746-8" aria-hidden="true" tabindex="-1"></a><span class="ex">q</span><span class="er">)</span> <span class="ex">Quit</span> config</span>-<span id="cb1746-9"><a href="#cb1746-9" aria-hidden="true" tabindex="-1"></a><span class="ex">e/n/d/r/c/s/q</span><span class="op">></span> n</span>-<span id="cb1746-10"><a href="#cb1746-10" aria-hidden="true" tabindex="-1"></a></span>-<span id="cb1746-11"><a href="#cb1746-11" aria-hidden="true" tabindex="-1"></a><span class="ex">Enter</span> name for new remote.</span>-<span id="cb1746-12"><a href="#cb1746-12" aria-hidden="true" tabindex="-1"></a><span class="ex">name</span><span class="op">></span> Shade</span>-<span id="cb1746-13"><a href="#cb1746-13" aria-hidden="true" tabindex="-1"></a></span>-<span id="cb1746-14"><a href="#cb1746-14" aria-hidden="true" tabindex="-1"></a><span class="ex">Option</span> Storage.</span>-<span id="cb1746-15"><a href="#cb1746-15" aria-hidden="true" tabindex="-1"></a><span class="ex">Type</span> of storage to configure.</span>-<span id="cb1746-16"><a href="#cb1746-16" aria-hidden="true" tabindex="-1"></a><span class="ex">Choose</span> a number from below, or type in your own value.</span>-<span id="cb1746-17"><a href="#cb1746-17" aria-hidden="true" tabindex="-1"></a><span class="ex">[OTHER</span> OPTIONS]</span>-<span id="cb1746-18"><a href="#cb1746-18" aria-hidden="true" tabindex="-1"></a><span class="ex">xx</span> / Shade FS</span>-<span id="cb1746-19"><a href="#cb1746-19" aria-hidden="true" tabindex="-1"></a> <span class="ex">\ </span><span class="er">(</span><span class="ex">shade</span><span class="kw">)</span></span>-<span id="cb1746-20"><a href="#cb1746-20" aria-hidden="true" tabindex="-1"></a><span class="ex">[OTHER</span> OPTIONS]</span>-<span id="cb1746-21"><a href="#cb1746-21" aria-hidden="true" tabindex="-1"></a><span class="ex">Storage</span><span class="op">></span> xx</span>-<span id="cb1746-22"><a href="#cb1746-22" aria-hidden="true" tabindex="-1"></a></span>-<span id="cb1746-23"><a href="#cb1746-23" aria-hidden="true" tabindex="-1"></a><span class="ex">Option</span> drive_id.</span>-<span id="cb1746-24"><a href="#cb1746-24" aria-hidden="true" tabindex="-1"></a><span class="ex">The</span> ID of your drive, see this in the drive settings. Individual rclone configs must be made per drive.</span>-<span id="cb1746-25"><a href="#cb1746-25" aria-hidden="true" tabindex="-1"></a><span class="ex">Enter</span> a value.</span>-<span id="cb1746-26"><a href="#cb1746-26" aria-hidden="true" tabindex="-1"></a><span class="ex">drive_id</span><span class="op">></span> <span class="pp">[</span><span class="ss">YOUR_ID</span><span class="pp">]</span></span>-<span id="cb1746-27"><a href="#cb1746-27" aria-hidden="true" tabindex="-1"></a></span>-<span id="cb1746-28"><a href="#cb1746-28" aria-hidden="true" tabindex="-1"></a><span class="ex">Option</span> api_key.</span>-<span id="cb1746-29"><a href="#cb1746-29" aria-hidden="true" tabindex="-1"></a><span class="ex">An</span> API key for your account.</span>-<span id="cb1746-30"><a href="#cb1746-30" aria-hidden="true" tabindex="-1"></a><span class="ex">Enter</span> a value.</span>-<span id="cb1746-31"><a href="#cb1746-31" aria-hidden="true" tabindex="-1"></a><span class="ex">api_key</span><span class="op">></span> <span class="pp">[</span><span class="ss">YOUR_API_KEY</span><span class="pp">]</span></span>-<span id="cb1746-32"><a href="#cb1746-32" aria-hidden="true" tabindex="-1"></a></span>-<span id="cb1746-33"><a href="#cb1746-33" aria-hidden="true" tabindex="-1"></a><span class="ex">Edit</span> advanced config<span class="pp">?</span></span>-<span id="cb1746-34"><a href="#cb1746-34" aria-hidden="true" tabindex="-1"></a><span class="ex">y</span><span class="er">)</span> <span class="ex">Yes</span></span>-<span id="cb1746-35"><a href="#cb1746-35" aria-hidden="true" tabindex="-1"></a><span class="ex">n</span><span class="er">)</span> <span class="ex">No</span> <span class="er">(</span><span class="ex">default</span><span class="kw">)</span></span>-<span id="cb1746-36"><a href="#cb1746-36" aria-hidden="true" tabindex="-1"></a><span class="ex">y/n</span><span class="op">></span> n</span>-<span id="cb1746-37"><a href="#cb1746-37" aria-hidden="true" tabindex="-1"></a></span>-<span id="cb1746-38"><a href="#cb1746-38" aria-hidden="true" tabindex="-1"></a><span class="ex">Configuration</span> complete.</span>-<span id="cb1746-39"><a href="#cb1746-39" aria-hidden="true" tabindex="-1"></a><span class="ex">Options:</span></span>-<span id="cb1746-40"><a href="#cb1746-40" aria-hidden="true" tabindex="-1"></a><span class="ex">-</span> type: shade</span>-<span id="cb1746-41"><a href="#cb1746-41" aria-hidden="true" tabindex="-1"></a><span class="ex">-</span> drive_id: <span class="pp">[</span><span class="ss">YOUR_ID</span><span class="pp">]</span></span>-<span id="cb1746-42"><a href="#cb1746-42" aria-hidden="true" tabindex="-1"></a><span class="ex">-</span> api_key: <span class="pp">[</span><span class="ss">YOUR_API_KEY</span><span class="pp">]</span></span>-<span id="cb1746-43"><a href="#cb1746-43" aria-hidden="true" tabindex="-1"></a><span class="ex">Keep</span> this <span class="st">"Shade"</span> remote<span class="pp">?</span></span>-<span id="cb1746-44"><a href="#cb1746-44" aria-hidden="true" tabindex="-1"></a><span class="ex">y</span><span class="er">)</span> <span class="ex">Yes</span> this is OK <span class="er">(</span><span class="ex">default</span><span class="kw">)</span></span>-<span id="cb1746-45"><a href="#cb1746-45" aria-hidden="true" tabindex="-1"></a><span class="ex">e</span><span class="er">)</span> <span class="ex">Edit</span> this remote</span>-<span id="cb1746-46"><a href="#cb1746-46" aria-hidden="true" tabindex="-1"></a><span class="ex">d</span><span class="er">)</span> <span class="ex">Delete</span> this remote</span>-<span id="cb1746-47"><a href="#cb1746-47" aria-hidden="true" tabindex="-1"></a><span class="ex">y/e/d</span><span class="op">></span> y</span></code></pre></div>-<h3 id="modification-times-and-hashes-31">Modification times and+<div class="sourceCode" id="cb1815"><pre+class="sourceCode sh"><code class="sourceCode bash"><span id="cb1815-1"><a href="#cb1815-1" aria-hidden="true" tabindex="-1"></a><span class="ex">$</span> rclone config</span>+<span id="cb1815-2"><a href="#cb1815-2" aria-hidden="true" tabindex="-1"></a><span class="ex">e</span><span class="er">)</span> <span class="ex">Edit</span> existing remote</span>+<span id="cb1815-3"><a href="#cb1815-3" aria-hidden="true" tabindex="-1"></a><span class="ex">n</span><span class="er">)</span> <span class="ex">New</span> remote</span>+<span id="cb1815-4"><a href="#cb1815-4" aria-hidden="true" tabindex="-1"></a><span class="ex">d</span><span class="er">)</span> <span class="ex">Delete</span> remote</span>+<span id="cb1815-5"><a href="#cb1815-5" aria-hidden="true" tabindex="-1"></a><span class="ex">r</span><span class="er">)</span> <span class="ex">Rename</span> remote</span>+<span id="cb1815-6"><a href="#cb1815-6" aria-hidden="true" tabindex="-1"></a><span class="ex">c</span><span class="er">)</span> <span class="ex">Copy</span> remote</span>+<span id="cb1815-7"><a href="#cb1815-7" aria-hidden="true" tabindex="-1"></a><span class="ex">s</span><span class="er">)</span> <span class="ex">Set</span> configuration password</span>+<span id="cb1815-8"><a href="#cb1815-8" aria-hidden="true" tabindex="-1"></a><span class="ex">q</span><span class="er">)</span> <span class="ex">Quit</span> config</span>+<span id="cb1815-9"><a href="#cb1815-9" aria-hidden="true" tabindex="-1"></a><span class="ex">e/n/d/r/c/s/q</span><span class="op">></span> n</span>+<span id="cb1815-10"><a href="#cb1815-10" aria-hidden="true" tabindex="-1"></a></span>+<span id="cb1815-11"><a href="#cb1815-11" aria-hidden="true" tabindex="-1"></a><span class="ex">Enter</span> name for new remote.</span>+<span id="cb1815-12"><a href="#cb1815-12" aria-hidden="true" tabindex="-1"></a><span class="ex">name</span><span class="op">></span> Shade</span>+<span id="cb1815-13"><a href="#cb1815-13" aria-hidden="true" tabindex="-1"></a></span>+<span id="cb1815-14"><a href="#cb1815-14" aria-hidden="true" tabindex="-1"></a><span class="ex">Option</span> Storage.</span>+<span id="cb1815-15"><a href="#cb1815-15" aria-hidden="true" tabindex="-1"></a><span class="ex">Type</span> of storage to configure.</span>+<span id="cb1815-16"><a href="#cb1815-16" aria-hidden="true" tabindex="-1"></a><span class="ex">Choose</span> a number from below, or type in your own value.</span>+<span id="cb1815-17"><a href="#cb1815-17" aria-hidden="true" tabindex="-1"></a><span class="ex">[OTHER</span> OPTIONS]</span>+<span id="cb1815-18"><a href="#cb1815-18" aria-hidden="true" tabindex="-1"></a><span class="ex">xx</span> / Shade FS</span>+<span id="cb1815-19"><a href="#cb1815-19" aria-hidden="true" tabindex="-1"></a> <span class="ex">\ </span><span class="er">(</span><span class="ex">shade</span><span class="kw">)</span></span>+<span id="cb1815-20"><a href="#cb1815-20" aria-hidden="true" tabindex="-1"></a><span class="ex">[OTHER</span> OPTIONS]</span>+<span id="cb1815-21"><a href="#cb1815-21" aria-hidden="true" tabindex="-1"></a><span class="ex">Storage</span><span class="op">></span> xx</span>+<span id="cb1815-22"><a href="#cb1815-22" aria-hidden="true" tabindex="-1"></a></span>+<span id="cb1815-23"><a href="#cb1815-23" aria-hidden="true" tabindex="-1"></a><span class="ex">Option</span> drive_id.</span>+<span id="cb1815-24"><a href="#cb1815-24" aria-hidden="true" tabindex="-1"></a><span class="ex">The</span> ID of your drive, see this in the drive settings. Individual rclone configs must be made per drive.</span>+<span id="cb1815-25"><a href="#cb1815-25" aria-hidden="true" tabindex="-1"></a><span class="ex">Enter</span> a value.</span>+<span id="cb1815-26"><a href="#cb1815-26" aria-hidden="true" tabindex="-1"></a><span class="ex">drive_id</span><span class="op">></span> <span class="pp">[</span><span class="ss">YOUR_ID</span><span class="pp">]</span></span>+<span id="cb1815-27"><a href="#cb1815-27" aria-hidden="true" tabindex="-1"></a></span>+<span id="cb1815-28"><a href="#cb1815-28" aria-hidden="true" tabindex="-1"></a><span class="ex">Option</span> api_key.</span>+<span id="cb1815-29"><a href="#cb1815-29" aria-hidden="true" tabindex="-1"></a><span class="ex">An</span> API key for your account.</span>+<span id="cb1815-30"><a href="#cb1815-30" aria-hidden="true" tabindex="-1"></a><span class="ex">Enter</span> a value.</span>+<span id="cb1815-31"><a href="#cb1815-31" aria-hidden="true" tabindex="-1"></a><span class="ex">api_key</span><span class="op">></span> <span class="pp">[</span><span class="ss">YOUR_API_KEY</span><span class="pp">]</span></span>+<span id="cb1815-32"><a href="#cb1815-32" aria-hidden="true" tabindex="-1"></a></span>+<span id="cb1815-33"><a href="#cb1815-33" aria-hidden="true" tabindex="-1"></a><span class="ex">Edit</span> advanced config<span class="pp">?</span></span>+<span id="cb1815-34"><a href="#cb1815-34" aria-hidden="true" tabindex="-1"></a><span class="ex">y</span><span class="er">)</span> <span class="ex">Yes</span></span>+<span id="cb1815-35"><a href="#cb1815-35" aria-hidden="true" tabindex="-1"></a><span class="ex">n</span><span class="er">)</span> <span class="ex">No</span> <span class="er">(</span><span class="ex">default</span><span class="kw">)</span></span>+<span id="cb1815-36"><a href="#cb1815-36" aria-hidden="true" tabindex="-1"></a><span class="ex">y/n</span><span class="op">></span> n</span>+<span id="cb1815-37"><a href="#cb1815-37" aria-hidden="true" tabindex="-1"></a></span>+<span id="cb1815-38"><a href="#cb1815-38" aria-hidden="true" tabindex="-1"></a><span class="ex">Configuration</span> complete.</span>+<span id="cb1815-39"><a href="#cb1815-39" aria-hidden="true" tabindex="-1"></a><span class="ex">Options:</span></span>+<span id="cb1815-40"><a href="#cb1815-40" aria-hidden="true" tabindex="-1"></a><span class="ex">-</span> type: shade</span>+<span id="cb1815-41"><a href="#cb1815-41" aria-hidden="true" tabindex="-1"></a><span class="ex">-</span> drive_id: <span class="pp">[</span><span class="ss">YOUR_ID</span><span class="pp">]</span></span>+<span id="cb1815-42"><a href="#cb1815-42" aria-hidden="true" tabindex="-1"></a><span class="ex">-</span> api_key: <span class="pp">[</span><span class="ss">YOUR_API_KEY</span><span class="pp">]</span></span>+<span id="cb1815-43"><a href="#cb1815-43" aria-hidden="true" tabindex="-1"></a><span class="ex">Keep</span> this <span class="st">"Shade"</span> remote<span class="pp">?</span></span>+<span id="cb1815-44"><a href="#cb1815-44" aria-hidden="true" tabindex="-1"></a><span class="ex">y</span><span class="er">)</span> <span class="ex">Yes</span> this is OK <span class="er">(</span><span class="ex">default</span><span class="kw">)</span></span>+<span id="cb1815-45"><a href="#cb1815-45" aria-hidden="true" tabindex="-1"></a><span class="ex">e</span><span class="er">)</span> <span class="ex">Edit</span> this remote</span>+<span id="cb1815-46"><a href="#cb1815-46" aria-hidden="true" tabindex="-1"></a><span class="ex">d</span><span class="er">)</span> <span class="ex">Delete</span> this remote</span>+<span id="cb1815-47"><a href="#cb1815-47" aria-hidden="true" tabindex="-1"></a><span class="ex">y/e/d</span><span class="op">></span> y</span></code></pre></div>+<h3 id="modification-times-and-hashes-33">Modification times andhashes</h3><p>Shade does not support hashes and writing mod times.</p><h3 id="transfers-5">Transfers</h3>@@ -62529,13 +65855,13 @@ be chunked and sent up to Shade concurrently. In order to configure howmany simultaneous uploads you want to use, upload the 'concurrency'option in the advanced config section. Note that this uses more memoryand initiates more http requests.</p>-<h3 id="deleting-files-7">Deleting files</h3>+<h3 id="deleting-files-8">Deleting files</h3><p>Please note that when deleting files in Shade via rclone it willdelete the file instantly, instead of sending it to the trash. Thismeans that it will not be recoverable.</p><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/shade/shade.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-58">Standard options</h3>+<h3 id="standard-options-60">Standard options</h3><p>Here are the Standard options specific to shade (Shade FS).</p><h4 id="shade-drive-id">--shade-drive-id</h4><p>The ID of your drive, see this in the drive settings. Individual@@ -62556,7 +65882,7 @@ rclone configs must be made per drive.</p><li>Type: string</li><li>Required: true</li></ul>-<h3 id="advanced-options-59">Advanced options</h3>+<h3 id="advanced-options-61">Advanced options</h3><p>Here are the Advanced options specific to shade (Shade FS).</p><h4 id="shade-endpoint">--shade-endpoint</h4><p>Endpoint for the service.</p>@@ -62643,7 +65969,7 @@ section in the overview</a> for more info.</p><li>Required: false</li></ul><!-- autogenerated options stop -->-<h2 id="limitations-44">Limitations</h2>+<h2 id="limitations-45">Limitations</h2><p>Note that Shade is case insensitive so you can't have a file called"Hello.doc" and one called "hello.doc".</p><p>Shade only supports filenames up to 255 characters in length.</p>@@ -62688,7 +66014,7 @@ href="https://rclone.org/local/#paths-on-windows">the local backend</a>on Windows can access SMB servers using UNC paths, by<code>\\server\share</code>. This doesn't apply to non-Windows OSes,such as Linux and macOS.</p>-<h2 id="configuration-56">Configuration</h2>+<h2 id="configuration-59">Configuration</h2><p>Here is an example of making a SMB configuration.</p><p>First run</p><pre class="console"><code>rclone config</code></pre>@@ -62765,7 +66091,7 @@ d) Delete this remotey/e/d> d</code></pre><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/smb/smb.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-59">Standard options</h3>+<h3 id="standard-options-61">Standard options</h3><p>Here are the Standard options specific to smb (SMB / CIFS).</p><h4 id="smb-host">--smb-host</h4><p>SMB server hostname to connect to.</p>@@ -62779,12 +66105,13 @@ y/e/d> d</code></pre></ul><h4 id="smb-user">--smb-user</h4><p>SMB username.</p>+<p>Leave blank to use the current user name.</p><p>Properties:</p><ul><li>Config: user</li><li>Env Var: RCLONE_SMB_USER</li><li>Type: string</li>-<li>Default: "$USER"</li>+<li>Required: false</li></ul><h4 id="smb-port">--smb-port</h4><p>SMB port number.</p>@@ -62843,8 +66170,22 @@ KRB5_CONFIG and KRB5CCNAME environment variables.</p><li>Type: bool</li><li>Default: false</li></ul>-<h3 id="advanced-options-60">Advanced options</h3>+<h3 id="advanced-options-62">Advanced options</h3><p>Here are the Advanced options specific to smb (SMB / CIFS).</p>+<h4 id="smb-workstation">--smb-workstation</h4>+<p>Workstation name sent in NTLM authentication.</p>+<p>The NetBIOS name sent as the client workstation in the NTLM+AUTHENTICATE message. Some servers (e.g. domain-joined Windows servers+with per-account workstation restrictions or NTLM allow-lists) require+this to match a specific name. Leave blank to send no workstation name+(the default).</p>+<p>Properties:</p>+<ul>+<li>Config: workstation</li>+<li>Env Var: RCLONE_SMB_WORKSTATION</li>+<li>Type: string</li>+<li>Required: false</li>+</ul><h4 id="smb-idle-timeout">--smb-idle-timeout</h4><p>Max time before closing idle connections.</p><p>If no connections have been returned to the connection pool in the@@ -63035,12 +66376,12 @@ upload</li>gateway</li></ul></li></ul>-<h2 id="configuration-57">Configuration</h2>+<h2 id="configuration-60">Configuration</h2><p>To make a new Storj configuration you need one of the following:</p><ul><li>Access Grant that someone else shared with you.</li><li><a-href="https://documentation.storj.io/getting-started/uploading-your-first-object/create-an-api-key">API+href="https://storj.dev/learn/concepts/access/access-grants/api-key">APIKey</a> of a Storj project you are a member of.</li></ul><p>Here is an example of how to make a remote called@@ -63142,7 +66483,7 @@ d) Delete this remotey/e/d> y</code></pre><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/storj/storj.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-60">Standard options</h3>+<h3 id="standard-options-62">Standard options</h3><p>Here are the Standard options specific to storj (Storj DecentralizedCloud Storage).</p><h4 id="storj-provider">--storj-provider</h4>@@ -63224,7 +66565,7 @@ passphrase.</li><li>Type: string</li><li>Required: false</li></ul>-<h3 id="advanced-options-61">Advanced options</h3>+<h3 id="advanced-options-63">Advanced options</h3><p>Here are the Advanced options specific to storj (Storj DecentralizedCloud Storage).</p><h4 id="storj-description">--storj-description</h4>@@ -63311,7 +66652,7 @@ deleted.</p><pre class="console"><code>rclone sync --interactive --progress remote-us:bucket/path/to/dir/ remote-europe:bucket/path/to/dir/</code></pre><p>Or even between another cloud storage and Storj.</p><pre class="console"><code>rclone sync --interactive --progress s3:bucket/path/to/dir/ storj:bucket/path/to/dir/</code></pre>-<h2 id="limitations-45">Limitations</h2>+<h2 id="limitations-46">Limitations</h2><p><code>rclone about</code> is not supported by the rclone Storjbackend. Backends without this capability cannot determine free spacefor an rclone mount or use policy <code>mfs</code> (most free space) as@@ -63340,7 +66681,7 @@ operating system manual.</p><p><a href="https://sugarsync.com">SugarSync</a> is a cloud service thatenables active synchronization of files across computers and otherdevices for file backup, access, syncing, and sharing.</p>-<h2 id="configuration-58">Configuration</h2>+<h2 id="configuration-61">Configuration</h2><p>The initial setup for SugarSync involves getting a token fromSugarSync which you can do with rclone. <code>rclone config</code> walksyou through it.</p>@@ -63413,13 +66754,13 @@ store them, it only uses them to get the initial token.</p><p><strong>NB</strong> you can't create files in the top level folderyou have to create a folder, which rclone will create as a "Sync Folder"with SugarSync.</p>-<h3 id="modification-times-and-hashes-32">Modification times and+<h3 id="modification-times-and-hashes-34">Modification times andhashes</h3><p>SugarSync does not support modification times or hashes, thereforesyncing will default to <code>--size-only</code> checking. Note thatusing <code>--update</code> will work as rclone can read the time fileswere uploaded.</p>-<h3 id="restricted-filename-characters-35">Restricted filename+<h3 id="restricted-filename-characters-39">Restricted filenamecharacters</h3><p>SugarSync replaces the <ahref="https://rclone.org/overview/#restricted-characters">default@@ -63427,7 +66768,7 @@ restricted characters set</a> except for DEL.</p><p>Invalid UTF-8 bytes will also be <ahref="https://rclone.org/overview/#invalid-utf8">replaced</a>, as theycan't be used in XML strings.</p>-<h3 id="deleting-files-8">Deleting files</h3>+<h3 id="deleting-files-9">Deleting files</h3><p>Deleted files will be moved to the "Deleted items" folder bydefault.</p><p>However you can supply the flag <code>--sugarsync-hard-delete</code>@@ -63435,7 +66776,7 @@ or set the config parameter <code>hard_delete = true</code> if you wouldlike files to be deleted straight away.</p><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/sugarsync/sugarsync.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-61">Standard options</h3>+<h3 id="standard-options-63">Standard options</h3><p>Here are the Standard options specific to sugarsync (Sugarsync).</p><h4 id="sugarsync-app-id">--sugarsync-app-id</h4><p>Sugarsync App ID.</p>@@ -63478,7 +66819,7 @@ files.</p><li>Type: bool</li><li>Default: false</li></ul>-<h3 id="advanced-options-62">Advanced options</h3>+<h3 id="advanced-options-64">Advanced options</h3><p>Here are the Advanced options specific to sugarsync (Sugarsync).</p><h4 id="sugarsync-refresh-token">--sugarsync-refresh-token</h4><p>Sugarsync refresh token.</p>@@ -63562,7 +66903,7 @@ section in the overview</a> for more info.</p><li>Required: false</li></ul><!-- autogenerated options stop -->-<h2 id="limitations-46">Limitations</h2>+<h2 id="limitations-47">Limitations</h2><p><code>rclone about</code> is not supported by the SugarSync backend.Backends without this capability cannot determine free space for anrclone mount or use policy <code>mfs</code> (most free space) as a@@ -63576,7 +66917,7 @@ href="https://rclone.org/commands/rclone_about/">rclone about</a>.</p><code>remote:directory/subdirectory</code>.</p><p>The initial setup for Uloz.to involves filling in the usercredentials. <code>rclone config</code> walks you through it.</p>-<h2 id="configuration-59">Configuration</h2>+<h2 id="configuration-62">Configuration</h2><p>Here is an example of how to make a remote called<code>remote</code>. First run:</p><pre class="console"><code>rclone config</code></pre>@@ -63641,7 +66982,7 @@ y/e/d> y</code></pre>and password. Uloz.to offers an API key as well, but it's reserved forthe use of Uloz.to's in-house application and using it in differentcircumstances is unreliable.</p>-<h3 id="modification-times-and-hashes-33">Modification times and+<h3 id="modification-times-and-hashes-35">Modification times andhashes</h3><p>Uloz.to doesn't allow the user to set a custom modification time, orretrieve the hashes after upload. As a result, the integration uses a@@ -63651,7 +66992,7 @@ and hashes. Timestamps are stored with microsecond precision.</p>Afterwards, the backend only serves the client-side calculated hashes.Hashes can also be retrieved upon creating a file download link, butit's impractical for <code>list</code>-like use cases.</p>-<h3 id="restricted-filename-characters-36">Restricted filename+<h3 id="restricted-filename-characters-40">Restricted filenamecharacters</h3><p>In addition to the <ahref="https://rclone.org/overview/#restricted-characters">default@@ -63681,12 +67022,12 @@ can't be used in JSON strings.</p>uploading large files a stable connection is necessary. Rclone willupload up to <code>--transfers</code> chunks at the same time (sharedamong all uploads).</p>-<h3 id="deleting-files-9">Deleting files</h3>+<h3 id="deleting-files-10">Deleting files</h3><p>By default, files are moved to the recycle bin whereas folders aredeleted immediately. Trashed files are permanently deleted after 30 daysin the recycle bin.</p><p>Emptying the trash is currently not implemented in rclone.</p>-<h3 id="root-folder-id-6">Root folder ID</h3>+<h3 id="root-folder-id-7">Root folder ID</h3><p>You can set the <code>root_folder_slug</code> for rclone. This is thefolder (identified by its <code>Folder slug</code>) that rcloneconsiders to be the root of your Uloz.to drive.</p>@@ -63707,7 +67048,7 @@ in the remote path. For example, if your remote's<code>ABSOLUTE_ULOZTO_ROOT/foo/bar/baz/qux</code>.</p><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/ulozto/ulozto.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-62">Standard options</h3>+<h3 id="standard-options-64">Standard options</h3><p>Here are the Standard options specific to ulozto (Uloz.to).</p><h4 id="ulozto-app-token">--ulozto-app-token</h4><p>The application token identifying the app. An app API key can be@@ -63741,7 +67082,7 @@ obscure</a>.</p><li>Type: string</li><li>Required: false</li></ul>-<h3 id="advanced-options-63">Advanced options</h3>+<h3 id="advanced-options-65">Advanced options</h3><p>Here are the Advanced options specific to ulozto (Uloz.to).</p><h4 id="ulozto-root-folder-slug">--ulozto-root-folder-slug</h4><p>If set, rclone will use this folder as the root folder for all@@ -63784,7 +67125,7 @@ section in the overview</a> for more info.</p><li>Required: false</li></ul><!-- autogenerated options stop -->-<h2 id="limitations-47">Limitations</h2>+<h2 id="limitations-48">Limitations</h2><p>Uloz.to file names can't have the <code>\</code> character in. rclonemaps this to and from an identical looking unicode equivalent<code>\</code> (U+FF3C Fullwidth Reverse Solidus).</p>@@ -63831,7 +67172,7 @@ named <code>backup</code> with the remotessegments. Invoking <code>rclone mkdir backup:../desktop</code> isexactly the same as invoking<code>rclone mkdir mydrive:private/backup/../desktop</code>.</p>-<h2 id="configuration-60">Configuration</h2>+<h2 id="configuration-63">Configuration</h2><p>Here is an example of how to make a union called <code>remote</code>for local folders. First run:</p><pre class="console"><code>rclone config</code></pre>@@ -63992,7 +67333,7 @@ used only if your upstreams support the respective quota fields.</p><p>To check if your upstream supports the field, run<code>rclone about remote: [flags]</code> and see if the required fieldexists.</p>-<h3 id="filters-2">Filters</h3>+<h3 id="filters-3">Filters</h3><p>Policies basically search upstream remotes and create a list of files/ paths for functions to work on. The policy is responsible forfiltering and sorting. The policy type defines the sorting but filtering@@ -64110,13 +67451,13 @@ upstream.</td><h3 id="writeback">Writeback</h3><p>The tag <code>:writeback</code> on an upstream remote can be used tomake a simple cache system like this:</p>-<div class="sourceCode" id="cb1787"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1787-1"><a href="#cb1787-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[union]</span></span>-<span id="cb1787-2"><a href="#cb1787-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> union</span></span>-<span id="cb1787-3"><a href="#cb1787-3" aria-hidden="true" tabindex="-1"></a><span class="dt">action_policy </span><span class="ot">=</span><span class="st"> all</span></span>-<span id="cb1787-4"><a href="#cb1787-4" aria-hidden="true" tabindex="-1"></a><span class="dt">create_policy </span><span class="ot">=</span><span class="st"> all</span></span>-<span id="cb1787-5"><a href="#cb1787-5" aria-hidden="true" tabindex="-1"></a><span class="dt">search_policy </span><span class="ot">=</span><span class="st"> ff</span></span>-<span id="cb1787-6"><a href="#cb1787-6" aria-hidden="true" tabindex="-1"></a><span class="dt">upstreams </span><span class="ot">=</span><span class="st"> /local:writeback remote:dir</span></span></code></pre></div>+<div class="sourceCode" id="cb1856"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1856-1"><a href="#cb1856-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[union]</span></span>+<span id="cb1856-2"><a href="#cb1856-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> union</span></span>+<span id="cb1856-3"><a href="#cb1856-3" aria-hidden="true" tabindex="-1"></a><span class="dt">action_policy </span><span class="ot">=</span><span class="st"> all</span></span>+<span id="cb1856-4"><a href="#cb1856-4" aria-hidden="true" tabindex="-1"></a><span class="dt">create_policy </span><span class="ot">=</span><span class="st"> all</span></span>+<span id="cb1856-5"><a href="#cb1856-5" aria-hidden="true" tabindex="-1"></a><span class="dt">search_policy </span><span class="ot">=</span><span class="st"> ff</span></span>+<span id="cb1856-6"><a href="#cb1856-6" aria-hidden="true" tabindex="-1"></a><span class="dt">upstreams </span><span class="ot">=</span><span class="st"> /local:writeback remote:dir</span></span></code></pre></div><p>When files are opened for read, if the file is in<code>remote:dir</code> but not <code>/local</code> then rclone willcopy the file entirely into <code>/local</code> before returning a@@ -64133,7 +67474,7 @@ other than writing files back to it. So if you need to expire old filesor manage the size then you will have to do this yourself.</p><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/union/union.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-63">Standard options</h3>+<h3 id="standard-options-65">Standard options</h3><p>Here are the Standard options specific to union (Union merges thecontents of several upstream fs).</p><h4 id="union-upstreams">--union-upstreams</h4>@@ -64184,7 +67525,7 @@ dir" upstreamb:', etc.</p><li>Type: int</li><li>Default: 120</li></ul>-<h3 id="advanced-options-64">Advanced options</h3>+<h3 id="advanced-options-66">Advanced options</h3><p>Here are the Advanced options specific to union (Union merges thecontents of several upstream fs).</p><h4 id="union-min-free-space">--union-min-free-space</h4>@@ -64217,7 +67558,7 @@ for more info.</p><p>Paths are specified as <code>remote:path</code></p><p>Paths may be as deep as required, e.g.<code>remote:directory/subdirectory</code>.</p>-<h2 id="configuration-61">Configuration</h2>+<h2 id="configuration-64">Configuration</h2><p>To configure the WebDAV remote you will need to have a URL for it,and a username and password. If you know what kind of system you areconnecting to then rclone can enable extra features.</p>@@ -64295,7 +67636,7 @@ y/e/d> y</code></pre><pre class="console"><code>rclone ls remote:</code></pre><p>To copy a local directory to an WebDAV directory called backup</p><pre class="console"><code>rclone copy /home/source remote:backup</code></pre>-<h3 id="modification-times-and-hashes-34">Modification times and+<h3 id="modification-times-and-hashes-36">Modification times andhashes</h3><p>Plain WebDAV does not support modified times. However when used withFastmail Files, ownCloud or Nextcloud rclone will support modified@@ -64304,10 +67645,12 @@ times.</p>Fastmail Files, ownCloud or Nextcloud rclone will support SHA1 and MD5hashes. Depending on the exact version of ownCloud or Nextcloud hashesmay appear on all objects, or only on objects which had a hash uploaded-with them.</p>+with them. With Nextcloud, rclone asks the server to calculate the SHA1+of uploads which had no hash to send, such as streamed uploads, and+after setting the modification time, which discards the stored hash.</p><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/webdav/webdav.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-64">Standard options</h3>+<h3 id="standard-options-66">Standard options</h3><p>Here are the Standard options specific to webdav (WebDAV).</p><h4 id="webdav-url">--webdav-url</h4><p>URL of http host to connect to.</p>@@ -64397,7 +67740,7 @@ obscure</a>.</p><li>Type: string</li><li>Required: false</li></ul>-<h3 id="advanced-options-65">Advanced options</h3>+<h3 id="advanced-options-67">Advanced options</h3><p>Here are the Advanced options specific to webdav (WebDAV).</p><h4 id="webdav-bearer-token-command">--webdav-bearer-token-command</h4><p>Command to run to get a bearer token.</p>@@ -64432,6 +67775,12 @@ used.</p>'"Cookie","name=value"'.</p><p>You can set multiple headers, e.g.'"Cookie","name=value","Authorization","xxx"'.</p>+<p>The headers are only sent to the host in the configured URL. If the+server redirects to another host (including a subdomain or a different+port) the headers are not sent to it, or to any further hop in that+redirect chain. The exception is the directory listing and file download+requests when auth_redirect is set, as that keeps all credentials across+redirects.</p><p>Properties:</p><ul><li>Config: headers</li>@@ -64543,7 +67892,7 @@ header.</p><p>This is configured in an identical way to ownCloud. Note thatNextcloud initially did not support streaming of files(<code>rcat</code>) whereas ownCloud did, but <a-href="https://github.com/nextcloud/nextcloud-snap/issues/365">this</a>+href="https://github.com/nextcloud-snap/nextcloud-snap/issues/365">this</a>seems to be fixed as of 2020-11-27 (tested with rclone v1.53.1 andNextcloud Server v19).</p><h3 id="owncloud-infinite-scale">ownCloud Infinite Scale</h3>@@ -64579,13 +67928,13 @@ and use your normal account email and password for <code>user</code> and<code>pass</code>. If you have 2FA enabled, you have to generate an apppassword. Set the <code>vendor</code> to <code>sharepoint</code>.</p><p>Your config file should look like this:</p>-<div class="sourceCode" id="cb1793"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1793-1"><a href="#cb1793-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[sharepoint]</span></span>-<span id="cb1793-2"><a href="#cb1793-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> webdav</span></span>-<span id="cb1793-3"><a href="#cb1793-3" aria-hidden="true" tabindex="-1"></a><span class="dt">url </span><span class="ot">=</span><span class="st"> https://[YOUR-DOMAIN]-my.sharepoint.com/personal/[YOUR-EMAIL]/Documents</span></span>-<span id="cb1793-4"><a href="#cb1793-4" aria-hidden="true" tabindex="-1"></a><span class="dt">vendor </span><span class="ot">=</span><span class="st"> sharepoint</span></span>-<span id="cb1793-5"><a href="#cb1793-5" aria-hidden="true" tabindex="-1"></a><span class="dt">user </span><span class="ot">=</span><span class="st"> YourEmailAddress</span></span>-<span id="cb1793-6"><a href="#cb1793-6" aria-hidden="true" tabindex="-1"></a><span class="dt">pass </span><span class="ot">=</span><span class="st"> encryptedpassword</span></span></code></pre></div>+<div class="sourceCode" id="cb1862"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1862-1"><a href="#cb1862-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[sharepoint]</span></span>+<span id="cb1862-2"><a href="#cb1862-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> webdav</span></span>+<span id="cb1862-3"><a href="#cb1862-3" aria-hidden="true" tabindex="-1"></a><span class="dt">url </span><span class="ot">=</span><span class="st"> https://[YOUR-DOMAIN]-my.sharepoint.com/personal/[YOUR-EMAIL]/Documents</span></span>+<span id="cb1862-4"><a href="#cb1862-4" aria-hidden="true" tabindex="-1"></a><span class="dt">vendor </span><span class="ot">=</span><span class="st"> sharepoint</span></span>+<span id="cb1862-5"><a href="#cb1862-5" aria-hidden="true" tabindex="-1"></a><span class="dt">user </span><span class="ot">=</span><span class="st"> YourEmailAddress</span></span>+<span id="cb1862-6"><a href="#cb1862-6" aria-hidden="true" tabindex="-1"></a><span class="dt">pass </span><span class="ot">=</span><span class="st"> encryptedpassword</span></span></code></pre></div><h3 id="sharepoint-with-ntlm-authentication">Sharepoint with NTLMAuthentication</h3><p>Use this option in case your (hosted) Sharepoint is not tied to@@ -64603,13 +67952,13 @@ class="uri">https://example.sharepoint.com/sites/12345/Documents</a></p><p>NTLM uses domain and user name combination for authentication, set<code>user</code> to <code>DOMAIN\username</code>.</p><p>Your config file should look like this:</p>-<div class="sourceCode" id="cb1794"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1794-1"><a href="#cb1794-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[sharepoint]</span></span>-<span id="cb1794-2"><a href="#cb1794-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> webdav</span></span>-<span id="cb1794-3"><a href="#cb1794-3" aria-hidden="true" tabindex="-1"></a><span class="dt">url </span><span class="ot">=</span><span class="st"> https://[YOUR-DOMAIN]/some-path-to/Documents</span></span>-<span id="cb1794-4"><a href="#cb1794-4" aria-hidden="true" tabindex="-1"></a><span class="dt">vendor </span><span class="ot">=</span><span class="st"> sharepoint-ntlm</span></span>-<span id="cb1794-5"><a href="#cb1794-5" aria-hidden="true" tabindex="-1"></a><span class="dt">user </span><span class="ot">=</span><span class="st"> DOMAIN\user</span></span>-<span id="cb1794-6"><a href="#cb1794-6" aria-hidden="true" tabindex="-1"></a><span class="dt">pass </span><span class="ot">=</span><span class="st"> encryptedpassword</span></span></code></pre></div>+<div class="sourceCode" id="cb1863"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1863-1"><a href="#cb1863-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[sharepoint]</span></span>+<span id="cb1863-2"><a href="#cb1863-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> webdav</span></span>+<span id="cb1863-3"><a href="#cb1863-3" aria-hidden="true" tabindex="-1"></a><span class="dt">url </span><span class="ot">=</span><span class="st"> https://[YOUR-DOMAIN]/some-path-to/Documents</span></span>+<span id="cb1863-4"><a href="#cb1863-4" aria-hidden="true" tabindex="-1"></a><span class="dt">vendor </span><span class="ot">=</span><span class="st"> sharepoint-ntlm</span></span>+<span id="cb1863-5"><a href="#cb1863-5" aria-hidden="true" tabindex="-1"></a><span class="dt">user </span><span class="ot">=</span><span class="st"> DOMAIN\user</span></span>+<span id="cb1863-6"><a href="#cb1863-6" aria-hidden="true" tabindex="-1"></a><span class="dt">pass </span><span class="ot">=</span><span class="st"> encryptedpassword</span></span></code></pre></div><h4 id="required-flags-for-sharepoint">Required Flags forSharePoint</h4><p>As SharePoint does some special things with uploaded documents, you@@ -64622,7 +67971,8 @@ datetime property to compare your documents:</p><pre class="text"><code>--ignore-size --ignore-checksum --update</code></pre><h2 id="rclone-1">Rclone</h2><p>Use this option if you are hosting remotes over WebDAV provided by-rclone. Read <a href="commands/rclone_serve_webdav/">rclone serve+rclone. Read <a+href="https://rclone.org/commands/rclone_serve_webdav/">rclone servewebdav</a> for more details.</p><p>rclone serve supports modified times using the<code>X-OC-Mtime</code> header.</p>@@ -64639,14 +67989,14 @@ access tokens.</p>username or password, instead enter your Macaroon as the<code>bearer_token</code>.</p><p>The config will end up looking something like this.</p>-<div class="sourceCode" id="cb1796"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1796-1"><a href="#cb1796-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[dcache]</span></span>-<span id="cb1796-2"><a href="#cb1796-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> webdav</span></span>-<span id="cb1796-3"><a href="#cb1796-3" aria-hidden="true" tabindex="-1"></a><span class="dt">url </span><span class="ot">=</span><span class="st"> https://dcache...</span></span>-<span id="cb1796-4"><a href="#cb1796-4" aria-hidden="true" tabindex="-1"></a><span class="dt">vendor </span><span class="ot">=</span><span class="st"> other</span></span>-<span id="cb1796-5"><a href="#cb1796-5" aria-hidden="true" tabindex="-1"></a><span class="dt">user </span><span class="ot">=</span></span>-<span id="cb1796-6"><a href="#cb1796-6" aria-hidden="true" tabindex="-1"></a><span class="dt">pass </span><span class="ot">=</span></span>-<span id="cb1796-7"><a href="#cb1796-7" aria-hidden="true" tabindex="-1"></a><span class="dt">bearer_token </span><span class="ot">=</span><span class="st"> your-macaroon</span></span></code></pre></div>+<div class="sourceCode" id="cb1865"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1865-1"><a href="#cb1865-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[dcache]</span></span>+<span id="cb1865-2"><a href="#cb1865-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> webdav</span></span>+<span id="cb1865-3"><a href="#cb1865-3" aria-hidden="true" tabindex="-1"></a><span class="dt">url </span><span class="ot">=</span><span class="st"> https://dcache...</span></span>+<span id="cb1865-4"><a href="#cb1865-4" aria-hidden="true" tabindex="-1"></a><span class="dt">vendor </span><span class="ot">=</span><span class="st"> other</span></span>+<span id="cb1865-5"><a href="#cb1865-5" aria-hidden="true" tabindex="-1"></a><span class="dt">user </span><span class="ot">=</span></span>+<span id="cb1865-6"><a href="#cb1865-6" aria-hidden="true" tabindex="-1"></a><span class="dt">pass </span><span class="ot">=</span></span>+<span id="cb1865-7"><a href="#cb1865-7" aria-hidden="true" tabindex="-1"></a><span class="dt">bearer_token </span><span class="ot">=</span><span class="st"> your-macaroon</span></span></code></pre></div><p>There is a <ahref="https://github.com/sara-nl/GridScripts/blob/master/get-macaroon">script</a>that obtains a Macaroon from a dCache WebDAV endpoint, and creates an@@ -64686,16 +68036,16 @@ the advanced config and enter the command to get a bearer token (e.g.,<p>The following example config shows a WebDAV endpoint that usesoidc-agent to supply an access token from the <em>XDC</em> OIDCProvider.</p>-<div class="sourceCode" id="cb1798"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1798-1"><a href="#cb1798-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[dcache]</span></span>-<span id="cb1798-2"><a href="#cb1798-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> webdav</span></span>-<span id="cb1798-3"><a href="#cb1798-3" aria-hidden="true" tabindex="-1"></a><span class="dt">url </span><span class="ot">=</span><span class="st"> https://dcache.example.org/</span></span>-<span id="cb1798-4"><a href="#cb1798-4" aria-hidden="true" tabindex="-1"></a><span class="dt">vendor </span><span class="ot">=</span><span class="st"> other</span></span>-<span id="cb1798-5"><a href="#cb1798-5" aria-hidden="true" tabindex="-1"></a><span class="dt">bearer_token_command </span><span class="ot">=</span><span class="st"> oidc-token XDC</span></span></code></pre></div>+<div class="sourceCode" id="cb1867"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1867-1"><a href="#cb1867-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[dcache]</span></span>+<span id="cb1867-2"><a href="#cb1867-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> webdav</span></span>+<span id="cb1867-3"><a href="#cb1867-3" aria-hidden="true" tabindex="-1"></a><span class="dt">url </span><span class="ot">=</span><span class="st"> https://dcache.example.org/</span></span>+<span id="cb1867-4"><a href="#cb1867-4" aria-hidden="true" tabindex="-1"></a><span class="dt">vendor </span><span class="ot">=</span><span class="st"> other</span></span>+<span id="cb1867-5"><a href="#cb1867-5" aria-hidden="true" tabindex="-1"></a><span class="dt">bearer_token_command </span><span class="ot">=</span><span class="st"> oidc-token XDC</span></span></code></pre></div><h1 id="yandex-disk">Yandex Disk</h1><p><a href="https://disk.yandex.com">Yandex Disk</a> is a cloud storagesolution created by <a href="https://yandex.com">Yandex</a>.</p>-<h2 id="configuration-62">Configuration</h2>+<h2 id="configuration-65">Configuration</h2><p>Here is an example of making a yandex configuration. First run</p><pre class="console"><code>rclone config</code></pre><p>This will guide you through an interactive setup process:</p>@@ -64759,7 +68109,7 @@ any excess files in the path.</p><pre class="console"><code>rclone sync --interactive /home/local/directory remote:directory</code></pre><p>Yandex paths may be as deep as required, e.g.<code>remote:directory/subdirectory</code>.</p>-<h3 id="modification-times-and-hashes-35">Modification times and+<h3 id="modification-times-and-hashes-37">Modification times andhashes</h3><p>Modified times are supported and are stored accurate to 1 ns incustom metadata called <code>rclone_modified</code> in RFC3339 with@@ -64774,7 +68124,7 @@ arguments.</p><p>To view your current quota you can use the<code>rclone about remote:</code> command which will display your usagelimit (quota) and the current usage.</p>-<h3 id="restricted-filename-characters-37">Restricted filename+<h3 id="restricted-filename-characters-41">Restricted filenamecharacters</h3><p>The <ahref="https://rclone.org/overview/#restricted-characters">default@@ -64782,9 +68132,19 @@ restricted characters set</a> are replaced.</p><p>Invalid UTF-8 bytes will also be <ahref="https://rclone.org/overview/#invalid-utf8">replaced</a>, as theycan't be used in JSON strings.</p>+<h3 id="application-folder">Application folder</h3>+<p>If you registered your own OAuth application with Yandex and only+granted it the "Application Folder" permission+(<code>cloud_api:disk.app_folder</code>) instead of full disk access,+then you must set the <code>app_folder</code> option to+<code>true</code>. Yandex Disk requires paths to be addressed with an+<code>app:/</code> prefix instead of the usual <code>disk:/</code>+prefix when using this restricted scope, and rclone will return a 403+Forbidden error on every request if this option isn't set correctly for+the token in use.</p><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/yandex/yandex.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-65">Standard options</h3>+<h3 id="standard-options-67">Standard options</h3><p>Here are the Standard options specific to yandex (Yandex Disk).</p><h4 id="yandex-client-id">--yandex-client-id</h4><p>OAuth Client Id.</p>@@ -64806,7 +68166,7 @@ can't be used in JSON strings.</p><li>Type: string</li><li>Required: false</li></ul>-<h3 id="advanced-options-66">Advanced options</h3>+<h3 id="advanced-options-68">Advanced options</h3><p>Here are the Advanced options specific to yandex (Yandex Disk).</p><h4 id="yandex-token">--yandex-token</h4><p>OAuth Access Token as a JSON blob.</p>@@ -64896,6 +68256,22 @@ normally enough to stop them, at the cost of slowing down uploads.</p><li>Type: Duration</li><li>Default: 0s</li></ul>+<h4 id="yandex-app-folder">--yandex-app-folder</h4>+<p>Use the application folder as the root.</p>+<p>If you registered your own OAuth application with Yandex with only+the "Application Folder" permission (cloud_api:disk.app_folder) instead+of full disk access, then rclone needs to address paths with the "app:/"+prefix instead of the usual "disk:/" prefix, otherwise all requests fail+with a 403 Forbidden error.</p>+<p>Set this to true if your OAuth token only has access to the+application folder.</p>+<p>Properties:</p>+<ul>+<li>Config: app_folder</li>+<li>Env Var: RCLONE_YANDEX_APP_FOLDER</li>+<li>Type: bool</li>+<li>Default: false</li>+</ul><h4 id="yandex-description">--yandex-description</h4><p>Description of the remote.</p><p>Properties:</p>@@ -64906,7 +68282,7 @@ normally enough to stop them, at the cost of slowing down uploads.</p><li>Required: false</li></ul><!-- autogenerated options stop -->-<h2 id="limitations-48">Limitations</h2>+<h2 id="limitations-49">Limitations</h2><p>When uploading very large files (bigger than about 5 GiB) you willneed to increase the <code>--timeout</code> parameter. This is becauseYandex pauses (perhaps to calculate the MD5SUM for the entire file)@@ -64933,7 +68309,7 @@ Rclone won't be able to complete any actions.</p><p><a href="https://www.zoho.com/workdrive/">Zoho WorkDrive</a> is acloud storage solution created by <ahref="https://zoho.com">Zoho</a>.</p>-<h2 id="configuration-63">Configuration</h2>+<h2 id="configuration-66">Configuration</h2><p>Here is an example of making a zoho configuration. First run</p><pre class="console"><code>rclone config</code></pre><p>This will guide you through an interactive setup process:</p>@@ -65016,7 +68392,7 @@ any excess files in the path.</p><pre class="console"><code>rclone sync --interactive /home/local/directory remote:directory</code></pre><p>Zoho paths may be as deep as required, eg<code>remote:directory/subdirectory</code>.</p>-<h3 id="modification-times-and-hashes-36">Modification times and+<h3 id="modification-times-and-hashes-38">Modification times andhashes</h3><p>Modified times are currently not supported for Zoho Workdrive</p><p>No hash algorithms are supported.</p>@@ -65024,14 +68400,14 @@ hashes</h3><p>To view your current quota you can use the<code>rclone about remote:</code> command which will display yourcurrent usage.</p>-<h3 id="restricted-filename-characters-38">Restricted filename+<h3 id="restricted-filename-characters-42">Restricted filenamecharacters</h3><p>Only control characters and invalid UTF-8 are replaced. In additionmost Unicode full-width characters are not supported at all and will beremoved from filenames during upload.</p><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/zoho/zoho.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="standard-options-66">Standard options</h3>+<h3 id="standard-options-68">Standard options</h3><p>Here are the Standard options specific to zoho (Zoho).</p><h4 id="zoho-client-id">--zoho-client-id</h4><p>OAuth Client Id.</p>@@ -65092,7 +68468,7 @@ browser.</p></ul></li></ul></li></ul>-<h3 id="advanced-options-67">Advanced options</h3>+<h3 id="advanced-options-69">Advanced options</h3><p>Here are the Advanced options specific to zoho (Zoho).</p><h4 id="zoho-token">--zoho-token</h4><p>OAuth Access Token as a JSON blob.</p>@@ -65297,12 +68673,12 @@ enable it in other regions.</p></li><pre class="console"><code>rclone sync --interactive /home/source /tmp/destination</code></pre><p>Will sync <code>/home/source</code> to<code>/tmp/destination</code>.</p>-<h2 id="configuration-64">Configuration</h2>+<h2 id="configuration-67">Configuration</h2><p>For consistencies sake one can also configure a remote of type<code>local</code> in the config file, and access the local filesystemusing rclone remote paths, e.g. <code>remote:path/to/wherever</code>,but it is probably easier not to.</p>-<h3 id="modification-times-10">Modification times</h3>+<h3 id="modification-times-11">Modification times</h3><p>Rclone reads and writes the modification times using an accuracydetermined by the OS. Typically this is 1ns on Linux, 10 ns on Windowsand 1 Second on OS X.</p>@@ -65594,7 +68970,7 @@ These only get replaced if they are the last character in the name:</p><p>Invalid UTF-8 bytes will also be <ahref="https://rclone.org/overview/#invalid-utf8">replaced</a>, as theycan't be converted to UTF-16.</p>-<h3 id="paths-on-windows">Paths on Windows</h3>+<h3 id="paths-on-windows-1">Paths on Windows</h3><p>On Windows there are many ways of specifying a path to a file systemresource. Local paths can be absolute, like<code>C:\path\to\wherever</code>, or relative, like@@ -65628,15 +69004,15 @@ drivers like <ahref="https://github.com/rclone/rclone/issues/261">EncFS</a>. To disableUNC conversion globally, add this to your <code>.rclone.conf</code>file:</p>-<div class="sourceCode" id="cb1814"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1814-1"><a href="#cb1814-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[local]</span></span>-<span id="cb1814-2"><a href="#cb1814-2" aria-hidden="true" tabindex="-1"></a><span class="dt">nounc </span><span class="ot">=</span><span class="st"> </span><span class="kw">true</span></span></code></pre></div>+<div class="sourceCode" id="cb1883"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1883-1"><a href="#cb1883-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[local]</span></span>+<span id="cb1883-2"><a href="#cb1883-2" aria-hidden="true" tabindex="-1"></a><span class="dt">nounc </span><span class="ot">=</span><span class="st"> </span><span class="kw">true</span></span></code></pre></div><p>If you want to selectively disable UNC, you can add it to a separateentry like this:</p>-<div class="sourceCode" id="cb1815"><pre-class="sourceCode ini"><code class="sourceCode ini"><span id="cb1815-1"><a href="#cb1815-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[nounc]</span></span>-<span id="cb1815-2"><a href="#cb1815-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> local</span></span>-<span id="cb1815-3"><a href="#cb1815-3" aria-hidden="true" tabindex="-1"></a><span class="dt">nounc </span><span class="ot">=</span><span class="st"> </span><span class="kw">true</span></span></code></pre></div>+<div class="sourceCode" id="cb1884"><pre+class="sourceCode ini"><code class="sourceCode ini"><span id="cb1884-1"><a href="#cb1884-1" aria-hidden="true" tabindex="-1"></a><span class="kw">[nounc]</span></span>+<span id="cb1884-2"><a href="#cb1884-2" aria-hidden="true" tabindex="-1"></a><span class="dt">type </span><span class="ot">=</span><span class="st"> local</span></span>+<span id="cb1884-3"><a href="#cb1884-3" aria-hidden="true" tabindex="-1"></a><span class="dt">nounc </span><span class="ot">=</span><span class="st"> </span><span class="kw">true</span></span></code></pre></div><p>And use rclone like this:</p><p><code>rclone copy c:\src nounc:z:\dst</code></p><p>This will use UNC paths on <code>c:\src</code> but not on@@ -65671,6 +69047,8 @@ this</p>6 two/three6 b/two6 b/one</code></pre>+<p>Symlinks which point to a parent directory would loop forever, so+rclone reports them as errors and skips them.</p><h4 id="local-links---links--l">--local-links, --links, -l</h4><p>Normally rclone will ignore symlinks or junction points (which behavelike symlinks under Windows).</p>@@ -65705,6 +69083,9 @@ $ tree /tmp/b/tmp/b├── file1 -> ./file4└── file2 -> /home/user/file3</code></pre>+<p>A <code>.rclonelink</code> file whose contents are too long to be a+path (more than 128 KiB) is refused rather than turned into a+symlink.</p><p>However, if copied back without '-l'</p><pre class="console"><code>$ rclone copyto remote:/tmp/a/ /tmp/b/@@ -65781,7 +69162,7 @@ systems. On systems where it isn't supported (e.g. Windows) it will beignored.</p><!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/local/local.go and run make backenddocs to verify --><!-- markdownlint-disable-line line-length -->-<h3 id="advanced-options-68">Advanced options</h3>+<h3 id="advanced-options-70">Advanced options</h3><p>Here are the Advanced options specific to local (Local Disk).</p><h4 id="local-nounc">--local-nounc</h4><p>Disable UNC (long path names) conversion on Windows.</p>@@ -65959,6 +69340,12 @@ deduplicated by default (i.e. having two identical files does notconsume more storage than having just one.) However, for use cases wheredata redundancy is preferable, --local-no-clone can be used to disablecloning and force "deep" copies.</p>+<p>Cloning also preserves macOS-specific metadata such as extended+attributes, which a normal copy does not, so this metadata is lost when+cloning is disabled.</p>+<p>If the destination file already exists, cloning replaces it with a+new file, even with --inplace, so hard links to it are broken. Use+--local-no-clone to update existing files in place.</p><p>Currently, cloning is only supported when using APFS on macOS(support for other platforms may be added in the future.)</p><p>Properties:</p>@@ -66033,10 +69420,10 @@ set this flag.</p></ul><h4 id="local-fatal-if-no-space">--local-fatal-if-no-space</h4><p>Make out-of-space errors fatal during transfers.</p>-<p>When enabled, an ENOSPC error during a write returns a fatal error so-that rclone aborts rather than retrying the operation. Useful for backup-scripts that should halt loudly on a full disk rather than spin-retrying.</p>+<p>When enabled, an out of space error while writing, creating a+directory, or creating a file returns a fatal error so that rclone+aborts rather than retrying the operation. Useful for backup scripts+that should halt loudly on a full disk rather than spin retrying.</p><p>Properties:</p><ul><li>Config: fatal_if_no_space</li>@@ -66130,8 +69517,10 @@ supported by all file systems) under the "user.*" prefix.</p><p>When restoring metadata with <code>--metadata</code> rclone appliesthe "mode", "uid" and "gid" from the source. These come from the sourceremote which may not be trusted, so restoring metadata as root from an-untrusted source can change file ownership and is not recommended. The-setuid, setgid and sticky bits are not restored by default - see the+untrusted source can change file ownership and is not recommended. A+user other than root keeps its own uid and sets the gid only to a group+it is a member of, as rsync does. The setuid, setgid and sticky bits are+not restored by default - see the<code>--local-metadata-restore-special-bits</code> flag.</p><p>Here are the possible system metadata items for the localbackend.</p>@@ -66229,6 +69618,1330 @@ the output.</p><!-- autogenerated options stop --><!-- markdownlint-disable line-length --><h1 id="changelog-1">Changelog</h1>+<h2 id="v1.76.0---2026-11-06">v1.76.0 - 2026-11-06</h2>+<p><a+href="https://github.com/rclone/rclone/compare/v1.75.0...v1.76.0">See+commits</a></p>+<ul>+<li>New backends+<ul>+<li><a href="https://rclone.org/dosya/">Dosya</a> (Kaya)</li>+<li><a href="https://rclone.org/kdrive/">Infomaniak kDrive</a>+(Guillaume MALLERET)</li>+</ul></li>+<li>New S3 Providers+<ul>+<li><a href="https://rclone.org/s3/#autonomys">Autonomys Auto Drive</a>+(Jim Counter)</li>+<li><a href="https://rclone.org/s3/#clever-cloud-cellar">Clever Cloud+Cellar</a> (David Legrand)</li>+<li><a href="https://rclone.org/s3/#infomaniak">Infomaniak Swiss+Backup</a> (Louis)</li>+<li><a href="https://rclone.org/s3/#neon">Neon Object Storage</a> (Rishi+Raj Jain)</li>+<li><a href="https://rclone.org/s3/#relaix">RelAix.Storage</a>+(mygrexit)</li>+</ul></li>+<li>New commands+<ul>+<li><a href="https://rclone.org/commands/rclone_index/">index</a>: Write+static directory listings into a remote (Nick Craig-Wood)</li>+</ul></li>+<li>Security+<ul>+<li>build+<ul>+<li>Disable Multipath TCP listeners by default to mitigate+CVE-2026-98432 (Nick Craig-Wood)</li>+<li>Fix multiple CVEs including CVE-2026-111943 by upgrading to go1.27.3+(Nick Craig-Wood)</li>+<li>Update <code>golang.org/x/net</code> to v0.62.0 to fix multiple CVEs+(Nick Craig-Wood)</li>+<li>Update <code>github.com/andybalholm/brotli</code> to v1.2.7 to fix+CVE-2026-111497 (Nick Craig-Wood)</li>+</ul></li>+<li>fshttp: Refuse HTTPS to HTTP redirects unless+<code>--allow-https-downgrade</code> GHSA-m97j-7x65-wh93 CVE-PENDING+(Nick Craig-Wood)</li>+<li>lib/http: Stop <code>--htpasswd</code> and <code>--user</code>+revealing which usernames exist through timing GHSA-hwj7-wp9g-wrv9+CVE-PENDING (Nick Craig-Wood)</li>+<li>local+<ul>+<li>Fix user metadata escaping the root through a planted symlink+GHSA-r5wv-vqqh-2r9c CVE-PENDING (Nick Craig-Wood)</li>+<li>Fix MkdirMetadata finding directories outside the root through a+planted symlink (Nick Craig-Wood)</li>+<li>Fix Remove and Rmdir deleting outside the root through a planted+symlink (Nick Craig-Wood)</li>+<li>Fix DirMove creating the destination parent through a planted+symlink (Nick Craig-Wood)</li>+</ul></li>+<li>netstorage: Fix API auth headers leaking to other hosts on redirect+GHSA-m97j-7x65-wh93 CVE-PENDING (Nick Craig-Wood)</li>+<li>rc+<ul>+<li>Limit request size when the rc server has no authentication+GHSA-f8p4-8g4j-q3m2 CVE-PENDING (Nick Craig-Wood)</li>+<li>Reject unauthorised and unknown calls before reading the request+GHSA-f8p4-8g4j-q3m2 CVE-PENDING (Nick Craig-Wood)</li>+</ul></li>+<li>seafile: Fix HTTPS to HTTP redirects leaking credentials+GHSA-m97j-7x65-wh93 CVE-PENDING (Nick Craig-Wood)</li>+<li>serve: Make auth proxy revocation take effect within 5 minutes+GHSA-vmgx-g9c5-669p CVE-PENDING (Nick Craig-Wood)</li>+<li>serve sftp+<ul>+<li>Fix auth proxy letting a public key be used as a password+GHSA-66qv-66mg-9h74 CVE-PENDING (Nick Craig-Wood)</li>+<li>Fix auth proxy running for public keys the client can't use+GHSA-66qv-66mg-9h74 CVE-PENDING (Nick Craig-Wood)</li>+</ul></li>+<li>swift+<ul>+<li>Fix auth token leaking to other hosts or over plaintext on redirect+GHSA-m97j-7x65-wh93 CVE-PENDING (Nick Craig-Wood)</li>+<li>Fix password being resent to another host on auth redirect+GHSA-m97j-7x65-wh93 CVE-PENDING (Nick Craig-Wood)</li>+</ul></li>+<li>ulozto: Fix auth tokens leaking on redirects from download links+GHSA-3xpx-hjg2-vrjh CVE-PENDING (Nick Craig-Wood)</li>+</ul></li>+<li>New Features+<ul>+<li>accounting+<ul>+<li>Stop <code>--max-transfer</code> overshooting the limit (Nick+Craig-Wood)</li>+<li>Add "Updated dirs" count of directory modtime/metadata updates to+the stats (Nick Craig-Wood)</li>+</ul></li>+<li>batcher: Log when committing uploads has finished (phatlc)</li>+<li>bisync+<ul>+<li>Make <code>--max-delete</code> aware of <code>--track-renames</code>+(Morax)</li>+<li>Fall back to <code>--download</code> when a checksum is missing+(nielash)</li>+<li>Reject <code>--max-delete</code> above 100 (nielash)</li>+</ul></li>+<li>build+<ul>+<li>Update to go1.27 and make go1.26 the minimum required version (Nick+Craig-Wood)</li>+<li>Update all dependencies (Nick Craig-Wood)</li>+</ul></li>+<li>cmd: Complete enum flag values from their choices (Demiao+Huang)</li>+<li>config: Only run <code>--password-command</code> once when using+<code>--daemon</code> (phatlc)</li>+<li>doc fixes (ashwinibailangadi, ferrumclaudepilgrim, frodleTian,+FTCHD, gomitrah, Nick Craig-Wood, qodelicious, Satvik Mishra,+ZainnQureshii)</li>+<li>docker: Make container user UID and GID configurable via build ARGs+(Vaibhav Mashal)</li>+<li>filter+<ul>+<li>Speed up large <code>--include-from</code> lists of directories+(kaii9)</li>+<li>Make <code>[!...]</code> in globs a negated character class+(kevin9327)</li>+<li>Make <code>[^...]</code> in globs exclude <code>/</code> and+document it (kevin9327)</li>+</ul></li>+<li>fs+<ul>+<li>Reject sizes too large for SizeSuffix instead of silently disabling+the limit (KBS)</li>+<li>Stop writing config for on the fly and environment variable remotes+into the config file (Nick Craig-Wood)</li>+<li>Reject durations like 300y that overflow instead of wrapping+(kevin9327)</li>+<li>Accept multi-unit relative times like 1d12h in Time options+(kevin)</li>+<li>Reject NaN sizes and CountSuffix values that overflow+(kevin9327)</li>+<li>Reject <code>--bwlimit</code> times without a colon or with signed+digits (kevin9327)</li>+<li>Reject NaN durations instead of wrapping (Soujas K)</li>+<li>Add MIME type for JPEG XL (.jxl) files (Seth Westphal)</li>+</ul></li>+<li>gui+<ul>+<li>Take the GUI out of beta (Nick Craig-Wood)</li>+<li>Update embedded release to 1.2.1 (Nick Craig-Wood)</li>+</ul></li>+<li>hash: Add CRC-32C (Castagnoli) support (Dirk Petersen)</li>+<li>index: Add partial runs which only re-index changed directories+(Nick Craig-Wood)</li>+<li>lib/http+<ul>+<li>Return 416 for ranges <code>--rc-serve</code> and serve restic can't+satisfy (Rohit Behera)</li>+<li>Replace <code>github.com/abbot/go-http-auth</code> with built in+htpasswd support (Nick Craig-Wood)</li>+<li>Accept the Basic auth scheme in any case (kevin)</li>+</ul></li>+<li>lib/proxy: Send the decoded user and password in Proxy-Authorization+(kevin)</li>+<li>lib/transform: Make titlecase capitalise words instead of+upper-casing them (kevin)</li>+<li>ncdu: Show progress and the item count when deleting a selection+(Lev Devaev)</li>+<li>operations+<ul>+<li>Check checksums in rcat with known size (phatlc)</li>+<li>Speed up checksum by hashing files in parallel (Teddy Tennant)</li>+<li>Fall back to a single stream when a multi-thread source ignores+ranges (Morax)</li>+<li>Show directory operations in the <code>--progress</code> display+(Nick Craig-Wood)</li>+<li>Add <code>operations/getfile</code> remote control API endpoint+(Hakan İSMAİL)</li>+<li>Make <code>--immutable</code> work with copyto, moveto and single+file copy (phatlc)</li>+<li>Respect <code>--backup-dir</code> in deletefile (Cao Yuhang)</li>+</ul></li>+<li>rc+<ul>+<li>Deprecate <code>--rc-web-gui</code> flag (Nick Craig-Wood)</li>+<li>Add sync report parameters to <code>sync/sync</code>,+<code>sync/copy</code> and <code>sync/move</code> (Nick Craig-Wood)</li>+</ul></li>+<li>rest: Reject negative sizes in HTTP response headers (Yousef+Issa)</li>+<li>serve http+<ul>+<li>Sort directory listings in the browser (Nick Craig-Wood)</li>+<li>Redesign the directory listing page (Nick Craig-Wood)</li>+</ul></li>+<li>serve s3+<ul>+<li>Report the object's modtime as Last-Modified after a PUT+(kevin9327)</li>+<li>Support changing object tags (Nick Craig-Wood)</li>+<li>Quote the ETags of objects in listings (Nick Craig-Wood)</li>+<li>Support listing with delimiters other than "/" (Nick+Craig-Wood)</li>+<li>Support directory markers and refuse file/directory conflicts (Nick+Craig-Wood)</li>+<li>Make the ETag of multipart uploads agree with HEAD and support+partNumber (Nick Craig-Wood)</li>+<li>Make conditional writes atomic against other writes through the+server (Nick Craig-Wood)</li>+<li>Make the ETag returned on upload agree with HEAD for any+<code>--etag-hash</code> (Nick Craig-Wood)</li>+<li>List empty directories as directory markers (Nick Craig-Wood)</li>+<li>Allow copying directory markers (Nick Craig-Wood)</li>+<li>Remove the directories made for a failed upload (Nick+Craig-Wood)</li>+<li>Keep the MD5 ETag of objects on remotes without MD5 (Nick+Craig-Wood)</li>+<li>Don't read directories needlessly when listing with a delimiter+(Nick Craig-Wood)</li>+<li>Don't compute the MD5 of multipart uploads unless the ETag is MD5+(Nick Craig-Wood)</li>+<li>List the directory markers put for directories holding other keys+(Nick Craig-Wood)</li>+<li>Let a failed CompleteMultipartUpload be retried (Nick+Craig-Wood)</li>+<li>List objects with the storage class they were put with (Nick+Craig-Wood)</li>+<li>Add <code>--directory-markers</code> to list every directory as a+marker (Nick Craig-Wood)</li>+<li>Remove a directory with its last key after its marker is deleted+(Nick Craig-Wood)</li>+<li>Make <code>--no-cleanup</code> a deprecated alias of+<code>--directory-markers</code> (Nick Craig-Wood)</li>+<li>Document which S3 features are supported and the design goals (Nick+Craig-Wood)</li>+</ul></li>+<li>serve webdav+<ul>+<li>Sort directory listings in the browser (Nick Craig-Wood)</li>+<li>Redesign the directory listing page (Nick Craig-Wood)</li>+</ul></li>+<li>size: Include source in JSON output (Jiacheng Xu)</li>+<li>sync: Add pattern-based transfer ordering (Shurong Cao)</li>+<li>touch: Parse documented timestamp formats (Nicholas Velten)</li>+</ul></li>+<li>Bug Fixes+<ul>+<li>accounting: Fix totalBytes counting server-side copies and moves+twice (ilya lavrenov)</li>+<li>bisync+<ul>+<li>Fix conflict winner missing from listings when Path2 wins+(nielash)</li>+<li>Fix <code>--conflict-resolve</code> comparing a zero size or modtime+(nielash)</li>+<li>Fix endless retry loop for skipped files (nielash)</li>+<li>Fix listings ignoring a modtime less than 1s older (nielash)</li>+<li>Fix rc ignoring global <code>--dry-run</code> and+<code>--max-delete</code> settings (nielash)</li>+<li>Fix rc ignoring legacy backupdir1/2 and invalid values+(nielash)</li>+<li>Fix internal error when a conflict rename reuses deleted names+(TastyHeadphones)</li>+</ul></li>+<li>fs: Fix durations like 1d12h being rejected (Rohit Behera)</li>+<li>lib/encoder: Fix decoding a trailing character after an escaped+quote (kevin9327)</li>+<li>list: Fix ext sort error sometimes being ignored at the end of a+listing (Nick Craig-Wood)</li>+<li>ncdu: Update to tcell v3 which fixes duplicated keys on Windows+(Nick Craig-Wood)</li>+<li>operations+<ul>+<li>Fix making a directory with modtime asking twice with+<code>--interactive</code> (Nick Craig-Wood)</li>+<li>Fix case- or Unicode-only renames deleting the file (nielash)</li>+<li>Fix moving a file onto its own name deleting it (nielash)</li>+<li>Fix cat <code>--offset</code> before the start of a file printing+its end (b84330808)</li>+<li>Fix oversized getfile tail requests (唐僧洗发用飘柔)</li>+</ul></li>+<li>rc: Fix LogLevel in <code>_config</code> being ignored (Nick+Craig-Wood)</li>+<li>rest: Fix retries reading the request body after the request has+finished (Nick Craig-Wood)</li>+<li>serve dlna: Fix crash on Browse with a negative StartingIndex or+RequestedCount (phatlc)</li>+<li>serve restic: Fix temporary errors confusing restic into leaving+locks behind (Azhar)</li>+<li>serve s3+<ul>+<li>Update to gofakes3 v0.0.8 which fixes 500 errors with go1.27 (Nick+Craig-Wood)</li>+<li>Upgrade gofakes3 to v0.0.11 to fix IsTruncated and Docker+Distribution Registry (DEVN)</li>+<li>Fix buckets being undeletable after an aborted multipart upload+(Nick Craig-Wood)</li>+<li>Fix CopyObject metadata including the source's and an mtime (Nick+Craig-Wood)</li>+</ul></li>+<li>serve sftp: Fix rmdir deleting a file instead of returning an error+(phatlc)</li>+<li>sync+<ul>+<li>Fix directory modtimes being set twice (Nick Craig-Wood)</li>+<li>Fix <code>--delete-empty-src-dirs</code> disabled by a prior error+in a reused stats group (ghwany)</li>+</ul></li>+<li>walk: Fix duplicate directories with <code>--fast-list</code> when+directories are listed after their contents (Nick Craig-Wood)</li>+</ul></li>+<li>Mount+<ul>+<li>Fix VFS instance leaks on mount failures and unmounts (Hakan+İSMAİL)</li>+</ul></li>+<li>VFS+<ul>+<li>Preserve shrinking truncate on uncached files (47th)</li>+<li>Fix crash renaming a directory while the dir cache expires+(Shreyansh Jain)</li>+<li>Fix virtual entries leak in removed directory (Biagio+Peccerillo)</li>+<li>Fix renamed items listed twice after a case or Unicode-only match+(nielash)</li>+<li>Fix uploads never starting while a dirty file keeps being read (Neil+Cawse)</li>+</ul></li>+<li>Local+<ul>+<li>Speed up default checksummed copies by writing in larger blocks+(Dominik Sander)</li>+<li>Make out of space errors fatal during multi-thread transfers+(ferrumclaudepilgrim)</li>+<li>Stop <code>--copy-links</code> following symlink loops (phatlc)</li>+<li>Limit the size of symlink targets read from .rclonelink files (Nick+Craig-Wood)</li>+<li>Fix server-side copy on macOS when the destination exists (Ayush+Sharma)</li>+<li>Set only the ownership a non-root user may set (Dirk Petersen)</li>+</ul></li>+<li>Crypt+<ul>+<li>Fix nil pointer panic in ObjectInfo.Hash (fixfon)</li>+</ul></li>+<li>Archive+<ul>+<li>Read symlinks stored in zip and squashfs archives (Lev Devaev)</li>+<li>Fix reading an archive in the root of a remote (Oshino Penguin)</li>+</ul></li>+<li>Azure Blob+<ul>+<li>Enable on Solaris (acoeur)</li>+<li>Fix directories on storage accounts with hierarchical namespace+(Nick Craig-Wood)</li>+<li>Fix NewObject on a directory returning "is not a regular file" (Nick+Craig-Wood)</li>+<li>Fix listing a file as a directory on hierarchical namespace accounts+(Nick Craig-Wood)</li>+<li>Fix sync at the root failing on soft deleted containers (Ludovic+Bondon)</li>+</ul></li>+<li>Box+<ul>+<li>Reuse multipart upload part buffers via the global pool (Nick+Craig-Wood)</li>+<li>Fix ~ and environment variables not being expanded in+<code>box_config_file</code> (Nick Craig-Wood)</li>+</ul></li>+<li>Chunker+<ul>+<li>Clean stale chunks after server-side overwrite+(hahahahahayesyeseys)</li>+</ul></li>+<li>Compress+<ul>+<li>Buffer small uploads in the global memory pool (Nick+Craig-Wood)</li>+<li>Stop buffering the compressibility check output in memory (Nick+Craig-Wood)</li>+</ul></li>+<li>Drime+<ul>+<li>Fix endless listing of directories with more than 20,000 entries+(Maksym Tymoshyk)</li>+</ul></li>+<li>Drive+<ul>+<li>Fetch metadata permissions concurrently (Loi Nguyen)</li>+<li>Reuse resumable-upload chunk buffers via the global memory pool+(SillyZir)</li>+<li>Stop a fast-list error from hanging the sync (Sash)</li>+</ul></li>+<li>Dropbox+<ul>+<li>Reuse upload chunk buffers via the global memory pool (Nick+Craig-Wood)</li>+</ul></li>+<li>Filelu+<ul>+<li>Reuse multipart upload buffers via the global pool and retry failed+parts (Nick Craig-Wood)</li>+<li>Encode multipart init paths (wjc)</li>+</ul></li>+<li>FTP+<ul>+<li>Discard connections that fail with non-FTP errors (Daniel Bae)</li>+<li>Fix ranged reads failing when the server answers 425 to an early+close (Robert Kubiś)</li>+</ul></li>+<li>Hasher+<ul>+<li>Keep cached hash when only the modtime changes (Aniket)</li>+</ul></li>+<li>Hidrive+<ul>+<li>Stop copying every upload chunk a second time (Nick Craig-Wood)</li>+<li>Buffer the start of each upload from the global memory pool (Nick+Craig-Wood)</li>+<li>Reuse chunked upload buffers via the global memory pool (Nick+Craig-Wood)</li>+<li>Buffer small file updates from the global memory pool (Nick+Craig-Wood)</li>+<li>Fix truncated file left behind when an upload hits+<code>--max-transfer</code> (Nick Craig-Wood)</li>+</ul></li>+<li>HTTP+<ul>+<li>Add metadata defaults for <code>--http-no-head</code>+(Giridhar)</li>+</ul></li>+<li>Huaweidrive+<ul>+<li>Build small uploads in pooled memory instead of a growing buffer+(Nick Craig-Wood)</li>+<li>Reuse resumable-upload chunk buffers via the global pool (Nick+Craig-Wood)</li>+<li>Fix creating, copying and renaming files with non-ASCII names (Nick+Craig-Wood)</li>+</ul></li>+<li>Iclouddrive+<ul>+<li>Mark the backend as case insensitive (berce)</li>+</ul></li>+<li>Imagekit+<ul>+<li>Stop multipart writers after failed uploads (Florian Josef+Altendorfer)</li>+</ul></li>+<li>Jottacloud+<ul>+<li>Buffer MD5 pre-reads in the global memory pool (Nick+Craig-Wood)</li>+<li>Fix server-side copy onto a trashed file failing as corrupted (Nick+Craig-Wood)</li>+<li>Fix server-side copy onto a trashed file keeping the old name case+(Nick Craig-Wood)</li>+<li>Fix authentication with whitelabel services elkjop and elko+(albertony)</li>+<li>Fix traditional authentication failing with invalid_client+(albertony)</li>+</ul></li>+<li>Linkbox+<ul>+<li>Buffer the hashed 10 MiB file prefix in the global memory pool (Nick+Craig-Wood)</li>+</ul></li>+<li>Mailru+<ul>+<li>Buffer speedup hashing in the global memory pool and fix upload+retries (Nick Craig-Wood)</li>+</ul></li>+<li>Onedrive+<ul>+<li>OneDrive for Business and SharePoint users must upgrade to v1.76++before April 1, 2027 (Nick Craig-Wood)</li>+<li>Authenticate SharePoint downloads, uploads and copies after tempauth+retirement (Nick Craig-Wood)</li>+<li>Return an error if a ranged download doesn't return the requested+range (Nick Craig-Wood)</li>+<li>Reuse upload chunk buffers via the global memory pool (Nick+Craig-Wood)</li>+<li>Allow <code>--onedrive-upload-cutoff</code> up to the documented+limit of 250 MiB (Can Arslan)</li>+<li>Add configurable SharePoint tenant API version (Mattias+Michaux)</li>+<li>Make <code>--onedrive-delta</code> quicker when not listing from the+drive root (Nick Craig-Wood)</li>+<li>Keep the delta token when a ChangeNotify poll fails+(xiaozou-wine)</li>+<li>Show the size of directories in listings (GingerAdonis)</li>+<li>Poll async copy status with the configured HTTP client (Soujas+K)</li>+</ul></li>+<li>Opendrive+<ul>+<li>Reuse upload chunk buffers via the global memory pool (Nick+Craig-Wood)</li>+</ul></li>+<li>Oracleobjectstorage+<ul>+<li>Add server-side object moves (tomaszni)</li>+<li>Encode carriage returns and line feeds (tomaszni)</li>+<li>Purge objects in batches (tomaszni)</li>+<li>Retry 409 ConcurrentObjectUpdate errors (berce)</li>+</ul></li>+<li>Pcloud+<ul>+<li>Stop multipart writers after failed uploads (Florian Josef+Altendorfer)</li>+</ul></li>+<li>Pikpak+<ul>+<li>Share upload chunk buffers with the global memory pool (Nick+Craig-Wood)</li>+<li>Use pooled memory for the gcid hash buffer and spool unknown-size+uploads to disk (Nick Craig-Wood)</li>+<li>Fix downloads that return an HTTP error page as the file+(Cestercian)</li>+<li>Stop multipart writers after failed uploads (Florian Josef+Altendorfer)</li>+<li>Fix uploads failing when the server is slow to complete them (Nick+Craig-Wood)</li>+</ul></li>+<li>Putio+<ul>+<li>Reuse upload chunk buffers via the global memory pool (Nick+Craig-Wood)</li>+</ul></li>+<li>S3+<ul>+<li>Avoid buffering Object Lock uploads when the source MD5 is known+(Nick Craig-Wood)</li>+<li>Add backend link command with signed response header overrides+(eliotee)</li>+<li>Add Fastly EU Central 1 region (Leon Brocard)</li>+<li>Add the MD5 of multipart chunks to the debug log+(interested.tortoise)</li>+<li>Report the real storage class when an object needs restoring (Dirk+Petersen)</li>+<li>Speed up multi-thread uploads by checksumming chunks in 1 MiB reads+(Roman Butor)</li>+<li>Add IT-MIL region to Scaleway (Alessio Margelli)</li>+<li>Add missing Huawei OBS regions (Enbiya Göral)</li>+<li>Fix rmdir of the root directory with+<code>--s3-directory-markers</code> (Nick Craig-Wood)</li>+<li>Fix presigned URLs for GCS, Linode and Ceph (Leon Brocard)</li>+</ul></li>+<li>Seafile+<ul>+<li>Fix operations on files whose names are not in Unicode NFC form+(Fred)</li>+<li>Stop multipart writers after failed uploads (Florian Josef+Altendorfer)</li>+<li>Fix double slash in the URL when getting a 2FA token (Nick+Craig-Wood)</li>+</ul></li>+<li>SFTP+<ul>+<li>Implement multi-thread uploads (Splainte)</li>+<li>Fix data race when using <code>--sftp-ssh</code> (Dirk+Petersen)</li>+<li>Fix hash check for names starting with "-" with relative+<code>--sftp-path-override</code> (Dirk Einecke)</li>+<li>Fix sync deleting files with control characters in their names+(nielash)</li>+<li>Fix hashing files with a newline in their name on rsync.net (Nick+Craig-Wood)</li>+</ul></li>+<li>Shade+<ul>+<li>Stop copying each upload chunk into a second heap buffer (Nick+Craig-Wood)</li>+</ul></li>+<li>Smb+<ul>+<li>Add workstation field to smb backend (Kalin Stoyanov)</li>+</ul></li>+<li>Swift+<ul>+<li>Add Infomaniak Swiss Backup provider (Louis)</li>+</ul></li>+<li>WebDAV+<ul>+<li>Reject ranged responses which ignore the range (Morax)</li>+<li>Retry Nextcloud chunk uploads individually (Stochastic)</li>+<li>Fix Nextcloud public share access by sending X-Requested-With+(drakeo338)</li>+<li>Fix memory leak uploading to Infinite Scale (jxj)</li>+</ul></li>+<li>Yandex+<ul>+<li>Add <code>app_folder</code> option to support+<code>cloud_api:disk.app_folder</code> OAuth scope (Sanjay Kanth A)</li>+</ul></li>+<li>Zoho+<ul>+<li>Stop multipart writers after failed uploads (Florian Josef+Altendorfer)</li>+</ul></li>+</ul>+<h2 id="v1.75.2---2026-10-09">v1.75.2 - 2026-10-09</h2>+<p><a+href="https://github.com/rclone/rclone/compare/v1.75.1...v1.75.2">See+commits</a></p>+<ul>+<li>Security+<ul>+<li>b2: Fix account token and SSE-C key leaking on redirects from+<code>--b2-download-url</code> GHSA-gpw7-77cq-9f72 CVE-2026-116196 (Nick+Craig-Wood)</li>+<li>build+<ul>+<li>Fix multiple CVEs by upgrading to go1.26.9 (Nick Craig-Wood)+<ul>+<li>CVE-2026-97032: net/http: HTTP/2 server crash due to HPACK encoder+race</li>+<li>CVE-2026-78659: net/http: HTTP/2 server memory exhaustion due to+Trailer headers</li>+<li>CVE-2026-97031: crypto/tls: reject malformed ECH outer extension+references</li>+<li>CVE-2026-94444: cmd/go: checksum bypass for golang.org/fips140</li>+<li>CVE-2026-94447: cmd/go: checksum database bypass for+golang.org/toolchain</li>+<li>CVE-2026-94448: html/template: reset context tracking on consecutive+template expressions</li>+<li>CVE-2026-97030: html/template: recognize |yield| as regexp preceder+keyword</li>+<li>CVE-2026-94440: net/textproto, mime/multipart: memory limit bypass+when parsing MIME headers</li>+<li>CVE-2026-56866: net/http: HTTP/1 client connection desynchronization+after CONNECT rejection</li>+<li>CVE-2026-94439: net/http: HTTP/1 server connection desynchronization+after 2xx CONNECT response</li>+<li>CVE-2026-78669: net/http: excessive CPU consumption from repeated+initial window changes</li>+<li>CVE-2026-78660: net/http: HTTP/2 transport accepts malformed+framing-related headers</li>+<li>CVE-2026-56857: os: Root.Mkdir(All) can follow junctions out of the+root on Windows</li>+<li>CVE-2026-78667: net/http: lack of limit on size of parsed Range+headers</li>+<li>CVE-2026-78663: net/http: double flow control refund on HTTP/2+server streams</li>+</ul></li>+<li>Update golang.org/x/net to v0.60.0 to fix multiple CVEs (Nick+Craig-Wood)+<ul>+<li>CVE-2026-78659: net/http: HTTP/2 server memory exhaustion due to+Trailer headers</li>+<li>CVE-2026-78669: net/http: excessive CPU consumption from repeated+initial window changes</li>+<li>CVE-2026-78663: net/http: double flow control refund on HTTP/2+server streams</li>+<li>CVE-2026-97032: net/http: HTTP/2 server crash due to HPACK encoder+race</li>+</ul></li>+<li>Update google.golang.org/grpc to fix CVE-2026-84445+(dependabot[bot])+<ul>+<li>CVE-2026-84445: xds: server panic on a request missing both+:authority and Host headers</li>+</ul></li>+</ul></li>+<li>fshttp: Drop <code>--header</code> values over HTTP after HTTPS+downgrade GHSA-rrwf-2qr6-p832 CVE-PENDING (Nick Craig-Wood)</li>+<li>http: Stop a path starting with // moving the remote to another host+GHSA-xqmq-v4wj-rw5m CVE-2026-116192 (Nick Craig-Wood)</li>+<li>lib/http: Require authentication for OPTIONS requests+GHSA-gj73-fh6v-92fj CVE-2026-116193 (Nick Craig-Wood)</li>+<li>local: Fix .rclonelink names escaping the root with+<code>--links</code> GHSA-55qp-jrwr-x827 CVE-2026-116195 (Nick+Craig-Wood)</li>+<li>netstorage+<ul>+<li>Stop a path starting with // moving the remote to another host+GHSA-xqmq-v4wj-rw5m CVE-2026-116192 (Nick Craig-Wood)</li>+<li>Stop a listing resume token moving signed requests to another host+GHSA-xqmq-v4wj-rw5m CVE-2026-116192 (Nick Craig-Wood)</li>+</ul></li>+<li>pcloud+<ul>+<li>Fix OAuth config accepting callbacks without state+GHSA-hvpr-p4pv-4f46 CVE-2026-116194 (Nick Craig-Wood)</li>+<li>Only accept pCloud hosts from the hostname in the OAuth callback+GHSA-hvpr-p4pv-4f46 (Nick Craig-Wood)</li>+</ul></li>+<li>selfupdate+<ul>+<li>Fix hash check trusting unsigned data in the signed SHA256SUMS+GHSA-w33h-3qgq-8mv8 CVE-2026-116189 (Nick Craig-Wood)</li>+<li>Fix release verification instructions trusting unsigned hashes+GHSA-w33h-3qgq-8mv8 (Nick Craig-Wood)</li>+</ul></li>+<li>serve nfs: Fix path traversal by refusing unsafe file names in the+VFS GHSA-44pm-2q5p-fm63 CVE-2026-116191 (Nick Craig-Wood)</li>+<li>serve s3+<ul>+<li>Fix presigned URL being turned into a copy of any object+GHSA-w3h7-jxxq-vfrp CVE-PENDING (Nick Craig-Wood)</li>+<li>Stop auth proxy users listing or using each other's multipart+uploads and metadata GHSA-35g6-9fjv-rmx5 CVE-2026-116187 (Nick+Craig-Wood)</li>+<li>Fix multipart buffer limits being bypassed by overflowing part sizes+GHSA-qjw8-8hmp-c3m6 CVE-2026-116188 (Nick Craig-Wood)</li>+<li>Add <code>--multipart-max-uploads</code> to limit uploads in+progress (Nick Craig-Wood)</li>+<li>Add <code>--multipart-streaming-buffer-total</code> to limit memory+across uploads (Nick Craig-Wood)</li>+<li>Add <code>--metadata-max-objects</code> to limit the metadata kept+in memory (Nick Craig-Wood)</li>+</ul></li>+<li>serve webdav: Fix path traversal above the served directory+GHSA-44pm-2q5p-fm63 CVE-2026-116191 (Nick Craig-Wood)</li>+<li>webdav: Fix headers option leaking to other hosts on redirect+GHSA-3rqf-7h69-gg5v CVE-2026-116190 (Nick Craig-Wood)</li>+</ul></li>+<li>Bug Fixes+<ul>+<li>accounting+<ul>+<li>Stop averaging when the last check finishes (Kunpeng Xie)</li>+<li>Keep speed averaging active when transfers overlap (kaii9)</li>+</ul></li>+<li>dedupe: Fix rename mode giving up after 100 names and make it faster+(Nick Craig-Wood)</li>+<li>doc fixes (Dhevenddra K G, enkvadrat, GhostCoder6969, James Truitt,+maximilize, n4n5, Nick Craig-Wood, PSR94, Rohit Behera, shaurya,+subomi)</li>+<li>docker: Fix files uploaded with the wrong mime type (hsdfat)</li>+<li>fs+<ul>+<li>Fix about showing a negative total when a quota reaches the int64+maximum (youdie006)</li>+<li>Fix bandwidth limits below 1 KiB being multiplied by 1024+(Roland)</li>+<li>Fix <code>--bwlimit</code> timetables written out of order using the+wrong limit (Rohit Behera)</li>+<li>Make BwTimetable.Set replace the timetable instead of appending to+it (youdie006)</li>+<li>Retry "network is unreachable" and "network is down" errors+(nielash)</li>+</ul></li>+<li>fserrors: Fix out of space detection on Windows+(ferrumclaudepilgrim)</li>+<li>lib/rest: Limit the size of HTTP response bodies read into memory+(Nick Craig-Wood)</li>+<li>log: Fix race when adding a log output while logging (Nick+Craig-Wood)</li>+<li>oauthutil: Fix Renew.Shutdown nil timer panic and race+(TastyHeadphones)</li>+<li>operations+<ul>+<li>Stop <code>--copy-dest</code> replacing files with+<code>--immutable</code> (phatlc)</li>+<li>Fix ignored error in rcat probe reads (Wang Chencheng)</li>+<li>Fix hang when deleting files and a fatal error occurs (Nick+Craig-Wood)</li>+<li>Fix hang and lost context when moving a directory file by file (Nick+Craig-Wood)</li>+<li>Don't sleep for a Retry-After error when the transfer is cancelled+(Nick Craig-Wood)</li>+</ul></li>+<li>rc+<ul>+<li>Make job/status and job/list require authentication (Nick+Craig-Wood)</li>+<li>Fix rc API accepting an out of range number and overflowing 64 bits+(KBS)</li>+<li>Reject out of range integer parameters instead of truncating them+(Nick Craig-Wood)</li>+<li>Fix large numeric parameters being rejected on 32 bit builds (Jeremy+Schoemaker)</li>+</ul></li>+<li>selfupdate: Fix <code>--version X.Y</code> depending on ./ in the+download site links (Nick Craig-Wood)</li>+<li>serve+<ul>+<li>Fix <code>--auth-proxy</code> stopping VFS cache uploads after 5+minutes (Nick Craig-Wood)</li>+<li>Fix <code>--auth-proxy</code> backends never being shut down after+simultaneous logins (Nick Craig-Wood)</li>+</ul></li>+<li>serve dlna: Log unescaped paths (phatlc)</li>+<li>serve docker: Fix volume path being lost when the plugin restarts+(phatlc)</li>+<li>serve ftp+<ul>+<li>Fix transfers failing after 5 minutes with <code>--auth-proxy</code>+(Nick Craig-Wood)</li>+<li>Fix VFS leak when the server fails to start (Nick Craig-Wood)</li>+</ul></li>+<li>serve http+<ul>+<li>Fix downloads failing after 5 minutes with <code>--auth-proxy</code>+(Nick Craig-Wood)</li>+<li>Escape filename in zip download Content-Disposition header (Harsh+Raj Singhania)</li>+</ul></li>+<li>serve restic+<ul>+<li>Fix repositories named data (kaii9)</li>+<li>Prevent concurrent append-only overwrites (Vasek Sraier)</li>+</ul></li>+<li>serve s3+<ul>+<li>Apply the multipart upload limits and+<code>--metadata-max-objects</code> to each user separately (Nick+Craig-Wood)</li>+<li>Stream in-order multipart parts without buffering them (Nick+Craig-Wood)</li>+<li>Start the upload to the backend on the first multipart part (Nick+Craig-Wood)</li>+<li>Don't buffer re-uploads of multipart parts already streamed (Nick+Craig-Wood)</li>+<li>Count buffered multipart parts in whole memory pages (Nick+Craig-Wood)</li>+<li>Time out multipart parts waiting for buffer space (Nick+Craig-Wood)</li>+<li>Limit the memory requests can use (Nick Craig-Wood)</li>+<li>Forget expired multipart uploads completely (Nick Craig-Wood)</li>+<li>Forget the metadata of deleted objects (Nick Craig-Wood)</li>+<li>List objects lazily so paging a deep hierarchy is fast (Nick+Craig-Wood)</li>+<li>Fix multipart upload corruption if the backend write of an early+part fails (Nick Craig-Wood)</li>+<li>Fix uploads failing when the auth proxy expires the user's backend+(Nick Craig-Wood)</li>+<li>Fix stale metadata being returned for objects changed elsewhere+(Nick Craig-Wood)</li>+<li>Fix <code>--etag-hash auto</code> crashing or using the wrong hash+with <code>--auth-proxy</code> (Nick Craig-Wood)</li>+<li>Fix CopyObject of a missing object not returning NoSuchKey (Nick+Craig-Wood)</li>+<li>Fix log messages with bad format strings (Nick Craig-Wood)</li>+<li>Document memory use and bring the docs up to date (Nick+Craig-Wood)</li>+</ul></li>+<li>serve webdav+<ul>+<li>Fix transfers failing after 5 minutes with <code>--auth-proxy</code>+(Nick Craig-Wood)</li>+<li>Fix crash when the server fails to start (Nick Craig-Wood)</li>+<li>Escape filename in zip download Content-Disposition header (Harsh+Raj Singhania)</li>+</ul></li>+<li>vfscache: Fix hang when the cache cleaner is disabled+(ferrumclaudepilgrim)</li>+</ul></li>+<li>Mount+<ul>+<li>Fix <code>--daemon</code> failing when <code>--rc</code> is enabled+(somaz)</li>+<li>Ignore com.apple.* xattrs on the macOS FSKit backend (Eugene)</li>+</ul></li>+<li>VFS+<ul>+<li>Fix reads returning zeros after a file changes during handle caching+(Neil Cawse)</li>+<li>Stop setting a file's modtime from discarding its cached data (Neil+Cawse)</li>+<li>Fix crash reading the metadata of a file which is being written+(Nick Craig-Wood)</li>+<li>Fix a VFS being reused while it is being shut down (Nick+Craig-Wood)</li>+<li>Fix AddVirtual ignoring isDir (phatlc)</li>+<li>Stop <code>--links</code> objects called "...rclonelink" appearing+as ".." in directory listings (Nick Craig-Wood)</li>+</ul></li>+<li>Local+<ul>+<li>Fix directories named *.rclonelink being treated as links with+<code>--links</code> (Nick Craig-Wood)</li>+</ul></li>+<li>Archive+<ul>+<li>Fix corrupt listings when listing a zip directory more than once+(Nick Craig-Wood)</li>+<li>Fix zip file entries named for a directory causing confusion (Nick+Craig-Wood)</li>+<li>Fix listing entries with a leading slash as if they were in the root+(Nick Craig-Wood)</li>+<li>Fix crash when creating archive to stdout (Vladimir Babin)</li>+<li>Fix a squashfs path to a single file listing its whole directory+(Nick Craig-Wood)</li>+</ul></li>+<li>Azure Blob+<ul>+<li>Use the released Azure SDK for Apache Arrow listing (Nick+Craig-Wood)</li>+</ul></li>+<li>Box+<ul>+<li>Fix log messages with bad format strings (Nick Craig-Wood)</li>+</ul></li>+<li>Chunker+<ul>+<li>Fix panic listing a directory after an interrupted upload with+meta_format none (Nick Craig-Wood)</li>+</ul></li>+<li>Compress+<ul>+<li>Fix crash on ranged reads when gzip metadata is corrupted+(Acts1631)</li>+</ul></li>+<li>Drime+<ul>+<li>Fix server-side copy over an existing file leaving a "name (1)" copy+(Nick Craig-Wood)</li>+<li>Fix deleted files and directories still being listed with+hard_delete (Nick Craig-Wood)</li>+</ul></li>+<li>Drive+<ul>+<li>Document the Branding step needed to publish your own client_id+(Sanjay Kanth A)</li>+</ul></li>+<li>Dropbox+<ul>+<li>Fix shared folder mount for roots nested more than one level deep+(phatlc)</li>+<li>Match shared-folder and received-file names case-insensitively+(phatlc)</li>+</ul></li>+<li>Filescom+<ul>+<li>Fix corrupted uploads after a retried upload error (Nick+Craig-Wood)</li>+<li>Document that names with / or in can't be stored (Nick+Craig-Wood)</li>+</ul></li>+<li>HTTP+<ul>+<li>Fix crash when the server fails to start (Nick Craig-Wood)</li>+</ul></li>+<li>Iclouddrive+<ul>+<li>Fix potential crash looking up items (Nick Craig-Wood)</li>+</ul></li>+<li>Imagekit+<ul>+<li>Fix potential crash in rmdir and purge (Nick Craig-Wood)</li>+</ul></li>+<li>Internxt+<ul>+<li>Fix lookups of files starting with a dot and dropped uploads+(jzunigax2)</li>+<li>Fix sync with <code>--backup-dir</code> or <code>--suffix</code>+deleting the backed up file (Nick Craig-Wood)</li>+<li>Fix server-side moves failing with "Not Found" or "already exists"+(Nick Craig-Wood)</li>+<li>Fix "directory not empty" and stale directories after moves and+deletes (Nick Craig-Wood)</li>+<li>Fix server-side directory move failing after a gateway timeout (Nick+Craig-Wood)</li>+</ul></li>+<li>Onedrive+<ul>+<li>Update docs for versions, links and time precision on personal+accounts (Nick Craig-Wood)</li>+<li>Document that shared with me shortcuts may fail to list on personal+(Nick Craig-Wood)</li>+<li>Update <code>--onedrive-hard-delete</code> docs to mention personal+account support (foecmke)</li>+</ul></li>+<li>Oracleobjectstorage+<ul>+<li>Fix SSE-C server-side copies (tomaszni)</li>+<li>Upload empty streams without multipart (tomaszni)</li>+</ul></li>+<li>Pixeldrain+<ul>+<li>Fix corrupted uploads after a retried upload error (Nick+Craig-Wood)</li>+</ul></li>+<li>Premiumizeme+<ul>+<li>Fix uploading files with ";" in their names (Nick Craig-Wood)</li>+</ul></li>+<li>S3+<ul>+<li>Disable signing Accept-Encoding for Ceph and Linode (Aditya)</li>+<li>Fix version-at listings with URL encoded keys (jxj)</li>+<li>Fix storage class missing from <code>--s3-versions</code> listings+(Dirk Petersen)</li>+</ul></li>+<li>Seafile+<ul>+<li>Fix corrupted uploads after a retried upload error (Nick+Craig-Wood)</li>+</ul></li>+<li>SFTP+<ul>+<li>Fix failed uploads leaving disk space in use on the server (Nick+Craig-Wood)</li>+</ul></li>+<li>Shade+<ul>+<li>Fix potential crash in directory move (Nick Craig-Wood)</li>+</ul></li>+<li>Smb+<ul>+<li>Save the user name in the config even if it matches the current user+(phatlc)</li>+</ul></li>+<li>WebDAV+<ul>+<li>Fix duplicated listing entries after retried PROPFIND (ZRHann)</li>+<li>Fix "XML syntax error" on paths which don't exist with ownCloud+10.16 (Nick Craig-Wood)</li>+</ul></li>+</ul>+<h2 id="v1.75.1---2026-09-04">v1.75.1 - 2026-09-04</h2>+<p><a+href="https://github.com/rclone/rclone/compare/v1.75.0...v1.75.1">See+commits</a></p>+<ul>+<li>Security+<ul>+<li>archive+<ul>+<li>Fix zip slip path traversal in untrusted zip files+GHSA-66hp-wgxq-6f5q CVE-2026-88014 (Nick Craig-Wood)</li>+<li>Hide any archive entry which escapes the directory being listed+GHSA-66hp-wgxq-6f5q (Nick Craig-Wood)</li>+<li>Reject unsafe entry names when mounting squashfs images+GHSA-66hp-wgxq-6f5q (Nick Craig-Wood)</li>+<li>Fix zip subdirectory root matching sibling directories+GHSA-66hp-wgxq-6f5q (Nick Craig-Wood)</li>+<li>Fix zip entry named "." hiding every other file GHSA-66hp-wgxq-6f5q+(Nick Craig-Wood)</li>+<li>Fix "directory not found" for archive paths containing "./" or "//"+GHSA-66hp-wgxq-6f5q (Nick Craig-Wood)</li>+</ul></li>+<li>build+<ul>+<li>Fix multiple CVEs by upgrading to go1.26.6 (Nick Craig-Wood)+<ul>+<li>CVE-2026-56860: net/url: quadratic complexity in resolvePath</li>+<li>CVE-2026-56858: html/template: JavaScript regexp context+tracking</li>+<li>CVE-2026-56862: crypto/tls: limit handshake messages accepted+post-handshake</li>+<li>CVE-2026-56853: net/http: apply ReadHeaderTimeout to unencrypted+HTTP/2 check</li>+<li>CVE-2026-56859: encoding/xml: recursion depth guard during+decode</li>+<li>CVE-2026-33818: encoding/asn1: enforce maximum recursion depth</li>+<li>CVE-2026-46600: net: panic parsing an invalid SVCB or HTTPS RR in+dnsmessage</li>+<li>CVE-2026-39821: net/http: reject ASCII-only Punycode-encoded labels+in idna</li>+</ul></li>+<li>Update golang.org/x/crypto to v0.56.0 to fix multiple CVEs (Nick+Craig-Wood)+<ul>+<li>CVE-2026-56854: ssh: source-address critical option not enforced for+non-public-key auth callbacks</li>+<li>CVE-2026-78662: ssh: a malicious peer could flood an undecided+channel's incoming requests, deadlocking the connection</li>+<li>CVE-2026-56855: ssh: a malicious peer could send crafted messages on+an established channel, deadlocking the connection</li>+</ul></li>+<li>Update golang.org/x/image to v0.45.0 to fix CVE-2026-46603 (Nick+Craig-Wood)+<ul>+<li>CVE-2026-46603: excessive memory allocation during VP8L+decoding</li>+</ul></li>+</ul></li>+<li>fs: Confine directory listing entries that escape the root+GHSA-3vxh-3pcx-9m8q CVE-2026-93986 GHSA-38xv-hf3p-h7mq CVE-2026-88046+(Nick Craig-Wood)</li>+<li>fshttp: Don't send <code>--header</code> values to other hosts on+redirect GHSA-486v-q2wf-fp2r CVE-2026-88013 (Nick Craig-Wood)</li>+<li>http: Don't leak configured headers to other hosts or over plaintext+on redirect GHSA-486v-q2wf-fp2r CVE-2026-88013 (Nick Craig-Wood)</li>+<li>lib/rest: Check HTTPS downgrades against the original request on+redirect GHSA-486v-q2wf-fp2r CVE-2026-88013 (Nick Craig-Wood)</li>+<li>local+<ul>+<li>Fix dir metadata escaping the root through a planted symlink+GHSA-f8g7-2xjc-7mfh CVE-2026-88016 (Nick Craig-Wood)</li>+<li>Fix btime escaping the root via a planted symlink+GHSA-f8g7-2xjc-7mfh CVE-2026-88016 (Nick Craig-Wood)</li>+<li>Fix panic on Range request past the end of a symlink+GHSA-p6m2-r3w9-mpxw CVE-2026-88015 (Nick Craig-Wood)</li>+</ul></li>+<li>serve docker+<ul>+<li>Reject volume names that escape the base directory+GHSA-p6vx-hf7p-98j6 CVE-2026-93987 (Nick Craig-Wood)</li>+<li>Reject volume names resolving to the base directory itself+GHSA-p6vx-hf7p-98j6 CVE-2026-93987 (Nick Craig-Wood)</li>+<li>Re-derive volume mountpoint from name when restoring state+GHSA-p6vx-hf7p-98j6 CVE-2026-93987 (Nick Craig-Wood)</li>+</ul></li>+<li>serve ftp: Fix auth-proxy sessions sharing credentials by username+GHSA-c476-6w5q-jw77 CVE-2026-88017 (Nick Craig-Wood)</li>+<li>serve s3+<ul>+<li>Fix memory exhaustion from client-declared multipart part size+GHSA-2p48-j3qc-rx9f CVE-2026-88045 (Nick Craig-Wood)</li>+<li>Reject bogus multipart part sizes in the reorder buffer+GHSA-2p48-j3qc-rx9f (Nick Craig-Wood)</li>+<li>Fix auth proxy accepting any request signed with an empty secret+GHSA-xwwr-4h3p-r22c CVE-2026-88018 (Nick Craig-Wood)+<ul>+<li><strong>NB</strong> the auth proxy protocol for+<code>serve s3</code> has changed - the proxy program is now given the+access key ID as <code>user</code> and must return the secret as+<code>_secret_access_key</code></li>+</ul></li>+<li>Fix each server accepting the <code>--auth-key</code> credentials of+all the others (Nick Craig-Wood)</li>+<li>Fix misleading anonymous access log when using an auth proxy via rc+GHSA-p569-5gjg-9cmj CVE-2026-88044 (Nick Craig-Wood)</li>+</ul></li>+<li>serve sftp: Fix auth proxy configured via rc being silently ignored+GHSA-p569-5gjg-9cmj CVE-2026-88044 (Nick Craig-Wood)</li>+</ul></li>+<li>Bug Fixes+<ul>+<li>accounting+<ul>+<li>Fix memory leak on long-running rcd (nielash)</li>+<li>Fix memory leak from stats groups on long-running rcd (nielash)</li>+<li>Fix bwlimit burst overflow (Rayan Salhab)</li>+</ul></li>+<li>bisync+<ul>+<li>Fix memory leak when running via the rc (nielash)</li>+<li>Fix failed transfers of empty files being recorded as synced (Nick+Craig-Wood)</li>+</ul></li>+<li>build: Make go1.26 the minimum required version as needed by+golang.org/x/crypto v0.56.0 (Nick Craig-Wood)</li>+<li>config: Redact env var config values in logs (Pastalikek65)</li>+<li>doc fixes (Anton Karpov, CAOShurong, Dean Chen, Nick Craig-Wood,+Recoordinate, Rodrigo Rodrigues, Shantanav Mukherjee, shaurya)</li>+<li>lib/batcher: Prevent commits racing shutdown (Loi Nguyen)</li>+<li>lib/transform: Fix panic in <code>truncate_keep_extension</code>+(VXNCXNX)</li>+<li>multipart: Fix chunked uploads storing truncated objects when the+source ends early (Nick Craig-Wood)</li>+<li>operations: Fix silent truncation of streaming uploads whose source+ends early (Nick Craig-Wood)</li>+<li>serve+<ul>+<li>Fix VFS instance leaks on server startup failures and shutdown+(Hakan İSMAİL)</li>+<li>Pass the client IP address to the auth proxy (am-at-enrollvb)</li>+</ul></li>+<li>serve http: Prevent scrolling to the top on page reload (Sune+Mølgaard)</li>+<li>serve nfs: Fix EIO when creating symlinks with+<code>--vfs-links</code> (SillyZir)</li>+<li>serve s3+<ul>+<li>Fix failed uploads deleting or corrupting the object at the key+(Nick Craig-Wood)</li>+<li>Fix crash when a multipart upload is aborted while a part is+uploading (Nick Craig-Wood)</li>+<li>Fix modtime not being set when only mtime metadata is supplied on+PUT (Nick Craig-Wood)</li>+<li>Upload all multipart uploads via the VFS so they obey+<code>--bwlimit</code> and show in stats (Nick Craig-Wood)</li>+<li>Reserve the <code>.rclone_temp_</code> prefix for temporary objects+(Nick Craig-Wood)</li>+<li>Clean up abandoned multipart uploads after+<code>--multipart-expiry</code> (Nick Craig-Wood)</li>+</ul></li>+<li>vfscache+<ul>+<li>Fix reader deadlock when the item size drops below the read offset+(Dave)</li>+<li>Fix log message growing without bound on repeated write errors+(Vijay Misal)</li>+</ul></li>+<li>walk: Stop directory traversal when the context is cancelled (Rahman+Yilmaz)</li>+</ul></li>+<li>VFS+<ul>+<li>Synchronize poll updates with shutdown (Loi Nguyen)</li>+<li>Make poll shutdown lifecycle deterministic (Loi Nguyen)</li>+</ul></li>+<li>Crypt+<ul>+<li>Fix hash mismatches with <code>no_data_encryption</code> on backends+which check upload hashes (Nick Craig-Wood)</li>+<li>Fix directory names which look like versioned file names+(TowyTowy)</li>+<li>Warn about directories with legacy version-like encrypted names+(Nick Craig-Wood)</li>+</ul></li>+<li>Azure Blob+<ul>+<li>Fix Entra ID server-side copy source authentication (Edward+Klesel)</li>+<li>Fix spurious vfs cache corruption errors during chunked reads (Nick+Craig-Wood)</li>+</ul></li>+<li>Azurefiles+<ul>+<li>Fix zero padded files being created when the source ends early (Nick+Craig-Wood)</li>+</ul></li>+<li>Box+<ul>+<li>Fix truncated files being uploaded successfully when the source ends+early (Rohit Behera)</li>+</ul></li>+<li>Compress+<ul>+<li>Fix corrupted objects being created when the source ends early (Nick+Craig-Wood)</li>+</ul></li>+<li>Drive+<ul>+<li>Don't list trashed files when removing a directory into the trash+(alliasgher)</li>+</ul></li>+<li>Dropbox+<ul>+<li>Preserve Paper export paths on lookup (Loi Nguyen)</li>+<li>Fix context cancellation (e.g. <code>--max-duration</code> limit)+not stopping in-flight requests (debaditya)</li>+<li>Fix chunked uploads of truncated files never finishing (Nick+Craig-Wood)</li>+<li>Don't retry chunked upload requests when the upload has been+cancelled (Nick Craig-Wood)</li>+<li>Decode received shared-file names (Sanjay Kanth A)</li>+<li>Fix ChangeNotify when the root's case differs from Dropbox's (Loi+Nguyen)</li>+</ul></li>+<li>Filelu+<ul>+<li>Fix truncated files being uploaded successfully when the source ends+early (Nick Craig-Wood)</li>+<li>Fix duplicate root path during multipart folder creation+(kingston125)</li>+</ul></li>+<li>Huaweidrive+<ul>+<li>Fix truncated files being uploaded successfully when the source ends+early (Rohit Behera)</li>+</ul></li>+<li>Iclouddrive+<ul>+<li>Fix uploads into an app container failing with 412 (Christian De+Santis)</li>+</ul></li>+<li>Internetarchive+<ul>+<li>Fix corrupted files being created when the source ends early (Nick+Craig-Wood)</li>+</ul></li>+<li>Internxt+<ul>+<li>Persist rotated token returned by the user info call+(0rangeSeaW0lf)</li>+</ul></li>+<li>Onedrive+<ul>+<li>Fix 403 Forbidden for configuration personal onedrive (machsix)</li>+<li>Fall back to manual drive ID entry when drive listing fails+(SillyZir)</li>+<li>Don't retry multipart upload chunk on 404 (upload session not found)+(water)</li>+</ul></li>+<li>Overview+<ul>+<li>Fix "internal error: no overview data found" on 32 bit architectures+(Nick Craig-Wood)</li>+</ul></li>+<li>Pikpak+<ul>+<li>Fix truncated files being created when the source ends early (Nick+Craig-Wood)</li>+<li>Fix truncated single part uploads reported as ok when source ends+early (Nick Craig-Wood)</li>+</ul></li>+<li>Protondrive+<ul>+<li>Fix files uploaded with v1.75.0 not being readable in the Proton+apps (Nick Craig-Wood)</li>+<li>Fix corrupted uploads after a retried upload error (Nick+Craig-Wood)</li>+</ul></li>+<li>Quatrix+<ul>+<li>Fix chunk upload retries and fix memory leak (Nick Craig-Wood)</li>+</ul></li>+<li>S3+<ul>+<li>Update Mega endpoints (Nick Craig-Wood)</li>+<li>Treat UploadPart success without ETag as retryable error+(CAOShurong)</li>+<li>Fix server side copy failing with <code>--s3-no-head-object</code>+(Anatoly Tarnavsky)</li>+</ul></li>+<li>Sia+<ul>+<li>Fix corrupted files being created when the source ends early (Nick+Craig-Wood)</li>+</ul></li>+<li>Smb+<ul>+<li>Reuse the upload connection for SetModTime (alliasgher)</li>+</ul></li>+<li>WebDAV+<ul>+<li>Fix SetModTime failing and hashes missing on Nextcloud (Nick+Craig-Wood)</li>+</ul></li>+<li>Yandex+<ul>+<li>Fix truncated files being uploaded successfully when the source ends+early (Rohit Behera)</li>+</ul></li>+</ul><h2 id="v1.75.0---2026-07-31">v1.75.0 - 2026-07-31</h2><p><ahref="https://github.com/rclone/rclone/compare/v1.74.0...v1.75.0">See@@ -66244,48 +70957,48 @@ ARTESCA)</li><li>Security<ul><li>archive: Don't crash on malformed squashfs images-GHSA-6jcg-q3wp-x2f4 CVE-PENDING (Nick Craig-Wood)</li>+GHSA-6jcg-q3wp-x2f4 CVE-2026-79775 (Nick Craig-Wood)</li><li>ftp: Fix ftp command injection when encoding doesn't include CRLF-GHSA-8c48-q9wj-3w37 CVE-PENDING (Nick Craig-Wood)</li>+GHSA-8c48-q9wj-3w37 CVE-2026-71311 (Nick Craig-Wood)</li><li>lib/http: Use TLS on all <code>--addr</code> listeners when<code>--cert</code> and <code>--key</code> are set GHSA-mfvx-7rcj-9m5g(Nick Craig-Wood)</li><li>lib/proxy: Fix unbounded HTTP CONNECT headers causing OOM-GHSA-xhf4-832v-7xcr CVE-PENDING (Nick Craig-Wood)</li>+GHSA-xhf4-832v-7xcr CVE-2026-71310 (Nick Craig-Wood)</li><li>local: Stop source file names escaping the destination directory-GHSA-7p4m-qxvv-g567 CVE-PENDING (Nick Craig-Wood)</li>+GHSA-7p4m-qxvv-g567 CVE-2026-71313 (Nick Craig-Wood)</li><li>rc<ul><li>Don't expose pprof debug handlers on an unauthenticated server-GHSA-mfvx-7rcj-9m5g CVE-PENDING (Nick Craig-Wood)</li>+GHSA-mfvx-7rcj-9m5g CVE-2026-79776 (Nick Craig-Wood)</li><li>Require authentication to list the remotes with<code>--rc-serve</code> GHSA-mfvx-7rcj-9m5g (Nick Craig-Wood)</li>-<li>Fix leaking stack traces on panics GHSA-gwfq-86j8-7qhv (Nick-Craig-Wood)</li>+<li>Fix leaking stack traces on panics GHSA-gwfq-86j8-7qhv+CVE-2026-79777 (Nick Craig-Wood)</li></ul></li><li>s3<ul><li>Fix redirect credential leaks, reject HTTPS->HTTP and strip-secrets GHSA-8mxv-9xhp-86h4 (Nick Craig-Wood)</li>+secrets GHSA-8mxv-9xhp-86h4 CVE-2026-79780 (Nick Craig-Wood)</li><li>Strip S3 Express session token on cross-host redirects-GHSA-8mxv-9xhp-86h4 (Nick Craig-Wood)</li>+GHSA-8mxv-9xhp-86h4 CVE-2026-79780 (Nick Craig-Wood)</li></ul></li><li>serve ftp: Use constant time comparison for password checkGHSA-mfvx-7rcj-9m5g (Nick Craig-Wood)</li><li>serve restic: Fix path traversal above the served directory-GHSA-45pq-889g-fcgh CVE-PENDING (Nick Craig-Wood)</li>+GHSA-45pq-889g-fcgh CVE-2026-71309 (Nick Craig-Wood)</li><li>serve sftp: Don't crash the whole server on a bad request-GHSA-6jcg-q3wp-x2f4 (Nick Craig-Wood)</li>+GHSA-6jcg-q3wp-x2f4 CVE-2026-79775 (Nick Craig-Wood)</li><li>sftp: Fix command injection via crafted filenames on PowerShell-remotes GHSA-2m8m-jhrm-w6j2 CVE-PENDING (Nick Craig-Wood)</li>+remotes GHSA-2m8m-jhrm-w6j2 CVE-2026-71312 (Nick Craig-Wood)</li><li>vfs: Don't crash the process if a backend panics on a background-goroutine GHSA-6jcg-q3wp-x2f4 (Nick Craig-Wood)</li>+goroutine GHSA-6jcg-q3wp-x2f4 CVE-2026-79775 (Nick Craig-Wood)</li><li>webdav<ul><li>Fix HTTPS to HTTP redirects leaking credentials GHSA-h4mf-4v27-hggj-(Nick Craig-Wood)</li>-<li>Tus: fix potential nil pointer crash GHSA-3x6r-wxxg-53vv (Nick-Craig-Wood)</li>+CVE-2026-79779 (Nick Craig-Wood)</li>+<li>Tus: fix potential nil pointer crash GHSA-3x6r-wxxg-53vv+CVE-2026-79778 (Nick Craig-Wood)</li></ul></li><li>Update <code>google.golang.org/grpc</code> to fix multiple securityproblems (Nick Craig-Wood)</li>@@ -66766,7 +71479,7 @@ CVE-2026-59733 (Nick Craig-Wood)</li><ul><li>Fix spurious 404 on HEAD/GET during VFS writeback (max)</li><li>Fix path traversal letting clients see files in the root-GHSA-8v25-v8p6-qf7v (Nick Craig-Wood)</li>+GHSA-8v25-v8p6-qf7v CVE-2026-79781 (Nick Craig-Wood)</li></ul></li><li>serve webdav: Fix MOVE overwrite failing without Overwrite header(Sanjay Santhanam)</li>@@ -66783,7 +71496,7 @@ Craig-Wood)</li><li>Stop <code>--links</code> symlinks escaping the destinationdirectory CVE-2026-54572 (Nick Craig-Wood)</li><li>Don't restore setuid/setgid/sticky bits from metadata by default-GHSA-945v-v9p3-v5xw (Nick Craig-Wood)</li>+GHSA-945v-v9p3-v5xw CVE-2026-79783 (Nick Craig-Wood)</li></ul></li><li>Drive<ul>@@ -79275,7 +83988,7 @@ installations</li><li>Project started</li></ul><h1 id="bugs-and-limitations">Bugs and Limitations</h1>-<h2 id="limitations-49">Limitations</h2>+<h2 id="limitations-50">Limitations</h2><h3 id="directory-timestamps-arent-preserved-on-some-backends">Directorytimestamps aren't preserved on some backends</h3><p>As of <code>v1.66</code>, rclone supports syncing directory modtimes,@@ -79430,7 +84143,7 @@ want it set for a single backend, eg<p>The FTP and SFTP backends have their own <code>http_proxy</code>settings to support an HTTP CONNECT proxy ( <ahref="https://rclone.org/ftp/#ftp-http-proxy">--ftp-http-proxy</a> and-<a href="https://rclone.org/ftp/#sftp-http-proxy">--sftp-http-proxy</a>+<a href="https://rclone.org/sftp/#sftp-http-proxy">--sftp-http-proxy</a>)</p><h3 id="rclone-gives-x509-ssl-root-certificates-error">Rclone gives x509SSL root certificates error</h3>@@ -79485,10 +84198,10 @@ formats</p>some.domain.com no such host</h3><p>This happens when rclone cannot resolve a domain. Please check thatyour DNS setup is generally working, e.g.</p>-<div class="sourceCode" id="cb1842"><pre-class="sourceCode sh"><code class="sourceCode bash"><span id="cb1842-1"><a href="#cb1842-1" aria-hidden="true" tabindex="-1"></a><span class="co"># both should print a long list of possible IP addresses</span></span>-<span id="cb1842-2"><a href="#cb1842-2" aria-hidden="true" tabindex="-1"></a><span class="ex">dig</span> www.googleapis.com <span class="co"># resolve using your default DNS</span></span>-<span id="cb1842-3"><a href="#cb1842-3" aria-hidden="true" tabindex="-1"></a><span class="ex">dig</span> www.googleapis.com @8.8.8.8 <span class="co"># resolve with Google's DNS server</span></span></code></pre></div>+<div class="sourceCode" id="cb1911"><pre+class="sourceCode sh"><code class="sourceCode bash"><span id="cb1911-1"><a href="#cb1911-1" aria-hidden="true" tabindex="-1"></a><span class="co"># both should print a long list of possible IP addresses</span></span>+<span id="cb1911-2"><a href="#cb1911-2" aria-hidden="true" tabindex="-1"></a><span class="ex">dig</span> www.googleapis.com <span class="co"># resolve using your default DNS</span></span>+<span id="cb1911-3"><a href="#cb1911-3" aria-hidden="true" tabindex="-1"></a><span class="ex">dig</span> www.googleapis.com @8.8.8.8 <span class="co"># resolve with Google's DNS server</span></span></code></pre></div><p>If you are using <code>systemd-resolved</code> (default on ArchLinux), ensure it is at version 233 or higher. Previous releases containa bug which causes not all domains to be resolved properly.</p>@@ -79511,8 +84224,8 @@ yyyy/mm/dd hh:mm:ss Fatal error: config failed to refresh token: failed to startwith opening the port on the host.</p><p>A simple solution may be restarting the Host Network Service with eg.Powershell</p>-<div class="sourceCode" id="cb1844"><pre-class="sourceCode powershell"><code class="sourceCode powershell"><span id="cb1844-1"><a href="#cb1844-1" aria-hidden="true" tabindex="-1"></a><span class="fu">Restart-Service</span> hns</span></code></pre></div>+<div class="sourceCode" id="cb1913"><pre+class="sourceCode powershell"><code class="sourceCode powershell"><span id="cb1913-1"><a href="#cb1913-1" aria-hidden="true" tabindex="-1"></a><span class="fu">Restart-Service</span> hns</span></code></pre></div><h3id="the-total-size-reported-in-the-stats-for-a-sync-is-wrong-and-keeps-changing">Thetotal size reported in the stats for a sync is wrong and keeps@@ -79582,20 +84295,20 @@ TLS (TLS 1.2/1.3) and ECDHE-based cipher suites you can re-enable legacyciphers via <code>GODEBUG</code>:</p><ul><li><p>Windows (cmd.exe):</p>-<div class="sourceCode" id="cb1845"><pre-class="sourceCode bat"><code class="sourceCode dosbat"><span id="cb1845-1"><a href="#cb1845-1" aria-hidden="true" tabindex="-1"></a><span class="bu">set</span> <span class="va">GODEBUG</span>=tlsrsakex=1</span>-<span id="cb1845-2"><a href="#cb1845-2" aria-hidden="true" tabindex="-1"></a>rclone copy ...</span></code></pre></div></li>+<div class="sourceCode" id="cb1914"><pre+class="sourceCode bat"><code class="sourceCode dosbat"><span id="cb1914-1"><a href="#cb1914-1" aria-hidden="true" tabindex="-1"></a><span class="bu">set</span> <span class="va">GODEBUG</span>=tlsrsakex=1</span>+<span id="cb1914-2"><a href="#cb1914-2" aria-hidden="true" tabindex="-1"></a>rclone copy ...</span></code></pre></div></li><li><p>Windows (PowerShell):</p>-<div class="sourceCode" id="cb1846"><pre-class="sourceCode powershell"><code class="sourceCode powershell"><span id="cb1846-1"><a href="#cb1846-1" aria-hidden="true" tabindex="-1"></a><span class="va">$env</span><span class="op">:</span>GODEBUG<span class="op">=</span><span class="st">"tlsrsakex=1"</span></span>-<span id="cb1846-2"><a href="#cb1846-2" aria-hidden="true" tabindex="-1"></a>rclone <span class="fu">copy</span> <span class="op">...</span></span></code></pre></div></li>+<div class="sourceCode" id="cb1915"><pre+class="sourceCode powershell"><code class="sourceCode powershell"><span id="cb1915-1"><a href="#cb1915-1" aria-hidden="true" tabindex="-1"></a><span class="va">$env</span><span class="op">:</span>GODEBUG<span class="op">=</span><span class="st">"tlsrsakex=1"</span></span>+<span id="cb1915-2"><a href="#cb1915-2" aria-hidden="true" tabindex="-1"></a>rclone <span class="fu">copy</span> <span class="op">...</span></span></code></pre></div></li><li><p>Linux/macOS:</p>-<div class="sourceCode" id="cb1847"><pre-class="sourceCode sh"><code class="sourceCode bash"><span id="cb1847-1"><a href="#cb1847-1" aria-hidden="true" tabindex="-1"></a><span class="va">GODEBUG</span><span class="op">=</span>tlsrsakex=1 <span class="ex">rclone</span> copy ...</span></code></pre></div></li>+<div class="sourceCode" id="cb1916"><pre+class="sourceCode sh"><code class="sourceCode bash"><span id="cb1916-1"><a href="#cb1916-1" aria-hidden="true" tabindex="-1"></a><span class="va">GODEBUG</span><span class="op">=</span>tlsrsakex=1 <span class="ex">rclone</span> copy ...</span></code></pre></div></li></ul><p>If the server only supports 3DES, try:</p>-<div class="sourceCode" id="cb1848"><pre-class="sourceCode sh"><code class="sourceCode bash"><span id="cb1848-1"><a href="#cb1848-1" aria-hidden="true" tabindex="-1"></a><span class="va">GODEBUG</span><span class="op">=</span>tls3des=1 <span class="ex">rclone</span> ...</span></code></pre></div>+<div class="sourceCode" id="cb1917"><pre+class="sourceCode sh"><code class="sourceCode bash"><span id="cb1917-1"><a href="#cb1917-1" aria-hidden="true" tabindex="-1"></a><span class="va">GODEBUG</span><span class="op">=</span>tls3des=1 <span class="ex">rclone</span> ...</span></code></pre></div><p>This applies to <strong>any rclone feature using TLS</strong> (HTTPS,FTPS, WebDAV over TLS, proxies with TLS interception, etc.). Use theseworkarounds only long enough to get the server/proxy updated.</p>@@ -82158,6 +86871,309 @@ class="email">[email protected]</a></li>class="email">[email protected]</a></li><li>Punya Jain <a href="mailto:[email protected]"class="email">[email protected]</a></li>+<li>acoeur <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>SillyZir <a+href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>am-at-enrollvb <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Loi Nguyen <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>debaditya <a+href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>alliasgher <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Anton Karpov <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Edward Klesel <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Christian De Santis <a+href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Rodrigo Rodrigues <a+href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Recoordinate <a+href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Jiacheng Xu <a+href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Teddy Tennant <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Dean Chen <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Pastalikek65 <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Morax <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Shurong Cao <a+href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Dave <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>max <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Hakan İSMAİL <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Shantanav Mukherjee <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>VXNCXNX <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Rohit Behera <a+href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Rahman Yilmaz <a+href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Sanjay Kanth A <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>machsix <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>CAOShurong <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>shaurya <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>water <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Rayan Salhab <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Vijay Misal <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Sune Mølgaard <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Anatoly Tarnavsky <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>cyphercodes <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>shaurya <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>no-hup <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Dhevenddra <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Splainte <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>PSR94 <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Shane McCarron <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Dhevenddra K G <a+href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Vaibhav Mashal <a+href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Murat Topcu <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>tomaszni <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>eliotee <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Aditya <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Nicholas Velten <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>youdie006 <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>subomi <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Can Arslan <a+href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Can Arslan <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>ZRHann <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>enkvadrat <a+href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Vladimir Babin <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Eugene <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Kunpeng Xie <a+href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>foecmke <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Vasek Sraier <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Kalin Stoyanov <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>NytePlus <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Jeremy Schoemaker <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>solunolab <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Mattias Michaux <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>jxj <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Harsh Raj Singhania <a+href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Roland <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>interested.tortoise <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>hsdfat <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Dirk Petersen <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>TastyHeadphones <a+href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Neil Cawse <a+href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Lev Devaev <a+href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>GhostCoder6969 <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>bounty-agent <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>mygrexit <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Kaya <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>somaz <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>xiaozou-wine <a+href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Roman Butor <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>kaii9 <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Stochastic <a+href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>James Truitt <a+href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>firatkaya <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>kevin9327 <a+href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>drakeo338 <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Alessio Margelli <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Ludovic Bondon <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>fmterrors <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>GingerAdonis <a+href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Satvik Mishra <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>wjc <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Maksym Tymoshyk <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Aniket <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Oshino Penguin <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Yousef Issa <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>ghwany <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>DEVN <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>47th <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Sash <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Shreyansh Jain <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Biagio Peccerillo <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>gomitrah <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Jim Counter <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Dirk Einecke <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>ASRodrigo1 <a+href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>chaogeshuodiannao <a+href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>David Legrand <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Azhar <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Daniel Bae <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>frodleTian <a+href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Cestercian <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Ayush Sharma <a+href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Dirk Petersen <a+href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Codex <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>bunnysayzz <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>hahahahahayesyeseys <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>qodelicious <a+href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>fixfon <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>berce <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Fred <a+href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Florian Josef Altendorfer <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>ashwinibailangadi <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>ilya lavrenov <a+href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Soujas K <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Robert Kubiś <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>b84330808 <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Guillaume MALLERET <a+href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Louis <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>唐僧洗发用飘柔 <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Seth Westphal <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>ZainnQureshii <a+href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Rishi Raj Jain <a href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Enbiya Göral <a+href="mailto:[email protected]"+class="email">[email protected]</a></li>+<li>Florian Josef Altendorfer <a+href="mailto:[email protected]"+class="email">[email protected]</a></li></ul><h1 id="contact-the-rclone-project">Contact the rclone project</h1><h2 id="forum">Forum</h2>diff --git a/MANUAL.md b/MANUAL.mdindex 6d194963b..7b75fda7c 100644--- a/MANUAL.md+++ b/MANUAL.md@@ -1,6 +1,6 @@% rclone(1) User Manual% Nick Craig-Wood-% Jul 31, 2026+% Nov 06, 2026# NAME@@ -39,6 +39,7 @@ Available commands:gui Open the web based GUI.hashsum Produces a hashsum file for all the objects in the path.help Show help for rclone commands, flags and backends.+ index Write static directory listings into a remote.link Generate public link to file/folder.listremotes List all the remotes in the config file and defined in environment variables.ls List the objects in the path with size and path.@@ -187,6 +188,7 @@ WebDAV or S3, that work out of the box.)- Akamai Netstorage- Alibaba Cloud (Aliyun) Object Storage System (OSS)- Amazon S3+- Autonomys Auto Drive- Bizfly Cloud Simple Storage- Backblaze B2- Box@@ -194,11 +196,13 @@ WebDAV or S3, that work out of the box.)- China Mobile Ecloud Elastic Object Storage (EOS)- Arvan Cloud Object Storage (AOS)- Citrix ShareFile+- Clever Cloud Cellar- Cloudflare R2- Cloudinary- Cubbit DS3- DigitalOcean Spaces- Digi Storage+- dosya.dev- Dreamhost- Drime- Dropbox@@ -232,8 +236,11 @@ WebDAV or S3, that work out of the box.)- IBM COS S3- IDrive e2- Impossible Cloud+- Infomaniak Swiss Backup (S3)+- Infomaniak Swiss Backup (Swift)- Intercolo Object Storage- IONOS Cloud+- kDrive- Koofr- Leviia Object Storage- Liara Object Storage@@ -249,6 +256,7 @@ WebDAV or S3, that work out of the box.)- Microsoft Azure Files Storage- Microsoft OneDrive- Minio+- Neon Object Storage- Nextcloud- Blomp Cloud Storage- OpenDrive@@ -273,6 +281,7 @@ WebDAV or S3, that work out of the box.)- RackCorp Object Storage- Rackspace Cloud Files- Rclone Serve S3+- RelAix.Storage- rsync.net- Scaleway- Scality (RING / ARTESCA)@@ -588,12 +597,6 @@ package is here.The rclone developers maintain a [docker image for rclone](https://hub.docker.com/r/rclone/rclone).-**Note:** We also now offer a paid version of rclone with-enterprise-grade security and zero CVEs through our partner-[SecureBuild](https://securebuild.com/blog/introducing-securebuild).-If you are interested, check out their website and the [Rclone-SecureBuild Image](https://securebuild.com/images/rclone).-These images are built as part of the release process based on aminimal Alpine Linux.@@ -707,7 +710,7 @@ not the rclone developers so it may be out of date. Its current version is as be## Source installation {#source}Make sure you have git and [Go](https://golang.org/) installed.-Go version 1.25 or newer is required, the latest release is recommended.+Go version 1.26 or newer is required, the latest release is recommended.You can get it from your package manager, or download it from[golang.org/dl](https://golang.org/dl/). Then you can run the following:@@ -1051,6 +1054,7 @@ See the following for detailed instructions for- [Crypt](https://rclone.org/crypt/) - to encrypt other remotes- [DigitalOcean Spaces](https://rclone.org/s3/#digitalocean-spaces)- [Digi Storage](https://rclone.org/koofr/#digi-storage)+- [dosya.dev](https://rclone.org/dosya/)- [Drime](https://rclone.org/drime/)- [Dropbox](https://rclone.org/dropbox/)- [Enterprise File Fabric](https://rclone.org/filefabric/)@@ -1072,6 +1076,7 @@ See the following for detailed instructions for- [Internet Archive](https://rclone.org/internetarchive/)- [Internxt](https://rclone.org/internxt/)- [Jottacloud](https://rclone.org/jottacloud/)+- [kDrive](https://rclone.org/kdrive/)- [Koofr](https://rclone.org/koofr/)- [Linkbox](https://rclone.org/linkbox/)- [Mail.ru Cloud](https://rclone.org/mailru/)@@ -2334,6 +2339,10 @@ Note that `ls` and `lsl` recurse by default - use `--max-depth 1` to stop the reThe other list commands `lsd`,`lsf`,`lsjson` do not recurse by default -use `-R` to make them recurse.+On the bucket-based remotes (e.g. s3, gcs, b2) a directory is a key prefix+rather than an object, so a listing without `-R` shows each prefix as one+directory entry and not the objects stored under it.+List commands prefer a recursive method that uses more memory but fewertransactions by default. Use `--disable ListR` to suppress the behavior.See [`--fast-list`](https://rclone.org/docs/#fast-list) for more details.@@ -2415,7 +2424,7 @@ Lists the directories in the source path to standard output. Does notrecurse by default. Use the `-R` flag to recurse.This command lists the total size of the directory (if known, -1 if-not), the modification time (if known, the current time if not), the+not), the modification time (if known, the `--default-time` if not), thenumber of objects in the directory (if known, -1 if not) and the nameof the directory, E.g.@@ -2455,6 +2464,10 @@ Note that `ls` and `lsl` recurse by default - use `--max-depth 1` to stop the reThe other list commands `lsd`,`lsf`,`lsjson` do not recurse by default -use `-R` to make them recurse.+On the bucket-based remotes (e.g. s3, gcs, b2) a directory is a key prefix+rather than an object, so a listing without `-R` shows each prefix as one+directory entry and not the objects stored under it.+List commands prefer a recursive method that uses more memory but fewertransactions by default. Use `--disable ListR` to suppress the behavior.See [`--fast-list`](https://rclone.org/docs/#fast-list) for more details.@@ -2565,6 +2578,10 @@ Note that `ls` and `lsl` recurse by default - use `--max-depth 1` to stop the reThe other list commands `lsd`,`lsf`,`lsjson` do not recurse by default -use `-R` to make them recurse.+On the bucket-based remotes (e.g. s3, gcs, b2) a directory is a key prefix+rather than an object, so a listing without `-R` shows each prefix as one+directory entry and not the objects stored under it.+List commands prefer a recursive method that uses more memory but fewertransactions by default. Use `--disable ListR` to suppress the behavior.See [`--fast-list`](https://rclone.org/docs/#fast-list) for more details.@@ -3734,6 +3751,7 @@ rclone bisync remote1:path1 remote2:path2 [flags]--force Bypass --max-delete safety check and run the sync. Consider using with --verbose-h, --help help for bisync--ignore-listing-checksum Do not use checksums for listings (add --ignore-checksum to additionally skip post-copy checksum checks)+ --max-delete-renames-aware Exclude tracked renames from the --max-delete safety check (requires --track-renames)--max-lock Duration Consider lock files older than this to be expired (default: 0 (never expire)) (minimum: 2m) (default 0s)--no-cleanup Retain working files (useful for troubleshooting and testing).--no-slow-hash Ignore listing checksums only on backends where they are slow@@ -4099,6 +4117,8 @@ Output completion script for a given shell.Generates a shell completion script for rclone.Run with `--help` to list the supported shells.+Enum-valued flags such as `--log-level` and `--vfs-cache-mode`+also complete their supported values.## Options@@ -5479,12 +5499,12 @@ rclone convmv "stories/The Quick Brown Fox!.txt" --name-transform "all,command=e```consolerclone convmv "stories/The Quick Brown Fox!" --name-transform "date=-{YYYYMMDD}"-// Output: stories/The Quick Brown Fox!-20260731+// Output: stories/The Quick Brown Fox!-20261106``````consolerclone convmv "stories/The Quick Brown Fox!" --name-transform "date=-{macfriendlytime}"-// Output: stories/The Quick Brown Fox!-2026-07-31 0340PM+// Output: stories/The Quick Brown Fox!-2026-11-06 0220PM``````console@@ -6178,9 +6198,11 @@ Remove a single file from remote.## Synopsis-Remove a single file from remote. Unlike `delete` it cannot be used to+Remove a single file from remote. Unlike `delete` it cannot be used toremove a directory and it doesn't obey include/exclude filters - if the-specified file exists, it will always be removed.+specified file exists, it will always be removed from its original location.++The `--backup-dir` and `--suffix` flags are honored when set.```rclone deletefile remote:path [flags]@@ -6195,6 +6217,26 @@ rclone deletefile remote:path [flags]Options shared with other commands are described next.See the [global flags page](https://rclone.org/flags/) for global options not listed here.+### Sync Options++Flags used for sync commands++```text+ --backup-dir string Make backups into hierarchy based in DIR+ --delete-after When synchronizing, delete files on destination after transferring (default)+ --delete-before When synchronizing, delete files on destination before transferring+ --delete-during When synchronizing, delete files during transfer+ --fix-case Force rename of case insensitive dest to match source+ --ignore-errors Delete even if there are I/O errors+ --list-cutoff int To save memory, sort directory listings on disk above this threshold (default 1000000)+ --max-delete int When synchronizing, limit the number of deletes (default -1)+ --max-delete-size SizeSuffix When synchronizing, limit the total size of deletes (default off)+ --suffix string Suffix to add to changed files+ --suffix-keep-extension Preserve the extension when using --suffix+ --track-renames When synchronizing, track file renames and do a server-side move if possible+ --track-renames-strategy string Strategies to use when synchronizing using track-renames hash|modtime|leaf (default "hash")+```+### Important OptionsImportant flags useful for most commands@@ -6448,11 +6490,6 @@ Flags to control the Remote Control API--rc-template string User-specified template--rc-user string User name for authentication--rc-user-from-header string User name from a defined HTTP header- --rc-web-fetch-url string URL to fetch the releases for webgui (default "https://api.github.com/repos/rclone/rclone-webui-react/releases/latest")- --rc-web-gui Launch WebGUI on localhost- --rc-web-gui-force-update Force update to latest version of web gui- --rc-web-gui-no-open-browser Don't open the browser automatically- --rc-web-gui-update Check and update to latest version of web gui```## See Also@@ -6502,6 +6539,7 @@ Supported hashes are:- blake3- xxh3- xxh128+- crc32c```Then@@ -6577,6 +6615,332 @@ Flags for listing directories* [rclone](https://rclone.org/commands/rclone/) - Show help for rclone commands, flags and backends.+<!-- markdownlint-restore -->++# rclone index++Write static directory listings into a remote.++## Synopsis++Write a directory listing, `index.html` by default, into every+directory of `remote:path` so that it can be browsed when it is served+as a static website, for example from an S3 website endpoint, a bucket+behind a CDN, or a plain web server.++This works like `rclone sync`: each run makes the minimum changes needed+to bring the listings into line with the files. Listings which haven't+changed aren't rewritten, so a run with no changes makes no uploads. The+listings never show the listing files themselves.++**Note** that `rclone index` treats every file with an output name+(`index.html` by default) in the directories it indexes as its own, and+will overwrite it or delete it without checking where it came from. To+protect a hand-written index, exclude its directory with+`--index-exclude`.++## Outputs++`--output NAME=FORMAT` says what to write in each directory and can be+repeated. NAME is the file name, and FORMAT is one of the built-in+formats below or the path of a Go template file.++| Format | MIME type | Contents |+|:-------|:----------|:---------|+| `html` | `text/html; charset=utf-8` | The HTML listing from the `rclone serve http` template |+| `json` | `application/json` | JSON in the format of `rclone lsjson` for the directory |+| `caddy` | `application/json` | JSON in the format of Caddy's `file_server browse` |++The default is `--output index.html=html`. The built-in formats are+templates, so they can be printed with `--print-template FORMAT`, copied+and changed. Outputs whose name ends in `.html` or `.htm` are rendered+with Go's `html/template` and everything else with `text/template`, which+has a `json` function for safe encoding. The template data is that of+`rclone serve http` with `.Static` set (see+[rclone serve http](https://rclone.org/commands/rclone_serve_http/#template)). Outputs are+uploaded with the MIME type of the built-in format, or that of the+output name's extension for template outputs, since static hosts serve+the stored type.++## Filters++The normal filters (`--include`, `--exclude`, `--filter-from`, `--max-age`, ...)+control what appears in the listings. An excluded file isn't listed, and+an excluded directory isn't listed or indexed.++The `--index-include`, `--index-exclude`, `--index-filter` flags and their+`-from` variants control which directories get listings. They take the same+patterns as the normal filters, applied to directory paths, and+`--index-include` implies excluding all other directories. A directory+excluded here still appears in its parent's listing, but nothing is written+or deleted inside it, so it can have its own hand-made index.+`--index-max-depth` limits how deep listings are written.++## Modification times++Listings show the modification time of each file. When using the s3,+oracleobjectstorage and swift backends rclone stores the modification time as+metadata which isn't returned by the bucket listing, so **every object needs+a HEAD request**, which can mean thousands of transactions per run. Avoid+this with one of:++- `--use-server-modtime` uses the time the object was uploaded instead,+ which is what the web server sends as `Last-Modified`. Recommended for s3+ unless the original modification times matter.+- `--no-modtime` shows no times at all, which is the cheapest option.++`--dir-time` sets the time shown for directories:++- `newest` (default): the newest modification time of anything below the+ directory. Note that this means one new file changes the listing of+ every directory above it.+- `dir`: the directory's own modification time. On most file systems+ writing a listing into a directory changes its time, so this can take+ more than one run to settle.+- `none`: no time.++The listing files are given the newest modification time below their+directory, so the `Last-Modified` header of a listing is meaningful.++## Links++Directory links are `dir/` by default, which works on hosts that serve+index documents: S3 website endpoints, GCS and Azure static websites,+nginx and Apache, and R2 custom domains with a URL rewrite rule. For+hosts that don't (plain S3 REST URLs, B2 friendly URLs) use+`--link-index` to make directory links `dir/index.html` instead.++## Partial runs++After a sync which changed a known set of files there is no need to+walk the whole remote. Tell `rclone index` what changed and it re-indexes+only the directories that could be affected:++ rclone index r2:bucket --changed v1.76.0/ --changed version.txt+ rclone index r2:bucket --changed-from changes.txt+ rclone sync ./public r2:bucket --exclude index.html --combined - | rclone index r2:bucket --changed-combined -++- `--changed PATH` names a changed file or directory relative to+ `remote:path` and can be repeated. A directory, given with a trailing+ slash or found to be one, is re-indexed completely.+- `--changed-from FILE` reads one path per line exactly as written.+- `--changed-combined FILE` reads the `--combined` report of+ `rclone sync`: lines starting with `+`, `-`, `*` and `!` are changes and+ `=` lines are ignored.+- Both files accept `-` for standard input and all three flags can be+ combined.++A changed file re-indexes its directory and every directory above it,+since their times and their listings can change too, and on bucket+based storage a directory can cease to exist. Each of those is one+directory listing. With `--dir-time newest` the directories above also+need the times of their unchanged subdirectories, which are read from+the modification times of those subdirectories' listings at the cost+of one small request each. With `--use-server-modtime` that is the time+the listing was written rather than the time of the newest file, so+partial runs can drift by the indexing delay until the next full run.+`--dir-time dir` and `none` need no such lookups.++Directories which are neither named nor above a named path are left+alone, as are directories deeper than `--max-depth`. An empty change+list does nothing, and naming the root does a full run. There is no+automatic cutover to a full run since that depends on the size of the+whole remote: a partial run costs one listing per affected directory,+a full run one request per thousand objects on S3-like storage or one+per directory elsewhere. The number+of directories a partial run will list is logged with `-v`, and+`--changed-max-dirs` makes it fall back to a full run above that+number. A good pattern is a partial run after each upload and a+scheduled full run to catch anything missed.++## Site icon++The listings carry no icon or branding of their own. To give a site an+icon, upload a `favicon.ico` to the root of the remote, which browsers+fetch from there for every page. Exclude it from the listings with+`--exclude /favicon.ico` so it stays on the site but isn't shown.++## Deleting listings++On backends which can't have empty directories (S3, B2 and other+bucket-based storage) a directory only exists while it has files in it.+Once the last file in a directory has been removed its listing is+deleted, since the directory only existed because of it. Backends which+can have empty directories keep a listing for each empty directory.++## Using with sync++A sync from a local build deletes files on the destination which aren't+in the source, which includes the listings. Either index the local build+before syncing:++ rclone index ./public+ rclone sync ./public r2:bucket++or exclude the listings from the sync so it neither uploads nor deletes+them, then index the remote:++ rclone sync ./public r2:bucket --exclude index.html+ rclone index r2:bucket++## Examples++An S3 website bucket, using the upload times so no object needs a HEAD+request:++ rclone index s3:my-site --use-server-modtime++HTML and rclone JSON listings, setting `Cache-Control` on them:++ rclone index s3:bucket --output index.html=html --output index.json=json --header-upload "Cache-Control: max-age=300"++Only list release files, and don't write listings under `/private`:++ rclone index r2:bucket --include "*.{zip,deb,rpm}" --index-exclude "/private/**"++A B2 bucket served from its friendly URLs:++ rclone index b2:bucket --link-index++See what would change:++ rclone index r2:bucket --dry-run -v++Listings which haven't changed aren't normally rewritten, so changing+the headers or metadata set on them needs `--index-rewrite`, which+writes every listing a run would render whether it changed or not:++ rclone index r2:bucket --index-rewrite --header-upload "Cache-Control: max-age=3600"++```+rclone index remote:path [flags]+```++## Options++```+ --changed stringArray Only re-index directories affected by this changed path (directories end in /)+ --changed-combined stringArray Read changed paths from a sync --combined report (use - to read from stdin)+ --changed-from stringArray Read changed paths from file, one per line (use - to read from stdin)+ --changed-max-dirs int Do a full index if a partial one would list more than this many directories (0 for no limit)+ --dir-time newest|dir|none How to work out the time shown for a directory (default newest)+ -h, --help help for index+ --index-exclude stringArray Don't write listings in directories matching pattern+ --index-exclude-from stringArray Read directory exclude patterns from file (use - to read from stdin)+ --index-filter stringArray Add a rule for which directories get listings+ --index-filter-from stringArray Read directory rules from a file (use - to read from stdin)+ --index-include stringArray Only write listings in directories matching pattern+ --index-include-from stringArray Read directory include patterns from file (use - to read from stdin)+ --index-max-depth int Only write listings this many directories deep (-1 for no limit) (default -1)+ --index-rewrite Write every listing even if it is unchanged+ --link-index Link to dir/NAME rather than dir/ for hosts without index documents+ --no-modtime Don't show modification times in listings+ --output stringArray Listing to write in each directory as NAME=FORMAT (default [index.html=html])+ --print-template string Print the built-in template for FORMAT and exit+```++Options shared with other commands are described next.+See the [global flags page](https://rclone.org/flags/) for global options not listed here.++### Copy Options++Flags for anything which can copy a file++```text+ --check-first Do all the checks before starting transfers+ -c, --checksum Check for changes with size & checksum (if available, or fallback to size only)+ --compare-dest stringArray Include additional server-side paths during comparison+ --copy-dest stringArray Implies --compare-dest but also copies files from paths into destination+ --cutoff-mode HARD|SOFT|CAUTIOUS Mode to stop transfers when reaching the max transfer limit HARD|SOFT|CAUTIOUS (default HARD)+ --ignore-case-sync Ignore case when synchronizing+ --ignore-checksum Skip post copy check of checksums+ --ignore-existing Skip all files that exist on destination+ --ignore-size Ignore size when skipping use modtime or checksum+ -I, --ignore-times Don't skip items that match size and time - transfer all unconditionally+ --immutable Do not modify files, fail if existing files have been modified+ --inplace Download directly to destination file instead of atomic download to temp/rename+ -l, --links Translate symlinks to/from regular files with a '.rclonelink' extension+ --max-backlog int Maximum number of objects in sync or check backlog (default 10000)+ --max-duration Duration Maximum duration rclone will transfer data for (default 0s)+ --max-transfer SizeSuffix Maximum size of data to transfer (default off)+ -M, --metadata If set, preserve metadata when copying objects+ --modify-window Duration Max time diff to be considered the same (default 1ns)+ --multi-thread-chunk-size SizeSuffix Chunk size for multi-thread downloads / uploads, if not set by filesystem (default 64Mi)+ --multi-thread-cutoff SizeSuffix Use multi-thread downloads for files above this size (default 256Mi)+ --multi-thread-streams int Number of streams to use for multi-thread downloads (default 4)+ --multi-thread-write-buffer-size SizeSuffix In memory buffer size for writing when in multi-thread mode (default 128Ki)+ --name-transform stringArray Transform paths during the copy process+ --no-check-dest Don't check the destination, copy regardless+ --no-traverse Don't traverse destination file system on copy+ --no-update-dir-modtime Don't update directory modification times+ --no-update-modtime Don't update destination modtime if files identical+ --order-by string Instructions on how to order the transfers, e.g. 'size,descending'+ --partial-suffix string Add partial-suffix to temporary file name when --inplace is not used (default ".partial")+ --refresh-times Refresh the modtime of remote files+ --server-side-across-configs Allow server-side operations (e.g. copy) to work across different configs+ --size-only Skip based on size only, not modtime or checksum+ --streaming-upload-cutoff SizeSuffix Cutoff for switching to chunked upload if file size is unknown, upload starts after reaching cutoff or when file ends (default 100Ki)+ -u, --update Skip files that are newer on the destination+```++### Important Options++Important flags useful for most commands++```text+ -n, --dry-run Do a trial run with no permanent changes+ -i, --interactive Enable interactive mode+ -v, --verbose count Print lots more stuff (repeat for more)+```++### Filter Options++Flags for filtering directory listings++```text+ --delete-excluded Delete files on dest excluded from sync+ --exclude stringArray Exclude files matching pattern+ --exclude-from stringArray Read file exclude patterns from file (use - to read from stdin)+ --exclude-if-present stringArray Exclude directories if filename is present+ --files-from stringArray Read list of source-file names from file (use - to read from stdin)+ --files-from-raw stringArray Read list of source-file names from file without any processing of lines (use - to read from stdin)+ --files-from0 stringArray Read list of source-file names from file using NUL as separator (use - to read from stdin)+ -f, --filter stringArray Add a file filtering rule+ --filter-from stringArray Read file filtering patterns from a file (use - to read from stdin)+ --hash-filter string Partition filenames by hash k/n or randomly @/n+ --ignore-case Ignore case in filters (case insensitive)+ --include stringArray Include files matching pattern+ --include-from stringArray Read file include patterns from file (use - to read from stdin)+ --max-age Duration Only transfer files younger than this in s or suffix ms|s|m|h|d|w|M|y (default off)+ --max-depth int If set limits the recursion depth to this (default -1)+ --max-size SizeSuffix Only transfer files smaller than this in KiB or suffix B|K|M|G|T|P (default off)+ --metadata-exclude stringArray Exclude metadatas matching pattern+ --metadata-exclude-from stringArray Read metadata exclude patterns from file (use - to read from stdin)+ --metadata-filter stringArray Add a metadata filtering rule+ --metadata-filter-from stringArray Read metadata filtering patterns from a file (use - to read from stdin)+ --metadata-include stringArray Include metadatas matching pattern+ --metadata-include-from stringArray Read metadata include patterns from file (use - to read from stdin)+ --min-age Duration Only transfer files older than this in s or suffix ms|s|m|h|d|w|M|y (default off)+ --min-size SizeSuffix Only transfer files bigger than this in KiB or suffix B|K|M|G|T|P (default off)+```++### Listing Options++Flags for listing directories++```text+ --default-time Time Time to show if modtime is unknown for files and directories (default 2000-01-01T00:00:00Z)+ --fast-list Use recursive list if available; uses more memory but fewer transactions+```++## See Also++<!-- markdownlint-capture -->+<!-- markdownlint-disable ul-style line-length -->++* [rclone](https://rclone.org/commands/rclone/) - Show help for rclone commands, flags and backends.++<!-- markdownlint-restore --># rclone link@@ -6844,6 +7208,10 @@ Note that `ls` and `lsl` recurse by default - use `--max-depth 1` to stop the reThe other list commands `lsd`,`lsf`,`lsjson` do not recurse by default -use `-R` to make them recurse.+On the bucket-based remotes (e.g. s3, gcs, b2) a directory is a key prefix+rather than an object, so a listing without `-R` shows each prefix as one+directory entry and not the objects stored under it.+List commands prefer a recursive method that uses more memory but fewertransactions by default. Use `--disable ListR` to suppress the behavior.See [`--fast-list`](https://rclone.org/docs/#fast-list) for more details.@@ -7032,6 +7400,10 @@ Note that `ls` and `lsl` recurse by default - use `--max-depth 1` to stop the reThe other list commands `lsd`,`lsf`,`lsjson` do not recurse by default -use `-R` to make them recurse.+On the bucket-based remotes (e.g. s3, gcs, b2) a directory is a key prefix+rather than an object, so a listing without `-R` shows each prefix as one+directory entry and not the objects stored under it.+List commands prefer a recursive method that uses more memory but fewertransactions by default. Use `--disable ListR` to suppress the behavior.See [`--fast-list`](https://rclone.org/docs/#fast-list) for more details.@@ -7402,6 +7774,11 @@ for all `mount` and `serve` commands on macOS. For details, see [vfs-case-sensit### NFS mount+For macOS (and other platforms where this path is supported), prefer the dedicated+[rclone nfsmount](https://rclone.org/commands/rclone_nfsmount/) command. It starts the NFS server and+performs the mount for you. `rclone mount` itself still uses FUSE (macFUSE/FUSE-T)+and does not switch to NFS via a flag.+This method spins up an NFS server using [serve nfs](https://rclone.org/commands/rclone_serve_nfs/)command and mounts it to the specified mountpoint. If you run this in backgroundmode using |--daemon|, you will need to send SIGTERM signal to the rclone process@@ -7739,7 +8116,7 @@ at all times. The buffered data is bound to one open file and won't beshared.This flag is a upper limit for the used memory per open file. The-buffer will only use memory for data that is downloaded but not not+buffer will only use memory for data that is downloaded but notyet read. If the buffer is empty, only a small amount of memory willbe used.@@ -7794,7 +8171,8 @@ longest. This cache flushing strategy is efficient and more relevantfiles are likely to remain cached.The `--vfs-cache-max-age` will evict files from the cache-after the set time since last access has passed. The default value of+after the set time since last access has passed; it is based on access time,+not on when the file was first added to the cache. The default value of1 hour will start evicting files from cache that haven't been accessedfor 1 hour. When a cached file is accessed the 1 hour timer is reset to 0and will wait for 1 more hour before evicting. Specify the time with@@ -7948,7 +8326,7 @@ Rclone reads `--vfs-read-chunk-streams` chunks of size`--vfs-read-chunk-size` concurrently. The size for each read will stayconstant.-This improves performance performance massively on high latency links+This improves performance massively on high latency linksor very high bandwidth links to high performance object stores.Some experimentation will be needed to find the optimum values of@@ -8908,6 +9286,11 @@ for all `mount` and `serve` commands on macOS. For details, see [vfs-case-sensit### NFS mount+For macOS (and other platforms where this path is supported), prefer the dedicated+[rclone nfsmount](https://rclone.org/commands/rclone_nfsmount/) command. It starts the NFS server and+performs the mount for you. `rclone mount` itself still uses FUSE (macFUSE/FUSE-T)+and does not switch to NFS via a flag.+This method spins up an NFS server using [serve nfs](https://rclone.org/commands/rclone_serve_nfs/)command and mounts it to the specified mountpoint. If you run this in backgroundmode using |--daemon|, you will need to send SIGTERM signal to the rclone process@@ -9245,7 +9628,7 @@ at all times. The buffered data is bound to one open file and won't beshared.This flag is a upper limit for the used memory per open file. The-buffer will only use memory for data that is downloaded but not not+buffer will only use memory for data that is downloaded but notyet read. If the buffer is empty, only a small amount of memory willbe used.@@ -9300,7 +9683,8 @@ longest. This cache flushing strategy is efficient and more relevantfiles are likely to remain cached.The `--vfs-cache-max-age` will evict files from the cache-after the set time since last access has passed. The default value of+after the set time since last access has passed; it is based on access time,+not on when the file was first added to the cache. The default value of1 hour will start evicting files from cache that haven't been accessedfor 1 hour. When a cached file is accessed the 1 hour timer is reset to 0and will wait for 1 more hour before evicting. Specify the time with@@ -9454,7 +9838,7 @@ Rclone reads `--vfs-read-chunk-streams` chunks of size`--vfs-read-chunk-size` concurrently. The size for each read will stayconstant.-This improves performance performance massively on high latency links+This improves performance massively on high latency linksor very high bandwidth links to high performance object stores.Some experimentation will be needed to find the optimum values of@@ -10058,6 +10442,13 @@ accept in the HTTP header.will overriding existing values. The flag may be repeated to add multipleheaders. Use the format `Header-Name: value`.+`--rc-allow-origin` enables cross-origin (CORS) requests from browsers+by setting the `Access-Control-Allow-Origin` header on every response to the+origin given, e.g. `--rc-allow-origin https://app.example.com`. Browser+CORS preflight OPTIONS requests are answered by rclone without needing+authentication as browsers never send credentials with them. Setting this+to `*` allows any origin which is a security risk on an authenticated server.+`--rc-baseurl` controls the URL prefix that rclone serves from. By defaultrclone will serve from the root. If you used `--rc-baseurl "/rclone"` thenrclone would serve from a URL starting with "/rclone/". This is@@ -10127,6 +10518,13 @@ to be used within the template to server pages:| .Sort | | The current sort used. This is changeable via '?sort=' parameter. Possible values: namedirfirst, name, size, time (default namedirfirst). || .Order | | The current ordering used. This is changeable via '?order=' parameter. Possible values: asc, desc (default asc). || .Query | | Currently unused. |+| .Path | | The path of the directory from the root with leading and trailing slashes, e.g. '/dir/'. |+| .IsRoot | | Boolean for if this is the root directory. |+| .UpLink | | The link to the parent directory. |+| .Static | | Boolean for if the listing is being written as a static page by rclone index. |+| .NumDirs | | The number of directories in the listing. |+| .NumFiles | | The number of files in the listing. |+| .TotalSize | | The total size in bytes of the files in the listing. || .Breadcrumb | | Allows for creating a relative navigation. || | .Link | The link of the Text relative to the root. || | .Text | The Name of the directory. |@@ -10136,6 +10534,7 @@ to be used within the template to server pages:| | .IsDir | Boolean for if an entry is a directory or not. || | .Size | Size in bytes of the entry. || | .ModTime | The UTC timestamp of an entry. |+| | .MimeType | The MIME type of an entry guessed from its name, or 'inode/directory'. |The server also makes the following functions available so that they can be usedwithin the template. These functions help extend the options for dynamic@@ -10176,8 +10575,20 @@ htpasswd -B htpasswd userhtpasswd -B htpasswd anotherUser```+Use the same hash format and cost for every entry in the file. If the+entries differ, the time taken to reject a wrong password can reveal+which usernames exist.+The password file can be updated while rclone is running.+When authentication is enabled every request must be authenticated,+including HTTP OPTIONS requests, as the answer to an OPTIONS request can+reveal whether a path exists. Health checks or monitoring which use+unauthenticated OPTIONS requests will need to send credentials. The+exception is browser CORS preflight requests when `--rc-allow-origin`+is set, which are answered without authentication as browsers never send+credentials with them.+Use `--rc-realm` to set the authentication realm.Use `--rc-salt` to change the password hashing salt from the default.@@ -10226,11 +10637,6 @@ Flags to control the Remote Control API--rc-template string User-specified template--rc-user string User name for authentication--rc-user-from-header string User name from a defined HTTP header- --rc-web-fetch-url string URL to fetch the releases for webgui (default "https://api.github.com/repos/rclone/rclone-webui-react/releases/latest")- --rc-web-gui Launch WebGUI on localhost- --rc-web-gui-force-update Force update to latest version of web gui- --rc-web-gui-no-open-browser Don't open the browser automatically- --rc-web-gui-update Check and update to latest version of web gui```## See Also@@ -10534,7 +10940,7 @@ at all times. The buffered data is bound to one open file and won't beshared.This flag is a upper limit for the used memory per open file. The-buffer will only use memory for data that is downloaded but not not+buffer will only use memory for data that is downloaded but notyet read. If the buffer is empty, only a small amount of memory willbe used.@@ -10589,7 +10995,8 @@ longest. This cache flushing strategy is efficient and more relevantfiles are likely to remain cached.The `--vfs-cache-max-age` will evict files from the cache-after the set time since last access has passed. The default value of+after the set time since last access has passed; it is based on access time,+not on when the file was first added to the cache. The default value of1 hour will start evicting files from cache that haven't been accessedfor 1 hour. When a cached file is accessed the 1 hour timer is reset to 0and will wait for 1 more hour before evicting. Specify the time with@@ -10743,7 +11150,7 @@ Rclone reads `--vfs-read-chunk-streams` chunks of size`--vfs-read-chunk-size` concurrently. The size for each read will stayconstant.-This improves performance performance massively on high latency links+This improves performance massively on high latency linksor very high bandwidth links to high performance object stores.Some experimentation will be needed to find the optimum values of@@ -11210,7 +11617,7 @@ at all times. The buffered data is bound to one open file and won't beshared.This flag is a upper limit for the used memory per open file. The-buffer will only use memory for data that is downloaded but not not+buffer will only use memory for data that is downloaded but notyet read. If the buffer is empty, only a small amount of memory willbe used.@@ -11265,7 +11672,8 @@ longest. This cache flushing strategy is efficient and more relevantfiles are likely to remain cached.The `--vfs-cache-max-age` will evict files from the cache-after the set time since last access has passed. The default value of+after the set time since last access has passed; it is based on access time,+not on when the file was first added to the cache. The default value of1 hour will start evicting files from cache that haven't been accessedfor 1 hour. When a cached file is accessed the 1 hour timer is reset to 0and will wait for 1 more hour before evicting. Specify the time with@@ -11419,7 +11827,7 @@ Rclone reads `--vfs-read-chunk-streams` chunks of size`--vfs-read-chunk-size` concurrently. The size for each read will stayconstant.-This improves performance performance massively on high latency links+This improves performance massively on high latency linksor very high bandwidth links to high performance object stores.Some experimentation will be needed to find the optimum values of@@ -11839,7 +12247,7 @@ at all times. The buffered data is bound to one open file and won't beshared.This flag is a upper limit for the used memory per open file. The-buffer will only use memory for data that is downloaded but not not+buffer will only use memory for data that is downloaded but notyet read. If the buffer is empty, only a small amount of memory willbe used.@@ -11894,7 +12302,8 @@ longest. This cache flushing strategy is efficient and more relevantfiles are likely to remain cached.The `--vfs-cache-max-age` will evict files from the cache-after the set time since last access has passed. The default value of+after the set time since last access has passed; it is based on access time,+not on when the file was first added to the cache. The default value of1 hour will start evicting files from cache that haven't been accessedfor 1 hour. When a cached file is accessed the 1 hour timer is reset to 0and will wait for 1 more hour before evicting. Specify the time with@@ -12048,7 +12457,7 @@ Rclone reads `--vfs-read-chunk-streams` chunks of size`--vfs-read-chunk-size` concurrently. The size for each read will stayconstant.-This improves performance performance massively on high latency links+This improves performance massively on high latency linksor very high bandwidth links to high performance object stores.Some experimentation will be needed to find the optimum values of@@ -12291,9 +12700,10 @@ This config generated must have this extra parameter- `_root` - root to use for the backend-And it may have this parameter+And it may have these parameters- `_obscure` - comma separated strings for parameters to obscure+- `_secret_access_key` - the secret for S3 access key auth (see below)If password authentication was used by the client, input to the proxyprocess (on STDIN) would look similar to this:@@ -12301,7 +12711,8 @@ process (on STDIN) would look similar to this:```json{"user": "me",- "pass": "mypassword"+ "pass": "mypassword",+ "client_ip": "192.168.1.1"}```@@ -12311,10 +12722,44 @@ proxy process (on STDIN) would look similar to this:```json{"user": "me",- "public_key": "AAAAB3NzaC1yc2EAAAADAQABAAABAQDuwESFdAe14hVS6omeyX7edc...JQdf"+ "public_key": "AAAAB3NzaC1yc2EAAAADAQABAAABAQDuwESFdAe14hVS6omeyX7edc...JQdf",+ "client_ip": "192.168.1.1"+}+```++If the client authenticated with an S3 access key (`rclone serve s3`),+the client never sends its secret, only a signature made with it, so+the input contains just the access key ID as the `user` with no `pass`+or `public_key`:++```json+{+ "user": "AKIAIOSFODNN7EXAMPLE",+ "client_ip": "192.168.1.1"}```+In this case the program must look up the secret access key for that+access key ID and return it in the `_secret_access_key` field of the+output. Rclone then uses that secret to verify the signature on the+request, refusing the request if it does not match. This means the+proxy program is the source of truth for both the credentials and the+backend they map to. If the program does not return+`_secret_access_key` or returns it empty the request is refused.++The program's answer for an access key ID is cached (see below) but+is checked with the program again after 5 minutes even if the access+key ID is in constant use, so revoking an access key ID in the+program takes effect within 5 minutes. A rotated secret takes effect+on the first request signed with it.++The `client_ip` key holds the IP address the client connected from,+without a port number. It can be used to restrict logins to certain+networks, or to log authentication attempts centrally. It is omitted if+the client has no IP address, for example when connecting over a unix+socket. Note that if rclone is behind a reverse proxy this will be the+address of the reverse proxy and not the original client.+And as an example return this on STDOUT```json@@ -12340,11 +12785,20 @@ to make proxy to many different sftp backends, you could make thein the output and the user to `user`. For security you'd probably wantto restrict the `host` to a limited list.-An internal cache of backends is keyed on the `user` and a hash of the-`pass` or `public_key`. This means that if a user's password or-public-key changes, or the proxy returns different config parameters-(eg a rotated `api_key`), a fresh backend will be created on the next-request rather than the cached one being reused.+An internal cache of backends is keyed on the `user`, the kind of+credential presented, a hash of the `pass` or `public_key`, and the+`client_ip`. This means that if a user's password or public-key+changes, the client connects from a new IP address, or the proxy+returns different config parameters (eg a rotated `api_key`), a fresh+backend will be created on the next request rather than the cached one+being reused. A backend created for a public key login is never+reused for a password login, so a public key cannot be sent as a+password to skip the program.++A cached answer is checked with the program again after 5 minutes even+if the user is in constant use, so revoking a user in the program+takes effect within 5 minutes. For `serve sftp` this applies to new+logins - an existing connection is not closed.This can be used to build general purpose proxies to any kind ofbackend that rclone supports.@@ -12489,6 +12943,13 @@ accept in the HTTP header.will overriding existing values. The flag may be repeated to add multipleheaders. Use the format `Header-Name: value`.+`--allow-origin` enables cross-origin (CORS) requests from browsers+by setting the `Access-Control-Allow-Origin` header on every response to the+origin given, e.g. `--allow-origin https://app.example.com`. Browser+CORS preflight OPTIONS requests are answered by rclone without needing+authentication as browsers never send credentials with them. Setting this+to `*` allows any origin which is a security risk on an authenticated server.+`--baseurl` controls the URL prefix that rclone serves from. By defaultrclone will serve from the root. If you used `--baseurl "/rclone"` thenrclone would serve from a URL starting with "/rclone/". This is@@ -12558,6 +13019,13 @@ to be used within the template to server pages:| .Sort | | The current sort used. This is changeable via '?sort=' parameter. Possible values: namedirfirst, name, size, time (default namedirfirst). || .Order | | The current ordering used. This is changeable via '?order=' parameter. Possible values: asc, desc (default asc). || .Query | | Currently unused. |+| .Path | | The path of the directory from the root with leading and trailing slashes, e.g. '/dir/'. |+| .IsRoot | | Boolean for if this is the root directory. |+| .UpLink | | The link to the parent directory. |+| .Static | | Boolean for if the listing is being written as a static page by rclone index. |+| .NumDirs | | The number of directories in the listing. |+| .NumFiles | | The number of files in the listing. |+| .TotalSize | | The total size in bytes of the files in the listing. || .Breadcrumb | | Allows for creating a relative navigation. || | .Link | The link of the Text relative to the root. || | .Text | The Name of the directory. |@@ -12567,6 +13035,7 @@ to be used within the template to server pages:| | .IsDir | Boolean for if an entry is a directory or not. || | .Size | Size in bytes of the entry. || | .ModTime | The UTC timestamp of an entry. |+| | .MimeType | The MIME type of an entry guessed from its name, or 'inode/directory'. |The server also makes the following functions available so that they can be usedwithin the template. These functions help extend the options for dynamic@@ -12607,8 +13076,20 @@ htpasswd -B htpasswd userhtpasswd -B htpasswd anotherUser```+Use the same hash format and cost for every entry in the file. If the+entries differ, the time taken to reject a wrong password can reveal+which usernames exist.+The password file can be updated while rclone is running.+When authentication is enabled every request must be authenticated,+including HTTP OPTIONS requests, as the answer to an OPTIONS request can+reveal whether a path exists. Health checks or monitoring which use+unauthenticated OPTIONS requests will need to send credentials. The+exception is browser CORS preflight requests when `--allow-origin`+is set, which are answered without authentication as browsers never send+credentials with them.+Use `--realm` to set the authentication realm.Use `--salt` to change the password hashing salt from the default.@@ -12676,7 +13157,7 @@ at all times. The buffered data is bound to one open file and won't beshared.This flag is a upper limit for the used memory per open file. The-buffer will only use memory for data that is downloaded but not not+buffer will only use memory for data that is downloaded but notyet read. If the buffer is empty, only a small amount of memory willbe used.@@ -12731,7 +13212,8 @@ longest. This cache flushing strategy is efficient and more relevantfiles are likely to remain cached.The `--vfs-cache-max-age` will evict files from the cache-after the set time since last access has passed. The default value of+after the set time since last access has passed; it is based on access time,+not on when the file was first added to the cache. The default value of1 hour will start evicting files from cache that haven't been accessedfor 1 hour. When a cached file is accessed the 1 hour timer is reset to 0and will wait for 1 more hour before evicting. Specify the time with@@ -12885,7 +13367,7 @@ Rclone reads `--vfs-read-chunk-streams` chunks of size`--vfs-read-chunk-size` concurrently. The size for each read will stayconstant.-This improves performance performance massively on high latency links+This improves performance massively on high latency linksor very high bandwidth links to high performance object stores.Some experimentation will be needed to find the optimum values of@@ -13128,9 +13610,10 @@ This config generated must have this extra parameter- `_root` - root to use for the backend-And it may have this parameter+And it may have these parameters- `_obscure` - comma separated strings for parameters to obscure+- `_secret_access_key` - the secret for S3 access key auth (see below)If password authentication was used by the client, input to the proxyprocess (on STDIN) would look similar to this:@@ -13138,7 +13621,8 @@ process (on STDIN) would look similar to this:```json{"user": "me",- "pass": "mypassword"+ "pass": "mypassword",+ "client_ip": "192.168.1.1"}```@@ -13148,10 +13632,44 @@ proxy process (on STDIN) would look similar to this:```json{"user": "me",- "public_key": "AAAAB3NzaC1yc2EAAAADAQABAAABAQDuwESFdAe14hVS6omeyX7edc...JQdf"+ "public_key": "AAAAB3NzaC1yc2EAAAADAQABAAABAQDuwESFdAe14hVS6omeyX7edc...JQdf",+ "client_ip": "192.168.1.1"}```+If the client authenticated with an S3 access key (`rclone serve s3`),+the client never sends its secret, only a signature made with it, so+the input contains just the access key ID as the `user` with no `pass`+or `public_key`:++```json+{+ "user": "AKIAIOSFODNN7EXAMPLE",+ "client_ip": "192.168.1.1"+}+```++In this case the program must look up the secret access key for that+access key ID and return it in the `_secret_access_key` field of the+output. Rclone then uses that secret to verify the signature on the+request, refusing the request if it does not match. This means the+proxy program is the source of truth for both the credentials and the+backend they map to. If the program does not return+`_secret_access_key` or returns it empty the request is refused.++The program's answer for an access key ID is cached (see below) but+is checked with the program again after 5 minutes even if the access+key ID is in constant use, so revoking an access key ID in the+program takes effect within 5 minutes. A rotated secret takes effect+on the first request signed with it.++The `client_ip` key holds the IP address the client connected from,+without a port number. It can be used to restrict logins to certain+networks, or to log authentication attempts centrally. It is omitted if+the client has no IP address, for example when connecting over a unix+socket. Note that if rclone is behind a reverse proxy this will be the+address of the reverse proxy and not the original client.+And as an example return this on STDOUT```json@@ -13177,11 +13695,20 @@ to make proxy to many different sftp backends, you could make thein the output and the user to `user`. For security you'd probably wantto restrict the `host` to a limited list.-An internal cache of backends is keyed on the `user` and a hash of the-`pass` or `public_key`. This means that if a user's password or-public-key changes, or the proxy returns different config parameters-(eg a rotated `api_key`), a fresh backend will be created on the next-request rather than the cached one being reused.+An internal cache of backends is keyed on the `user`, the kind of+credential presented, a hash of the `pass` or `public_key`, and the+`client_ip`. This means that if a user's password or public-key+changes, the client connects from a new IP address, or the proxy+returns different config parameters (eg a rotated `api_key`), a fresh+backend will be created on the next request rather than the cached one+being reused. A backend created for a public key login is never+reused for a password login, so a public key cannot be sent as a+password to skip the program.++A cached answer is checked with the program again after 5 minutes even+if the user is in constant use, so revoking a user in the program+takes effect within 5 minutes. For `serve sftp` this applies to new+logins - an existing connection is not closed.This can be used to build general purpose proxies to any kind ofbackend that rclone supports.@@ -13456,7 +13983,7 @@ at all times. The buffered data is bound to one open file and won't beshared.This flag is a upper limit for the used memory per open file. The-buffer will only use memory for data that is downloaded but not not+buffer will only use memory for data that is downloaded but notyet read. If the buffer is empty, only a small amount of memory willbe used.@@ -13511,7 +14038,8 @@ longest. This cache flushing strategy is efficient and more relevantfiles are likely to remain cached.The `--vfs-cache-max-age` will evict files from the cache-after the set time since last access has passed. The default value of+after the set time since last access has passed; it is based on access time,+not on when the file was first added to the cache. The default value of1 hour will start evicting files from cache that haven't been accessedfor 1 hour. When a cached file is accessed the 1 hour timer is reset to 0and will wait for 1 more hour before evicting. Specify the time with@@ -13665,7 +14193,7 @@ Rclone reads `--vfs-read-chunk-streams` chunks of size`--vfs-read-chunk-size` concurrently. The size for each read will stayconstant.-This improves performance performance massively on high latency links+This improves performance massively on high latency linksor very high bandwidth links to high performance object stores.Some experimentation will be needed to find the optimum values of@@ -14095,6 +14623,13 @@ accept in the HTTP header.will overriding existing values. The flag may be repeated to add multipleheaders. Use the format `Header-Name: value`.+`--allow-origin` enables cross-origin (CORS) requests from browsers+by setting the `Access-Control-Allow-Origin` header on every response to the+origin given, e.g. `--allow-origin https://app.example.com`. Browser+CORS preflight OPTIONS requests are answered by rclone without needing+authentication as browsers never send credentials with them. Setting this+to `*` allows any origin which is a security risk on an authenticated server.+`--baseurl` controls the URL prefix that rclone serves from. By defaultrclone will serve from the root. If you used `--baseurl "/rclone"` thenrclone would serve from a URL starting with "/rclone/". This is@@ -14179,8 +14714,20 @@ htpasswd -B htpasswd userhtpasswd -B htpasswd anotherUser```+Use the same hash format and cost for every entry in the file. If the+entries differ, the time taken to reject a wrong password can reveal+which usernames exist.+The password file can be updated while rclone is running.+When authentication is enabled every request must be authenticated,+including HTTP OPTIONS requests, as the answer to an OPTIONS request can+reveal whether a path exists. Health checks or monitoring which use+unauthenticated OPTIONS requests will need to send credentials. The+exception is browser CORS preflight requests when `--allow-origin`+is set, which are answered without authentication as browsers never send+credentials with them.+Use `--realm` to set the authentication realm.Use `--salt` to change the password hashing salt from the default.@@ -14249,6 +14796,21 @@ docs](https://docs.aws.amazon.com/general/latest/gr/signature-version-4.html)).`--auth-key` is not provided then `serve s3` will allow anonymousaccess.+Alternatively `--auth-proxy` can be used to look up the secret for each+access key ID and choose the backend it maps to (see [Auth+Proxy](#auth-proxy) below). When an auth proxy is in use `--auth-key`+is ignored and every request must be signed with the secret the proxy+returns for its access key ID. Each access key ID is a separate user,+whose multipart uploads and object metadata are private to it, even if+the proxy maps several access key IDs to the same backend.++As on AWS, every `x-amz-*` header in an authenticated request must be+in its signed headers list; a request with an unsigned `x-amz-*`+header is refused with `AccessDenied`. S3 clients sign these headers+as a matter of course, but a header added after signing (for example+by a proxy in front of the server, or `x-amz-meta-*` headers sent+with a presigned URL which only signs the host) is refused.+Like all rclone flags `--auth-key` can be set via environmentvariables, in this case `RCLONE_AUTH_KEY`. Since this flag can berepeated, the input to `RCLONE_AUTH_KEY` is CSV encoded. Because the@@ -14283,6 +14845,14 @@ part of the hostname (such as mybucket.local)Use `--etag-hash` if you want to change the hash uses for the `ETag`.Note that using anything other than `MD5` (the default) is likely tocause problems for S3 clients which rely on the Etag being the MD5.+Use `--etag-hash auto` to use the best hash the backend supports - with+`--auth-proxy` that of each user's backend.++The `ETag` of an object made by a multipart upload is its hash, like+any other object, rather than the `-N` suffixed `ETag` AWS S3 gives+these. If the remote doesn't have the MD5 of objects, `HeadObject`+and listings only return the MD5 of objects uploaded through `serve+s3` while the object's metadata is kept (see below).## Quickstart@@ -14324,29 +14894,78 @@ access_key_id = ACCESS_KEY_IDsecret_access_key = SECRET_ACCESS_KEY```-## Multipart uploads+## Object uploads (PUT)++A `PutObject` upload only ever changes the object at its key atomically, on+success, a failed or interrupted PUT neither removes nor overwrites the+object already stored at the key, and never leaves a partial object visible+at it.-By default `serve s3` **streams** each multipart upload, in part-number-order, into a single `PutStream` upload to the underlying remote, so the-whole file is never buffered in memory - memory use stays bounded by the-parts in flight. The remote then performs its own internal upload (for-example its own multipart upload, still with bounded memory). This works-for any remote that supports `PutStream`, which is nearly all of them,-including through `crypt`.+Remotes that upload atomically (e.g. object stores such as `s3`) are streamed+straight to the destination. On remotes where a partial upload would+otherwise be visible (e.g. `local`), and whenever `--vfs-cache-mode` is+`writes` or above, the upload is written to a temporary object that is+renamed into place on success; these remotes need to support a server-side+move or copy for this (nearly all do - without move or copy the upload is+written directly and a failed PUT may leave a partial object at the key). If+`serve s3` is killed part-way through an upload the temporary object (named+with a leading `.rclone_temp_put_`) may be left behind; it is hidden from+S3 listings but must be removed manually.-The upload is atomic so the destination object only ever changes on a+## Multipart uploads++Multipart uploads are written, in part-number order, to a temporary+object which is renamed into place, server-side, on completion, so the+upload is atomic. The object at the key only ever changes on asuccessful completion. A failed or aborted upload never affects any-object already stored under that name. Remotes that upload atomically-already (object stores such as `s3`) are streamed straight to the-destination. On remotes where a partial upload would otherwise be visible-(such as `local`), the parts are streamed to a temporary object that is-moved into place, server-side, on completion; these remotes therefore-also need to support a server-side move or copy.+object already stored under that name and a partly-uploaded object+never becomes visible under it.++With the default `--vfs-cache-mode off` `serve s3` **streams** each+multipart upload, in part-number order, into a single streaming upload+to the underlying remote, so the whole file is never buffered in+memory: only parts which arrive out of order are buffered, within the+limits described below. The remote then performs its own internal upload (for example its own multipart+upload, still with bounded memory). Remotes that don't support+streaming uploads (those that must know the file size before the+upload starts, such as `onedrive`, `pcloud`, `jottacloud`, `mailru`,+`opendrive`, `putio`, `protondrive` and `zoho`) have the parts spooled+to a temporary file on **local disk** instead, and uploaded with the+size then known on completion, so they need local disk space for the+largest objects in flight rather than memory.++With `--vfs-cache-mode writes` (or `full`) the parts are written to a+temporary file in the VFS cache and uploaded by the VFS write-back -+see [Multipart uploads and the VFS+cache](#multipart-uploads-and-the-vfs-cache) below.++The rename into place needs the remote to support a server-side move+or copy, which nearly all do. It is a cheap rename on most remotes,+but on object stores without a real rename (such as `s3` itself) the+move is performed as a server-side copy and delete of the whole+object, which can take time and API calls for large objects.+Concurrent multipart uploads of the same key (which S3 permits) are+safe. Each writes its own temporary object and the last to complete+wins.++On the few remotes that support neither server side move nor copy, the+parts are written straight to the destination object instead and never+buffered in memory. This is at some cost in atomicity - the incomplete+object is visible under its final name while the upload is in flight,+as it also is for a plain object PUT on such remotes, and concurrent+multipart uploads of the same key write to the same object and can+interleave. A failed or aborted upload still leaves any pre-existing+object untouched provided the remote uploads atomically and the VFS+cache is off; on a remote where partial uploads are visible it may+leave partial data at the key (like a plain PUT there), and with+`--vfs-cache-mode writes` (or `full`) a write to the cache cannot be+abandoned, so an aborted upload's partial data is written back to the+remote as if it had completed.**Features**-- The whole object is never buffered in memory; memory use is bounded by- the parts in flight, not the upload size.+- The whole object is never buffered in memory; only parts arriving out+ of order are, within fixed limits, whatever the upload size.- Parts can be any size. Clients that don't produce uniform-sized partswork fine - for example PostgreSQL backup tools such as **pgBarman**and **pgBackRest**, which flush an upload buffer once it grows past@@ -14356,69 +14975,275 @@ also need to support a server-side move or copy.as one continuous stream.- The destination object only ever changes atomically, on completion: anaborted or failed upload leaves any pre-existing object of the same- name untouched, and a partly-uploaded object never becomes visible.-- Backend-agnostic - it only needs the remote to support `PutStream`- (plus a server-side move or copy on remotes that don't upload- atomically).+ name untouched, and a partly-uploaded object never becomes visible+ (except on the few remotes with no server-side move or copy, as+ above).+- Multipart uploads go through the VFS like any other upload, so they+ show in rclone's transfer stats and obey `--bwlimit`.+- Backend-agnostic - it only needs the remote to support a server-side+ move or copy for the rename into place, which nearly all do; a remote+ without streaming upload support spools to local disk as above.**Limitations**-- Parts must arrive in ascending, contiguous part-number order- (1, 2, 3, ...). Parts the client uploads concurrently or out of order- are buffered until their turn. The memory used for this buffering is- capped, per upload, by `--multipart-streaming-buffer-limit` (default- `256M`, `0` for no limit): a part that would take the buffer over the- limit is stalled until the stream drains, so a client that uploads- faster than the remote can accept sees backpressure rather than- unbounded server memory use. Since a stalled part holds its HTTP- request open, clients whose upload concurrency times chunk size- exceeds the limit may need a longer read timeout when the remote is- slow. Non-contiguous part numbers are rejected on completion.- Configure the client to upload in part order, ideally with low- concurrency, for the lowest memory use.+- Parts must arrive in ascending, contiguous part-number order (1, 2, 3,+ ...). The part the stream needs next is written straight through as it+ arrives, so it is never buffered, whatever its size. Parts the client+ uploads concurrently or out of order are buffered until their turn.+ The memory used for this buffering is capped, per upload, by+ `--multipart-streaming-buffer-limit` (default `256M`, `0` for no+ limit), and across all the uploads of each user (each access key) by+ `--multipart-streaming-buffer-total` (default `1G`, `0` for no+ limit), counting each buffered part in the+ whole 1 MiB memory pages it occupies: a part that would take the+ buffer over either limit is stalled until the stream drains (or, if it+ is bigger than the limits, until it is the next part), so a client+ that uploads faster than the remote can accept sees backpressure+ rather than unbounded server memory use. A stalled part holds its HTTP+ request open, and one stalled for more than a minute is failed with a+ `SlowDown` error, which S3 clients retry. Clients whose upload+ concurrency times part size exceeds the limits will see these stalls+ and retries when the remote is slower than the client. Non-contiguous+ part numbers are rejected on completion. Configure the client to+ upload in part order, ideally with low concurrency, for the lowest+ memory use.- A part uploaded again before completion - typically a client retryingafter a timeout - is accepted: if the earlier copy is still bufferedit is replaced, and if it has already been streamed an identical- re-upload is a no-op. What isn't possible is replacing a part that has- already been streamed with *different* content - that is rejected. A- failure in the stream to the remote itself still aborts the whole- upload and the client must start it again. (The remote's own upload+ re-upload is a no-op (a re-upload which arrives while the part is+ being streamed waits for it to finish first). If a part fails part way+ through being streamed+ (for example the client's connection drops) the client's retry of it+ carries on where it stopped, provided it has the same content. What+ isn't possible is replacing a part that has already been streamed,+ in whole or in part, with *different* content - that is rejected. A+ part which arrives in full but is then rejected, for example because+ it doesn't match the `Content-MD5` the client declared, has already+ been streamed, so the whole upload is failed and the client must+ start it again. So is a failure in the stream to the remote itself. (The remote's own uploadstill retries its internal chunks.)+- A `CompleteMultipartUpload` which fails because the parts it lists+ don't match those uploaded, or there is a gap in the part numbers,+ leaves the upload open, as on S3, so the client can upload any+ missing parts and complete it again. If writing the completed object+ to the remote fails, the upload is failed and must be started again.- Parts are serialised into one stream, so ingest from the client is- effectively single-threaded, although the remote's own upload still- runs concurrently.-- On remotes that don't upload atomically (such as `local`), the- completed object is moved into place with a server-side operation.- This is a cheap rename on most such remotes. On these remotes, if- `serve s3` is killed part-way through an upload the temporary object- (named with a leading `.rclone_multipart_upload_`) may be left behind;- it is hidden from S3 listings but must be removed manually.+ effectively single-threaded. When streaming, the remote's own upload+ runs concurrently with the parts arriving; with the local disk spool+ or the VFS cache the upload to the remote only starts on completion.+- If `serve s3` is killed part-way through an upload the temporary+ object (named with a leading `.rclone_temp_multipart_`) may be left+ behind; it is hidden from S3 listings but must be removed manually.++### Multipart uploads and the VFS cache++With `--vfs-cache-mode writes` (or `full`) multipart uploads do not+stream to the remote at all. The parts are written, in part-number+order, to a temporary file in the VFS cache. On completion the file is+renamed into place and uploaded by the VFS write-back, exactly like a+plain object PUT. This needs no streaming upload support from the+remote. The rename normally happens in the cache before the upload has+started, but the VFS requires the remote to support a server-side move+or copy to rename files at all (and uses one if the temporary file has+already been written back, e.g. with `--vfs-write-back 0`). On remotes+without either, the parts are written to the cache directly under the+final key instead: the upload still never touches memory, but it loses+its atomicity - the in-flight upload is visible at the key, and an+aborted upload cannot be abandoned once in the cache, so its partial+data is written back to the remote as if it were a completed object.++Remotes that benefit from `--vfs-cache-mode writes`:++- **Remotes over slow or unreliable links.** A failure in a streamed+ upload aborts the whole multipart upload and the client must start+ again from the first part; a failed write-back upload is retried by+ the VFS (see `--vfs-cache-max-age` and friends) without the client+ being involved. Ingest from the client also runs at local disk speed+ rather than being throttled to the remote's pace.+- **Workloads that read back or overwrite what they just wrote.** The+ completed object stays in the cache, so subsequent `GET`/`HEAD`+ requests are served locally, and plain PUTs and multipart uploads to+ the same key go through the same cache entry so the last write wins+ regardless of upload style.++The trade-offs of the VFS cache:++- The whole object lands on local disk, so the cache (`--cache-dir`)+ needs space for the largest objects in flight; `--vfs-cache-max-size`+ cannot evict files which are still being uploaded.+- The `200 OK` for `CompleteMultipartUpload` means the data is safely+ in the **local cache**, not yet on the remote - the same durability+ the cache gives plain PUTs. If an acknowledgement must mean the data+ has reached the remote (for example WAL archiving), use the default+ `--vfs-cache-mode off`.+- The upload to the remote only starts on completion, rather than+ overlapping with the parts arriving, so the data reaches the remote+ later than with streaming.+- If `serve s3` is killed part-way through an upload, the temporary+ file survives in the cache and the VFS cache recovery uploads it to+ the remote on restart as a temporary object (named with a leading+ `.rclone_temp_multipart_`); as with the streaming path, it is+ hidden from S3 listings but must be removed manually.++### Cleaning up temporary objects++If `serve s3` is killed part-way through an upload it can leave a+temporary object behind, named with a leading `.rclone_temp_`. This+whole prefix is reserved: any object whose name (the last+`/`-separated segment of its key) starts with `.rclone_temp_` is+hidden from S3 listings, so don't give real objects such names - an+existing object with such a name disappears from listings (though it+stays accessible directly by its key: only listings hide reserved+names, `GET`, `HEAD` and `DELETE` of the exact key still work). A+temporary object never holds acknowledged data - uploads whose+temporary object survived were never confirmed to the client - so old+ones are safe to delete:++ rclone delete --min-age 24h --include ".rclone_temp_*" remote:path++The `--min-age` protects uploads which are still in progress: make sure+it is longer than your longest upload, especially if several `serve s3`+instances share the same remote.++rclone v1.75 named its temporary multipart objects+`.rclone_multipart_upload_*`; leftovers from an older server are also+hidden from listings and can be cleaned up the same way.++### Abandoned uploads++A client which starts a multipart upload and vanishes without either+completing or aborting it would otherwise hold on to its resources+forever.++An incomplete multipart upload which has had no activity for+`--multipart-expiry` (default `24h`) is therefore aborted and cleaned+up, exactly as if the client had called `AbortMultipartUpload`, and a+`NOTICE` is logged.++An upload with a part still being received is never expired, however+slowly the part is arriving, and each completed part restarts the+clock, so the expiry only needs to outlast the client's pauses+*between* parts, not the whole upload.++Late operations on an expired upload fail with `NoSuchUpload`, as they+do on real S3 when a lifecycle rule has aborted the upload. Set+`--multipart-expiry 0` to keep incomplete uploads forever - but note+that abandoned uploads are then never cleaned up and go on counting+towards `--multipart-max-uploads` below, so enough of them will stop+the user starting any new multipart upload until the server is+restarted.++Each multipart upload in progress holds server resources - for example+the buffers the remote's own upload uses once the first part has+arrived - so each user (each access key) can have no more than+`--multipart-max-uploads` (default `1000`, `0` for no limit) in+progress at once. Starting another fails with a `SlowDown` error,+which S3 clients retry, and a one-off `NOTICE` is logged.### Disabling streaming-If you pass `--disable-multipart-streaming`, or the remote doesn't-support `PutStream` (or doesn't upload atomically and can't move or copy-server-side), multipart uploads are instead **buffered in memory**-by the underlying S3 library: every part is held in memory and the whole-object is written out in one go when the upload completes (the previous-behaviour). This removes the in-order/contiguous-part restriction above,-so parts can be uploaded in any order, but **memory use grows with the-size of the upload**, so it is only suitable for small objects. A one-off-`NOTICE` is logged the first time this happens.--Alternatively, if the client is an rclone `s3` remote (like the-`[serves3]` example above), you can set `use_multipart_uploads = false`-on it so it uploads each object as a single stream and skips multipart-uploads altogether.+If you pass `--disable-multipart-streaming`, multipart uploads are+instead **buffered in memory** by the underlying S3 library: every+part is held in memory and the whole object is written out in one go+when the upload completes. This removes the in-order/contiguous-part+restriction above, so parts can be uploaded in any order, but **memory+use grows with the size of the upload**, so it is only suitable for+small objects. A one-off `NOTICE` is logged the first time this+happens. This flag is the only thing that makes multipart uploads+buffer in memory - it is never done because of missing remote+capabilities. The limits on multipart uploads above+(`--multipart-streaming-buffer-limit`,+`--multipart-streaming-buffer-total` and `--multipart-max-uploads`) do+not apply to uploads buffered in memory, so don't use this flag if the+server is open to clients you don't trust. Consider+`--vfs-cache-mode writes` instead, which buffers the upload in the VFS+cache on disk and takes precedence over+`--disable-multipart-streaming`.++## Memory use++`serve s3` never buffers whole objects in memory (except with+`--disable-multipart-streaming`, above), and limits the memory each+kind of request can use, so that a client, even a hostile one, can't+make it use unbounded memory.++- **Object data.** `PutObject` bodies and multipart parts arriving in+ order are streamed straight through to the remote. Parts arriving out+ of order are buffered, up to `--multipart-streaming-buffer-limit`+ per upload and `--multipart-streaming-buffer-total` per user.+- **Uploads in progress.** Each user can have at most+ `--multipart-max-uploads` multipart uploads in progress. An upload+ holds little memory until its+ first part arrives. After that, with `--vfs-cache-mode off`, it holds+ a `--streaming-upload-cutoff` sized buffer plus whatever the remote's+ own upload buffers, which for many remotes is up to their chunk size+ times their upload concurrency (for example `--s3-chunk-size` times+ `--s3-upload-concurrency`). Plain `PutObject` uploads use the same+ while they are in progress.+- **Metadata.** The metadata of at most `--metadata-max-objects`+ objects of each user is kept, each limited to 2 KB.+- **Request bodies.** The XML bodies of requests such as+ `CompleteMultipartUpload` and `DeleteObjects` are limited to 10 MiB,+ and `DeleteObjects` to 1000 keys. A browser form (`POST`) upload+ holds at most 1 MiB in memory, spooling the rest to a temporary file,+ and is limited to 5 GB.+- **Listings.** Each page of a `ListObjects` listing reads only the+ directories needed to fill it, and returns at most 1000 keys.+ Directory listings are cached by the [VFS directory+ cache](#vfs-directory-cache).++The largest of these is usually the remote's upload buffers, which can+be up to `--multipart-max-uploads` times the buffers of one upload for+each user if clients start many uploads and send part of each. The+limits apply to each user separately, so that one user can't stop+others uploading, which means the memory use grows with the number of+users. To limit it, lower+`--multipart-max-uploads`, the remote's chunk size or upload+concurrency, or set `--max-buffer-memory` to limit the memory used by+the buffers of all uploads to remotes which use rclone's buffer pool+(including `s3`, `azureblob` and `b2`). Buffered multipart parts come+from the same pool, so if you set `--max-buffer-memory` make it+comfortably bigger than `--multipart-streaming-buffer-total` times the+number of users uploading at once, otherwise+buffered parts can use up all the memory the remote needs to upload the+parts which would free them.++The number of connections is not limited, so if the server is open to+clients you don't trust, use `--auth-key` or `--auth-proxy` and+consider putting it behind a reverse proxy which limits connections.++## Design goals++These goals explain many of the limitations below.++- **Objects are plain files on the remote.** A bucket is a directory+ in the root and a key is a path in it, so the remote stays usable by+ other tools, such as `rclone mount` or `rclone sync`. So keys must+ be paths (see [Object names](#object-names)).+- **No database.** `serve s3` stores nothing but the objects. What the+ remote can't hold, such as user metadata, tags, the MD5 `ETag` of an+ object on a remote without MD5 and the directory markers put, is+ kept in memory as a best effort, and lost on restart.+- **Works with any rclone remote.** What `serve s3` can do depends on+ what the remote supports, such as the hashes used for `ETag`s and+ modification times, rather than needing a particular backend.+- **Bounded memory.** Uploads, including multipart uploads, are+ streamed to the remote and the memory each request can use is+ limited (see [Memory use](#memory-use)). Nothing is written to local+ disk unless the VFS cache is used or the remote can't stream uploads+ (see [Multipart uploads](#multipart-uploads)).+- **Refuse rather than ignore.** Unsupported features are refused, or+ answered as S3 answers when the feature isn't configured, so a+ client is never told something was done which wasn't (see+ [Compatibility](#compatibility)).+- **A storage gateway, not a full S3.** `serve s3` implements the parts+ of the S3 API clients use to store and fetch objects. Versioning,+ ACLs, bucket policies, lifecycle rules and the like are out of scope.+- **Compatible with rclone's s3 backend.** This includes its directory+ markers, which the tests check.## Bugs-Multipart server side copies do not work (see-[#7454](https://github.com/rclone/rclone/issues/7454)). These take a-very long time and eventually fail. The default threshold for-multipart server side copies is 5G which is the maximum it can be, so-files above this side will fail to be server side copied.-For a current list of `serve s3` bugs see the [serves3](https://github.com/rclone/rclone/labels/serve%20s3) bug categoryon GitHub.@@ -14427,57 +15252,124 @@ on GitHub.`serve s3` will treat all directories in the root as buckets andignore all files in the root. You can use `CreateBucket` to create-folders under the root, but you can't create empty folders under other-folders not in the root.+folders under the root, and put a directory marker to create an empty+folder below them (see [Object names](#object-names)).When using `PutObject` or `DeleteObject`, rclone will automaticallycreate or clean up empty folders. If you don't want to clean up empty-folders automatically, use `--no-cleanup`.--When using `ListObjects`, rclone will use `/` when the delimiter is-empty. This reduces backend requests with no effect on most-operations, but if the delimiter is something other than `/` and-empty, rclone will do a full recursive search of the backend, which-can take some time.--Versioning is not currently supported.--Metadata will only be saved in memory other than the rclone `mtime`-metadata which will be set as the modification time of the file.+folders automatically, use `--directory-markers` (see [Object+names](#object-names)). The deprecated `--no-cleanup` does the same.++When using `ListObjects` with the delimiter `/`, rclone lists the+directory the prefix is in, showing its subdirectories as common+prefixes, including empty ones. Without a delimiter the listing is+recursive: rclone walks the directory tree below the prefix, reading+only as many directories as it needs to fill each page, and lists+empty directories as their directory markers (see below). Any other+delimiter is supported too, but needs the whole directory tree below+the prefix walking, unless the delimiter is in the directory names.++Versioning is not supported (see [Design goals](#design-goals)).++Conditional writes, `PutObject` and `CompleteMultipartUpload` with+`If-None-Match` or `If-Match`, are supported. The conditions are+checked before the object is written, and again as it is moved into+place when it is written to a temporary object first. This happens for+multipart uploads and, for `PutObject`, with remotes which can upload+partial files (such as local disk) or with `--vfs-cache-mode minimal`+or above. The second check is atomic against other writes through+`serve s3`, but neither is atomic against changes made to the remote+some other way. If the remote doesn't support the `--etag-hash`,+objects have no `ETag` so `If-Match` with an `ETag` always fails.++Metadata will only be saved in memory (see [Design+goals](#design-goals)) other than the rclone `mtime` metadata which+will be set as the modification time of the file. The+metadata of at most `--metadata-max-objects` objects (default `100000`,+`0` for no limit) of each user (each access key) is kept; after that+the metadata of the user's least recently used object is forgotten. The metadata of an object is+ignored once the object has been changed some other way, for example+by another `--auth-proxy` user or directly on the backend. Object tags+are kept with the metadata in the same way.## Object names`serve s3` stores objects as files in the backend, so object keys aremapped to file paths rather than treated as the opaque strings AWS S3allows. Keys must be in canonical path form: keys that contain `..` or-`.` path segments, repeated slashes (`//`), or a leading or trailing-slash are rejected with a `400 Bad Request` (`InvalidArgument`)-instead of being normalised, since normalising them could alias two-distinct keys to the same file or resolve a key outside its bucket.-This matches the behaviour of other S3 servers such as MinIO.--## Supported operations--`serve s3` currently supports the following operations.--- Bucket- - `ListBuckets`- - `CreateBucket`- - `DeleteBucket`-- Object- - `HeadObject`- - `ListObjects`- - `GetObject`- - `PutObject`- - `DeleteObject`- - `DeleteObjects`- - `CreateMultipartUpload`- - `CompleteMultipartUpload`- - `AbortMultipartUpload`- - `CopyObject`- - `UploadPart`--Other operations will return error `Unimplemented`.+`.` path segments, repeated slashes (`//`), or a leading slash are+rejected with a `400 Bad Request` (`InvalidArgument`) instead of being+normalised, since normalising them could alias two distinct keys to+the same file or resolve a key outside its bucket. This matches the+behaviour of other S3 servers such as MinIO.++A key ending in `/`, such as `dir/`, is a directory marker, standing+for the directory `dir`. Putting an empty directory marker creates the+directory (one with data is rejected with `400 Bad Request`), and+deleting it removes the directory if it is empty. Directory markers+are found by `HeadObject`, `GetObject` and `CopyObject` for any+directory, but only the markers of empty directories, and of the+directories a marker was put for, are listed as objects, as other+directories may not have been made by putting a marker. The markers+put are remembered in memory with the metadata (see+[Limitations](#limitations)), so after a restart, or once forgotten,+the marker of a directory holding other keys is no longer listed.+While a marker is remembered its directory isn't removed when the+last key below it is deleted.++With `--directory-markers` every directory is listed as a directory+marker, so the markers are listed after a restart too, as needed by+clients which rely on them, such as rclone's s3 backend with+`--s3-directory-markers`. As every directory has a marker, a directory+isn't removed when the last key below it is deleted. Deleting the+marker of a directory which holds other keys stops it being listed,+and the directory is then removed with the last key below it, but this+is only remembered in memory, like the markers put without+`--directory-markers`.++As keys are file paths, a key can't be both an object and a directory+holding other objects, for example `a/b` and `a/b/c`. Writing an+object which would need this fails with `400 Bad Request`+(`InvalidRequest`).++## Compatibility++`serve s3` implements the parts of the S3 API that S3 clients commonly+use to store and fetch objects. Features it doesn't support are+refused with `501 NotImplemented`, or answered as AWS S3 answers for a+bucket without that feature configured, rather than being silently+ignored.++| Feature | Supported | Notes |+| ------- | --------- | ----- |+| `ListBuckets`, `CreateBucket`, `DeleteBucket`, `HeadBucket`, `GetBucketLocation` | Yes | Buckets are the directories in the root of the remote. |+| `ListObjects`, `ListObjectsV2` | Yes | Prefixes, delimiters, markers, `start-after`, continuation tokens, `fetch-owner` and `encoding-type=url` all work. Delimiters other than `/` need the whole directory tree walking. |+| `HeadObject`, `GetObject` | Yes | Including `Range`, `partNumber`, the `response-*` header overrides and the `If-Match`, `If-None-Match`, `If-Modified-Since` and `If-Unmodified-Since` conditions. |+| `PutObject` | Yes | Including `Content-MD5`, `aws-chunked` uploads and uploads with no `Content-Length`. |+| `CopyObject` | Yes | Including the metadata and tagging directives and the `x-amz-copy-source-if-*` conditions. |+| `DeleteObject`, `DeleteObjects` | Yes | |+| Multipart uploads | Yes | `CreateMultipartUpload`, `UploadPart`, `UploadPartCopy`, `CompleteMultipartUpload`, `AbortMultipartUpload`, `ListMultipartUploads` and `ListParts`. The `ETag` of the object is its hash, not the `-N` suffixed one AWS gives. The 5 MiB minimum part size isn't enforced. |+| Conditional writes | Partial | `If-Match` and `If-None-Match` on `PutObject` and `CompleteMultipartUpload` are checked, but not atomically against changes made to the remote some other way (see [Limitations](#limitations)). |+| Authentication | Partial | AWS Signature Version 4, in headers, presigned URLs and browser form (`POST`) uploads. Signature Version 2 isn't supported. |+| Anonymous access | Partial | All requests are allowed without `--auth-key` or `--auth-proxy`, and none are with them, as buckets and objects can't be made public. |+| Metadata | Partial | `x-amz-meta-*`, the storage class and the standard headers such as `Content-Type` are kept in memory only (see [Limitations](#limitations)), except the modification time. The storage class is only recorded, so doesn't change how the object is stored. |+| Directory markers | Partial | Keys ending in `/` are stored as directories, so can't hold data. A marker put for a directory holding other keys is only listed while it is remembered in memory, unless `--directory-markers` is set (see [Object names](#object-names)). |+| Checksums | Partial | The `x-amz-checksum-*` of a `PutObject` (CRC32, CRC32C, CRC64NVME, SHA1 or SHA256) is checked and returned with the metadata. The `Content-MD5` and `x-amz-checksum-*` of the bodies of other requests, such as `DeleteObjects`, are checked. The checksums of multipart uploads aren't supported. |+| `GetObjectAttributes` | Partial | Everything but the checksums of multipart uploads. |+| Object tagging | Partial | `GetObjectTagging`, `PutObjectTagging` and `DeleteObjectTagging` work, but tags are kept in memory with the metadata. |+| Bucket owner | No | `serve s3` has no AWS account IDs, so `x-amz-expected-bucket-owner` and `x-amz-source-expected-bucket-owner` are ignored. |+| ACLs | Partial | `GetBucketAcl` and `GetObjectAcl` give the owner full control. Only the `private` ACL can be set. |+| CORS | Partial | Preflight (`OPTIONS`) requests are answered, allowing any origin. CORS configuration (`PutBucketCors`) isn't supported. |+| Server side encryption | No | SSE-S3 (`AES256`) is accepted but the data isn't encrypted. SSE-KMS and SSE-C are refused. |+| Versioning | No | Buckets are reported as unversioned and `ListObjectVersions` lists each object as version `null`. Enabling versioning is refused. |+| Object lock, retention and legal hold | No | |+| Bucket configuration | No | Bucket policies, tagging, lifecycle, logging, encryption, website, replication, notifications and the like. Getting them fails as for a bucket without them configured. |+| `SelectObjectContent`, `RestoreObject`, torrents | No | |++The [ceph s3-tests](https://github.com/ceph/s3-tests) can be run+against `serve s3` to check its compatibility. See+[cmd/serve/s3/s3tests](https://github.com/rclone/rclone/tree/master/cmd/serve/s3/s3tests)+for how.### Authentication@@ -14507,8 +15399,20 @@ htpasswd -B htpasswd userhtpasswd -B htpasswd anotherUser```+Use the same hash format and cost for every entry in the file. If the+entries differ, the time taken to reject a wrong password can reveal+which usernames exist.+The password file can be updated while rclone is running.+When authentication is enabled every request must be authenticated,+including HTTP OPTIONS requests, as the answer to an OPTIONS request can+reveal whether a path exists. Health checks or monitoring which use+unauthenticated OPTIONS requests will need to send credentials. The+exception is browser CORS preflight requests when `--allow-origin`+is set, which are answered without authentication as browsers never send+credentials with them.+Use `--realm` to set the authentication realm.Use `--salt` to change the password hashing salt from the default.@@ -14540,6 +15444,13 @@ accept in the HTTP header.will overriding existing values. The flag may be repeated to add multipleheaders. Use the format `Header-Name: value`.+`--allow-origin` enables cross-origin (CORS) requests from browsers+by setting the `Access-Control-Allow-Origin` header on every response to the+origin given, e.g. `--allow-origin https://app.example.com`. Browser+CORS preflight OPTIONS requests are answered by rclone without needing+authentication as browsers never send credentials with them. Setting this+to `*` allows any origin which is a security risk on an authenticated server.+`--baseurl` controls the URL prefix that rclone serves from. By defaultrclone will serve from the root. If you used `--baseurl "/rclone"` thenrclone would serve from a URL starting with "/rclone/". This is@@ -14659,7 +15570,7 @@ at all times. The buffered data is bound to one open file and won't beshared.This flag is a upper limit for the used memory per open file. The-buffer will only use memory for data that is downloaded but not not+buffer will only use memory for data that is downloaded but notyet read. If the buffer is empty, only a small amount of memory willbe used.@@ -14714,7 +15625,8 @@ longest. This cache flushing strategy is efficient and more relevantfiles are likely to remain cached.The `--vfs-cache-max-age` will evict files from the cache-after the set time since last access has passed. The default value of+after the set time since last access has passed; it is based on access time,+not on when the file was first added to the cache. The default value of1 hour will start evicting files from cache that haven't been accessedfor 1 hour. When a cached file is accessed the 1 hour timer is reset to 0and will wait for 1 more hour before evicting. Specify the time with@@ -14868,7 +15780,7 @@ Rclone reads `--vfs-read-chunk-streams` chunks of size`--vfs-read-chunk-size` concurrently. The size for each read will stayconstant.-This improves performance performance massively on high latency links+This improves performance massively on high latency linksor very high bandwidth links to high performance object stores.Some experimentation will be needed to find the optimum values of@@ -15085,6 +15997,135 @@ If the file has no metadata it will be returned as `{}` and if thereis an error reading the metadata the error will be returned as`{"error":"error string"}`.+## Auth Proxy++If you supply the parameter `--auth-proxy /path/to/program` then+rclone will use that program to generate backends on the fly which+then are used to authenticate incoming requests. This uses a simple+JSON based protocol with input on STDIN and output on STDOUT.++**PLEASE NOTE:** `--auth-proxy` and `--authorized-keys` cannot be used+together, if `--auth-proxy` is set the authorized keys option will be+ignored.++There is an example program+[bin/test_proxy.py](https://github.com/rclone/rclone/blob/master/bin/test_proxy.py)+in the rclone source code.++The program's job is to take a `user` and `pass` on the input and turn+those into the config for a backend on STDOUT in JSON format. This+config will have any default parameters for the backend added, but it+won't use configuration from environment variables or command line+options - it is the job of the proxy program to make a complete+config.++This config generated must have this extra parameter++- `_root` - root to use for the backend++And it may have these parameters++- `_obscure` - comma separated strings for parameters to obscure+- `_secret_access_key` - the secret for S3 access key auth (see below)++If password authentication was used by the client, input to the proxy+process (on STDIN) would look similar to this:++```json+{+ "user": "me",+ "pass": "mypassword",+ "client_ip": "192.168.1.1"+}+```++If public-key authentication was used by the client, input to the+proxy process (on STDIN) would look similar to this:++```json+{+ "user": "me",+ "public_key": "AAAAB3NzaC1yc2EAAAADAQABAAABAQDuwESFdAe14hVS6omeyX7edc...JQdf",+ "client_ip": "192.168.1.1"+}+```++If the client authenticated with an S3 access key (`rclone serve s3`),+the client never sends its secret, only a signature made with it, so+the input contains just the access key ID as the `user` with no `pass`+or `public_key`:++```json+{+ "user": "AKIAIOSFODNN7EXAMPLE",+ "client_ip": "192.168.1.1"+}+```++In this case the program must look up the secret access key for that+access key ID and return it in the `_secret_access_key` field of the+output. Rclone then uses that secret to verify the signature on the+request, refusing the request if it does not match. This means the+proxy program is the source of truth for both the credentials and the+backend they map to. If the program does not return+`_secret_access_key` or returns it empty the request is refused.++The program's answer for an access key ID is cached (see below) but+is checked with the program again after 5 minutes even if the access+key ID is in constant use, so revoking an access key ID in the+program takes effect within 5 minutes. A rotated secret takes effect+on the first request signed with it.++The `client_ip` key holds the IP address the client connected from,+without a port number. It can be used to restrict logins to certain+networks, or to log authentication attempts centrally. It is omitted if+the client has no IP address, for example when connecting over a unix+socket. Note that if rclone is behind a reverse proxy this will be the+address of the reverse proxy and not the original client.++And as an example return this on STDOUT++```json+{+ "type": "sftp",+ "_root": "",+ "_obscure": "pass",+ "user": "me",+ "pass": "mypassword",+ "host": "sftp.example.com"+}+```++This would mean that an SFTP backend would be created on the fly for+the `user` and `pass`/`public_key` returned in the output to the host given. Note+that since `_obscure` is set to `pass`, rclone will obscure the `pass`+parameter before creating the backend (which is required for sftp+backends).++The program can manipulate the supplied `user` in any way, for example+to make proxy to many different sftp backends, you could make the+`user` be `[email protected]` and then set the `host` to `example.com`+in the output and the user to `user`. For security you'd probably want+to restrict the `host` to a limited list.++An internal cache of backends is keyed on the `user`, the kind of+credential presented, a hash of the `pass` or `public_key`, and the+`client_ip`. This means that if a user's password or public-key+changes, the client connects from a new IP address, or the proxy+returns different config parameters (eg a rotated `api_key`), a fresh+backend will be created on the next request rather than the cached one+being reused. A backend created for a public key login is never+reused for a password login, so a public key cannot be sent as a+password to skip the program.++A cached answer is checked with the program again after 5 minutes even+if the user is in constant use, so revoking a user in the program+takes effect within 5 minutes. For `serve sftp` this applies to new+logins - an existing connection is not closed.++This can be used to build general purpose proxies to any kind of+backend that rclone supports.+```rclone serve s3 remote:path [flags]```@@ -15101,7 +16142,8 @@ rclone serve s3 remote:path [flags]--client-ca string Client certificate authority to verify clients with--dir-cache-time Duration Time to cache directory entries for (default 5m0s)--dir-perms FileMode Directory permissions (default 777)- --disable-multipart-streaming Buffer multipart uploads in memory instead of streaming them to the backend (see the Multipart uploads docs section)+ --directory-markers List every directory as a directory marker object+ --disable-multipart-streaming Buffer multipart uploads in memory instead of streaming them to the backend--etag-hash string Which hash to use for the ETag, or auto or blank for off (default "MD5")--file-perms FileMode File permissions (default 666)--force-path-style If true use path style access if false use virtual hosted style (default true)@@ -15111,10 +16153,14 @@ rclone serve s3 remote:path [flags]--key string TLS PEM Private key--link-perms FileMode Link permissions (default 666)--max-header-bytes int Maximum size of request header (default 4096)+ --metadata-max-objects int Maximum number of objects of a user whose metadata is kept in memory, 0 for unlimited (default 100000)--min-tls-version string Minimum TLS version that is acceptable (default "tls1.0")- --multipart-streaming-buffer-limit SizeSuffix Maximum memory buffered per streamed multipart upload for parts arriving out of order, 0 for unlimited (see the Multipart uploads docs section) (default 256Mi)+ --multipart-expiry Duration Abort incomplete multipart uploads idle for longer than this, 0 to keep forever (default 1d)+ --multipart-max-uploads int Maximum number of streamed multipart uploads a user can have in progress at once, 0 for unlimited (default 1000)+ --multipart-streaming-buffer-limit SizeSuffix Maximum memory buffered per streamed multipart upload for parts arriving out of order, 0 for unlimited (default 256Mi)+ --multipart-streaming-buffer-total SizeSuffix Maximum memory buffered by all the streamed multipart uploads of a user for parts arriving out of order, 0 for unlimited (default 1Gi)--no-checksum Don't compare checksums on up/download- --no-cleanup Not to cleanup empty folder after object is deleted+ --no-cleanup Deprecated: use --directory-markers instead--no-modtime Don't read/write the modification time (can speed things up)--no-seek Don't allow seeking in files--pass string Password for authentication@@ -15332,7 +16378,7 @@ at all times. The buffered data is bound to one open file and won't beshared.This flag is a upper limit for the used memory per open file. The-buffer will only use memory for data that is downloaded but not not+buffer will only use memory for data that is downloaded but notyet read. If the buffer is empty, only a small amount of memory willbe used.@@ -15387,7 +16433,8 @@ longest. This cache flushing strategy is efficient and more relevantfiles are likely to remain cached.The `--vfs-cache-max-age` will evict files from the cache-after the set time since last access has passed. The default value of+after the set time since last access has passed; it is based on access time,+not on when the file was first added to the cache. The default value of1 hour will start evicting files from cache that haven't been accessedfor 1 hour. When a cached file is accessed the 1 hour timer is reset to 0and will wait for 1 more hour before evicting. Specify the time with@@ -15541,7 +16588,7 @@ Rclone reads `--vfs-read-chunk-streams` chunks of size`--vfs-read-chunk-size` concurrently. The size for each read will stayconstant.-This improves performance performance massively on high latency links+This improves performance massively on high latency linksor very high bandwidth links to high performance object stores.Some experimentation will be needed to find the optimum values of@@ -15784,9 +16831,10 @@ This config generated must have this extra parameter- `_root` - root to use for the backend-And it may have this parameter+And it may have these parameters- `_obscure` - comma separated strings for parameters to obscure+- `_secret_access_key` - the secret for S3 access key auth (see below)If password authentication was used by the client, input to the proxyprocess (on STDIN) would look similar to this:@@ -15794,7 +16842,8 @@ process (on STDIN) would look similar to this:```json{"user": "me",- "pass": "mypassword"+ "pass": "mypassword",+ "client_ip": "192.168.1.1"}```@@ -15804,10 +16853,44 @@ proxy process (on STDIN) would look similar to this:```json{"user": "me",- "public_key": "AAAAB3NzaC1yc2EAAAADAQABAAABAQDuwESFdAe14hVS6omeyX7edc...JQdf"+ "public_key": "AAAAB3NzaC1yc2EAAAADAQABAAABAQDuwESFdAe14hVS6omeyX7edc...JQdf",+ "client_ip": "192.168.1.1"}```+If the client authenticated with an S3 access key (`rclone serve s3`),+the client never sends its secret, only a signature made with it, so+the input contains just the access key ID as the `user` with no `pass`+or `public_key`:++```json+{+ "user": "AKIAIOSFODNN7EXAMPLE",+ "client_ip": "192.168.1.1"+}+```++In this case the program must look up the secret access key for that+access key ID and return it in the `_secret_access_key` field of the+output. Rclone then uses that secret to verify the signature on the+request, refusing the request if it does not match. This means the+proxy program is the source of truth for both the credentials and the+backend they map to. If the program does not return+`_secret_access_key` or returns it empty the request is refused.++The program's answer for an access key ID is cached (see below) but+is checked with the program again after 5 minutes even if the access+key ID is in constant use, so revoking an access key ID in the+program takes effect within 5 minutes. A rotated secret takes effect+on the first request signed with it.++The `client_ip` key holds the IP address the client connected from,+without a port number. It can be used to restrict logins to certain+networks, or to log authentication attempts centrally. It is omitted if+the client has no IP address, for example when connecting over a unix+socket. Note that if rclone is behind a reverse proxy this will be the+address of the reverse proxy and not the original client.+And as an example return this on STDOUT```json@@ -15833,11 +16916,20 @@ to make proxy to many different sftp backends, you could make thein the output and the user to `user`. For security you'd probably wantto restrict the `host` to a limited list.-An internal cache of backends is keyed on the `user` and a hash of the-`pass` or `public_key`. This means that if a user's password or-public-key changes, or the proxy returns different config parameters-(eg a rotated `api_key`), a fresh backend will be created on the next-request rather than the cached one being reused.+An internal cache of backends is keyed on the `user`, the kind of+credential presented, a hash of the `pass` or `public_key`, and the+`client_ip`. This means that if a user's password or public-key+changes, the client connects from a new IP address, or the proxy+returns different config parameters (eg a rotated `api_key`), a fresh+backend will be created on the next request rather than the cached one+being reused. A backend created for a public key login is never+reused for a password login, so a public key cannot be sent as a+password to skip the program.++A cached answer is checked with the program again after 5 minutes even+if the user is in constant use, so revoking a user in the program+takes effect within 5 minutes. For `serve sftp` this applies to new+logins - an existing connection is not closed.This can be used to build general purpose proxies to any kind ofbackend that rclone supports.@@ -16059,6 +17151,13 @@ accept in the HTTP header.will overriding existing values. The flag may be repeated to add multipleheaders. Use the format `Header-Name: value`.+`--allow-origin` enables cross-origin (CORS) requests from browsers+by setting the `Access-Control-Allow-Origin` header on every response to the+origin given, e.g. `--allow-origin https://app.example.com`. Browser+CORS preflight OPTIONS requests are answered by rclone without needing+authentication as browsers never send credentials with them. Setting this+to `*` allows any origin which is a security risk on an authenticated server.+`--baseurl` controls the URL prefix that rclone serves from. By defaultrclone will serve from the root. If you used `--baseurl "/rclone"` thenrclone would serve from a URL starting with "/rclone/". This is@@ -16128,6 +17227,13 @@ to be used within the template to server pages:| .Sort | | The current sort used. This is changeable via '?sort=' parameter. Possible values: namedirfirst, name, size, time (default namedirfirst). || .Order | | The current ordering used. This is changeable via '?order=' parameter. Possible values: asc, desc (default asc). || .Query | | Currently unused. |+| .Path | | The path of the directory from the root with leading and trailing slashes, e.g. '/dir/'. |+| .IsRoot | | Boolean for if this is the root directory. |+| .UpLink | | The link to the parent directory. |+| .Static | | Boolean for if the listing is being written as a static page by rclone index. |+| .NumDirs | | The number of directories in the listing. |+| .NumFiles | | The number of files in the listing. |+| .TotalSize | | The total size in bytes of the files in the listing. || .Breadcrumb | | Allows for creating a relative navigation. || | .Link | The link of the Text relative to the root. || | .Text | The Name of the directory. |@@ -16137,6 +17243,7 @@ to be used within the template to server pages:| | .IsDir | Boolean for if an entry is a directory or not. || | .Size | Size in bytes of the entry. || | .ModTime | The UTC timestamp of an entry. |+| | .MimeType | The MIME type of an entry guessed from its name, or 'inode/directory'. |The server also makes the following functions available so that they can be usedwithin the template. These functions help extend the options for dynamic@@ -16177,8 +17284,20 @@ htpasswd -B htpasswd userhtpasswd -B htpasswd anotherUser```+Use the same hash format and cost for every entry in the file. If the+entries differ, the time taken to reject a wrong password can reveal+which usernames exist.+The password file can be updated while rclone is running.+When authentication is enabled every request must be authenticated,+including HTTP OPTIONS requests, as the answer to an OPTIONS request can+reveal whether a path exists. Health checks or monitoring which use+unauthenticated OPTIONS requests will need to send credentials. The+exception is browser CORS preflight requests when `--allow-origin`+is set, which are answered without authentication as browsers never send+credentials with them.+Use `--realm` to set the authentication realm.Use `--salt` to change the password hashing salt from the default.@@ -16246,7 +17365,7 @@ at all times. The buffered data is bound to one open file and won't beshared.This flag is a upper limit for the used memory per open file. The-buffer will only use memory for data that is downloaded but not not+buffer will only use memory for data that is downloaded but notyet read. If the buffer is empty, only a small amount of memory willbe used.@@ -16301,7 +17420,8 @@ longest. This cache flushing strategy is efficient and more relevantfiles are likely to remain cached.The `--vfs-cache-max-age` will evict files from the cache-after the set time since last access has passed. The default value of+after the set time since last access has passed; it is based on access time,+not on when the file was first added to the cache. The default value of1 hour will start evicting files from cache that haven't been accessedfor 1 hour. When a cached file is accessed the 1 hour timer is reset to 0and will wait for 1 more hour before evicting. Specify the time with@@ -16455,7 +17575,7 @@ Rclone reads `--vfs-read-chunk-streams` chunks of size`--vfs-read-chunk-size` concurrently. The size for each read will stayconstant.-This improves performance performance massively on high latency links+This improves performance massively on high latency linksor very high bandwidth links to high performance object stores.Some experimentation will be needed to find the optimum values of@@ -16698,9 +17818,10 @@ This config generated must have this extra parameter- `_root` - root to use for the backend-And it may have this parameter+And it may have these parameters- `_obscure` - comma separated strings for parameters to obscure+- `_secret_access_key` - the secret for S3 access key auth (see below)If password authentication was used by the client, input to the proxyprocess (on STDIN) would look similar to this:@@ -16708,7 +17829,8 @@ process (on STDIN) would look similar to this:```json{"user": "me",- "pass": "mypassword"+ "pass": "mypassword",+ "client_ip": "192.168.1.1"}```@@ -16718,10 +17840,44 @@ proxy process (on STDIN) would look similar to this:```json{"user": "me",- "public_key": "AAAAB3NzaC1yc2EAAAADAQABAAABAQDuwESFdAe14hVS6omeyX7edc...JQdf"+ "public_key": "AAAAB3NzaC1yc2EAAAADAQABAAABAQDuwESFdAe14hVS6omeyX7edc...JQdf",+ "client_ip": "192.168.1.1"}```+If the client authenticated with an S3 access key (`rclone serve s3`),+the client never sends its secret, only a signature made with it, so+the input contains just the access key ID as the `user` with no `pass`+or `public_key`:++```json+{+ "user": "AKIAIOSFODNN7EXAMPLE",+ "client_ip": "192.168.1.1"+}+```++In this case the program must look up the secret access key for that+access key ID and return it in the `_secret_access_key` field of the+output. Rclone then uses that secret to verify the signature on the+request, refusing the request if it does not match. This means the+proxy program is the source of truth for both the credentials and the+backend they map to. If the program does not return+`_secret_access_key` or returns it empty the request is refused.++The program's answer for an access key ID is cached (see below) but+is checked with the program again after 5 minutes even if the access+key ID is in constant use, so revoking an access key ID in the+program takes effect within 5 minutes. A rotated secret takes effect+on the first request signed with it.++The `client_ip` key holds the IP address the client connected from,+without a port number. It can be used to restrict logins to certain+networks, or to log authentication attempts centrally. It is omitted if+the client has no IP address, for example when connecting over a unix+socket. Note that if rclone is behind a reverse proxy this will be the+address of the reverse proxy and not the original client.+And as an example return this on STDOUT```json@@ -16747,11 +17903,20 @@ to make proxy to many different sftp backends, you could make thein the output and the user to `user`. For security you'd probably wantto restrict the `host` to a limited list.-An internal cache of backends is keyed on the `user` and a hash of the-`pass` or `public_key`. This means that if a user's password or-public-key changes, or the proxy returns different config parameters-(eg a rotated `api_key`), a fresh backend will be created on the next-request rather than the cached one being reused.+An internal cache of backends is keyed on the `user`, the kind of+credential presented, a hash of the `pass` or `public_key`, and the+`client_ip`. This means that if a user's password or public-key+changes, the client connects from a new IP address, or the proxy+returns different config parameters (eg a rotated `api_key`), a fresh+backend will be created on the next request rather than the cached one+being reused. A backend created for a public key login is never+reused for a password login, so a public key cannot be sent as a+password to skip the program.++A cached answer is checked with the program again after 5 minutes even+if the user is in constant use, so revoking a user in the program+takes effect within 5 minutes. For `serve sftp` this applies to new+logins - an existing connection is not closed.This can be used to build general purpose proxies to any kind ofbackend that rclone supports.@@ -17180,7 +18345,7 @@ This command runs a series of uploads and downloads to the remote, measuringand printing the speed of each test using varying file sizes and numbers offiles.-Test time can be innaccurate with small file caps and large files. As it+Test time can be inaccurate with small file caps and large files. As ituses the results of an initial test to determine how many files to use ineach subsequent test.@@ -17247,9 +18412,11 @@ This will touch `--transfers` files concurrently.If `--timestamp` is used then sets the modification time to thattime instead of the current time. Times may be specified as one of:-- 'YYMMDD' - e.g. 17.10.30+- 'YYMMDD' - e.g. 171030+- 'YYYY-MM-DD' - e.g. 2006-01-02- 'YYYY-MM-DDTHH:MM:SS' - e.g. 2006-01-02T15:04:05- 'YYYY-MM-DDTHH:MM:SS.SSS' - e.g. 2006-01-02T15:04:05.123456789+- 'YYYY-MM-DDTHH:MM:SSZ' - e.g. 2006-01-02T15:04:05ZNote that value of `--timestamp` is in UTC. If you want local timethen add the `--localtime` flag.@@ -17707,6 +18874,11 @@ possible to write in all of them. This is mostly a problem on Windows, wherethe console traditionally uses a non-Unicode character set - definedby the so-called "code page".+If you want to configure a remote using environment variables then its name+becomes part of the environment variable name, which in practice restricts the+characters you can use in it. See the [Config file](#config-file) section+under Environment variables for details.+Do not use single character names on Windows as it creates ambiguity with Windowsdrives' names, e.g.: remote called `C` is indistinguishable from `C` drive. Rclonewill always assume that single letter name refers to a drive.@@ -17753,6 +18925,7 @@ Not all global variables make sense to be overridden like this as the config isonly applied during the remote creation. Here is a non exhaustive list of oneswhich might be useful:+- `allow_https_downgrade`- `bind_addr`- `ca_cert`- `client_cert`@@ -18043,6 +19216,135 @@ modification time of the source object.Hashes are not included in system metadata as there is a well definedway of reading those already.+## Performance {#performance}++Rclone's defaults are fairly conservative. They are chosen to work on+small machines, slow links and with cloud providers which rate limit+heavily. If you have plenty of bandwidth, memory and a fast storage+system then rclone can usually be made to run a great deal faster by+adjusting a few flags.++The right values depend on the size of your files, your network and+the storage systems at each end, so treat the numbers below as+starting points. Change one thing at a time and measure the result.++### Measuring++Use `rclone copy` with [-P/--progress](#p-progress) to watch the+transfer rate while you are experimenting. For unattended runs log the+stats instead, e.g. `--stats 1m --stats-one-line --stats-log-level+NOTICE`.++A transfer is always limited by something: the network bandwidth, the+rate at which the storage system will accept requests, disk IO, CPU or+memory. Tuning is working out which limit you are hitting and raising+it until you hit the next one. Watch the network, disk, CPU and memory+use of the machine rclone is running on as well as rclone's stats.++### Lots of files++The time taken to transfer small files is dominated by the round trip+time of the requests rather than by the bandwidth, so the way to make+them go faster is to do more of them at once.++[--transfers](#transfers-int) sets the number of files transferred in+parallel. The default is 4. Double this until the transfer rate stops+increasing or the storage system starts returning errors. Values of 32+to 64 are common and values in the hundreds work well with the large+object storage systems (e.g. s3, azureblob, googlecloudstorage, b2)+when transferring lots of small files.++[--checkers](#checkers-int) sets the number of directory listings and+file comparisons done in parallel. The default is 8. Raising this+speeds up syncs where most of the files are already present on the+destination and stops the checking being the bottleneck when+`--transfers` is large. 16 to 64 is a reasonable range.++Not all storage systems like lots of parallel requests. If you see rate+limiting errors (e.g. HTTP 429 or 503 errors) or lots of low level+retries in the log then you have gone too far, so reduce `--transfers`+and `--checkers` or use [--tpslimit](#tpslimit-float) to cap the+request rate. Local disks, especially spinning ones, and network file+systems may get slower rather than faster with too much parallelism.++### Large files++A single stream often won't fill a fast link, so rclone can transfer a+large file as several chunks in parallel. These are called multi-thread+transfers and are used for files bigger than+[--multi-thread-cutoff](#multi-thread-cutoff) on backends which+support them.++The number of chunks transferred in parallel is set by+[--multi-thread-streams](#multi-thread-streams-int) (default 4) and+this works for all backends so increase this first to make large files+transfer faster.++The size of the chunks is set by the backend, usually with a+`--backend-chunk-size` flag, e.g. `--s3-chunk-size`. Backends without+a natural chunk size (e.g. local, smb) use+[--multi-thread-chunk-size](#multi-thread-chunk-size-sizesuffix).+Larger chunks mean fewer requests which increases throughput on big+files. See the docs for the backend you are using for details.++Chunked uploads are usually buffered in memory, so in the worst case+rclone will use++```text+--transfers * upload concurrency * chunk size+```++of memory for buffers. For example `--transfers 256` with an upload+concurrency of 8 and a chunk size of 16 MiB could use up to 32 GiB.+When you raise `--transfers` to speed up small files you may therefore+need to lower the upload concurrency, or set+[--max-buffer-memory](#max-buffer-memory) which limits the total+memory used for buffers and makes transfers wait when the limit is+reached.++[--use-mmap](#use-mmap) makes rclone return buffer memory to the OS as+soon as it has finished with it which is helpful when using lots of+transfers or large chunks.++### A mixture of large and small files++By default rclone transfers files in roughly the order it finds them.+With a mixture of file sizes this means that at times all the+transfers are large files, which fill the bandwidth but make few+requests, and at other times all the transfers are small files, which+make lots of requests but use little of the bandwidth.++Using [--order-by](#order-by-string) `size,mixed,50` makes half of the+transfers take the smallest files in the backlog and half of them take+the largest. This keeps both large and small files moving at once so+the bandwidth and the request rate are both kept full for the whole of+the run.++This works best if rclone has lots of files to choose from so use it+with a big [--max-backlog](#max-backlog-int), e.g. `--max-backlog+2000000`. Each item in the backlog uses in the order of 1 KiB of memory.++### Doing less work++Fewer requests to the storage system means a faster transfer.++- If the source and destination are on the same remote then rclone+ will use [server-side copy](#server-side-copy) where it is supported+ so the data doesn't need to be downloaded and uploaded again.+- [--fast-list](#fast-list) lists bucket-based remotes with fewer+ transactions at the cost of using more memory. This may be faster or+ slower than the default so it is worth testing.+- [--no-traverse](#no-traverse) stops rclone listing the destination+ which helps when copying a small number of files into a large+ destination.+- [--size-only](#size-only) or [--checksum](#c-checksum) avoid reading+ the modification time. On some object storage systems (e.g. s3,+ swift) this needs an extra request for each file. Alternatively use+ [--use-server-modtime](#use-server-modtime) with `--update`.+- [--no-check-dest](#no-check-dest) skips checking the destination+ entirely. Only use this if you know that none of the files are+ there already.+## OptionsRclone has a number of options to control its behaviour. These are@@ -18110,7 +19412,7 @@ as a number of seconds. The following suffixes are valid:- `M` - Months- `y` - Years-Examples: "10", "300ms", "-1.5h" or "2h45m".+Examples: "10", "300ms", "-1.5h", "2h45m" or "1d12h".### Size options@@ -18125,6 +19427,21 @@ See also [--human-readable](#human-readable).## Main options+### --allow-https-downgrade++By default rclone refuses to follow a redirect from an `https://` URL+to a plain `http://` URL, as any credentials or other headers in the+request would then be sent unencrypted, and the response could be+altered by anyone on the network path. The request fails with an error+which mentions this flag.++Set `--allow-https-downgrade` to follow such redirects, for example+when downloading from a mirror which redirects to plain HTTP. Some+backends still refuse to follow them, as doing so would send their+credentials in cleartext.++This option defaults to `false`.+### --backup-dir stringWhen using [sync](https://rclone.org/commands/rclone_sync/), [copy](https://rclone.org/commands/rclone_copy/) or@@ -18256,7 +19573,7 @@ example:Is equivalent to this:```text---bwlimit "Mon-00:00,512Mon-12:00,1M Tue-12:00,1M Wed-12:00,1M Thu-12:00,1M Fri-12:00,1M Sat-12:00,1M Sun-12:00,1M Sun-20:00,off"+--bwlimit "Mon-00:00,512 Mon-12:00,1M Tue-12:00,1M Wed-12:00,1M Thu-12:00,1M Fri-12:00,1M Sat-12:00,1M Sun-12:00,1M Sun-20:00,off"```Bandwidth limit apply to the data transfer for all backends. For most@@ -18309,7 +19626,8 @@ will use this much memory for buffering.When using `mount` or `cmount` each open file descriptor will use this muchmemory for buffering.-See the [mount](https://rclone.org/commands/rclone_mount/#file-buffering) documentation for more details.+See the [mount](https://rclone.org/commands/rclone_mount/#vfs-file-buffering) documentation for+more details.Set to `0` to disable the buffering for the minimum memory usage.@@ -18568,7 +19886,7 @@ for more information as to what these options mean.### --default-time Time-If a file or directory does have a modification time rclone can read+If a file or directory does not have a modification time rclone can readthen rclone will display this fixed time instead.The default is `2000-01-01 00:00:00 UTC`. This can be configured in@@ -18729,6 +20047,12 @@ This flag is supported for all HTTP based backends even those notsupported by `--header-upload` and `--header-download` so may be usedas a workaround for those with care.+The headers are only sent to the host of the original request. If the+server redirects to another host (including a subdomain or a different+port), or redirects an `https` request to `http`, the headers are not+sent to it, or to any further hop in that redirect chain. Rclone logs+this at debug level (`-vv`).+```consolerclone ls remote:test --header "X-Rclone: Foo" --header "X-LetMeIn: Yes"```@@ -18915,6 +20239,12 @@ Note that on the local file system if you don't use `--inplace` hardlinks (Unix only) will be broken. And if you do use `--inplace` youwon't be able to update in use executables.+On macOS, if rclone clones a file (see+[--local-no-clone](https://rclone.org/local/#local-no-clone)) and the destination+already exists, the existing file is replaced with a new one, even+with `--inplace`, so hard links to it will be broken. Use+`--local-no-clone` to keep them.+Note also that versions of rclone prior to v1.63.0 behave as if the`--inplace` flag is always supplied.@@ -18967,7 +20297,7 @@ like symlinks under Windows). Ignored files won't be copied, moved ordeleted in a sync.If you supply this flag then rclone will copy symbolic links from any-supported backend backend, and store them as text files, with a+supported backend, and store them as text files, with a`.rclonelink` suffix in the destination.The text file will contain the target of the symbolic link.@@ -19499,7 +20829,7 @@ rclone will use multiple threads to transfer the file (default 256M).Capable backends are marked in the[overview](https://rclone.org/overview/#optional-features) as `MultithreadUpload`. (Theyneed to implement either the `OpenWriterAt` or `OpenChunkWriter`-internal interfaces). These include include, `local`, `s3`,+internal interfaces). These include `local`, `s3`,`azureblob`, `b2`, `oracleobjectstorage` and `smb` at the time ofwriting.@@ -19526,7 +20856,7 @@ Most multi-thread transfers do not take additional memory, but some doat maximum `--transfers` \* `--multi-thread-chunk-size` \*`--multi-thread-streams` or specifically for the s3 backend`--transfers` \* `--s3-chunk-size` \* `--s3-concurrency`. However you-can use the the [--max-buffer-memory](https://rclone.org/docs/#max-buffer-memory) flag+can use the [--max-buffer-memory](https://rclone.org/docs/#max-buffer-memory) flagto control the maximum memory used here.**NB** that this **only** works with supported backends as the@@ -19644,8 +20974,10 @@ describes what aspect is being measured:- `size` - order by the size of the files- `name` - order by the full path of the files- `modtime` - order by the modification date of the files+- `pattern` - order by a list of path glob patterns-This can have a modifier appended with a comma:+The `size`, `name` and `modtime` comparisons can have a modifier appended with+a comma:- `ascending` or `asc` - order so that the smallest (or oldest) is processed first- `descending` or `desc` - order so that the largest (or newest) is processed first@@ -19664,11 +20996,18 @@ processed continuously.If no modifier is supplied then the order is `ascending`.+For `pattern`, append one or more comma-separated patterns. Patterns use the+[filter pattern syntax](https://rclone.org/filtering/#patterns). Files matching the first pattern+are processed first, followed by files matching the second pattern, and so on.+Files which do not match a pattern are processed last. Within each group, files+are ordered by full path. Patterns cannot contain a comma.+For example- `--order-by size,desc` - send the largest files first- `--order-by modtime,ascending` - send the oldest files first- `--order-by name` - send the files with alphabetically by path first+- `--order-by 'pattern,*.tar.gz,*.md5'` - send archives before checksum filesIf the `--order-by` flag is not supplied or it is supplied with anempty string then the default ordering will be used which is as@@ -19710,8 +21049,9 @@ setting the config password for the first time.The argument to this should be a command with a space separated listof arguments. If one of the arguments has a space in then enclose itin `"`, if you want a literal `"` in an argument then enclose the-argument in `"` and double the `"`. See [CSV encoding](https://godoc.org/encoding/csv)-for more info.+argument in `"` and double the `"`. This includes the command itself:+if the path to the executable contains a space, it must be quoted too.+See [CSV encoding](https://godoc.org/encoding/csv) for more info.Eg@@ -19719,6 +21059,7 @@ Eg--password-command "echo hello"--password-command 'echo "hello with space"'--password-command 'echo "hello with ""quotes"" and space"'+--password-command '"/path with a space/get-password.sh"'```Note that when changing the configuration password the environment@@ -20752,9 +22093,11 @@ for each backend.To find the name of the environment variable, you need to set, take`RCLONE_CONFIG_` + name of remote + `_` + name of config file optionand make it all uppercase.-Note one implication here is the remote's name must be-convertible into a valid environment variable name,-so it can only contain letters, digits, or the `_` (underscore) character.+Note that the name of the remote is only converted to uppercase - characters+such as `-` and `.` are not changed to `_` - so a remote called `my-remote`+needs `RCLONE_CONFIG_MY-REMOTE_TYPE`. Many shells only allow letters, digits+and `_` (underscore) in environment variable names, so it is best to use only+those characters in the name of a remote you want to configure this way.For example, to configure an S3 remote named `mys3:` without a configfile (using unix ways of setting environment variables):@@ -21009,7 +22352,7 @@ Rclone matching rules follow a glob style:* matches any sequence of non-separator (/) characters** matches any sequence of characters including / separators? matches any single non-separator (/) character-[ [ ! ] { character-range } ]+[ [ ! | ^ ] { character-range } ]character class (must be non-empty){ pattern-list }pattern alternatives@@ -22049,6 +23392,10 @@ When you run `rclone gui` this is what happens- The URL already contains the username & password, in which case theGUI will use those values and log you in automatically.+The rc address is passed to the GUI via the `url` query parameter (the+URL printed in the terminal); append `?url=http://...` to point it at a+different rc server.+## SecurityIt's important to think first about what rclone has access to and what@@ -22132,7 +23479,7 @@ Consequently:reason to change it.- **Do not use `--rc-no-auth` on a non-loopback bind.** It disablesauthentication on the endpoints that access remotes — see- [`--rc-no-auth`](#--rc-no-auth).+ [`--rc-no-auth`](#rc-no-auth).- **Use authentication and TLS** (`--rc-user`/`--rc-pass` or `--rc-htpasswd`,plus `--rc-cert`/`--rc-key`) whenever the port is reachable by anyone you donot fully trust, and raise `--rc-min-tls-version`.@@ -22259,9 +23606,20 @@ Default Off.Set the allowed Access-Control-Allow-Origin for rc requests.-Can be used with --rc-web-gui if the rclone is running on different IP than the web-gui.+This enables cross-origin (CORS) requests from browsers on the origin+given, for example `--rc-allow-origin https://app.example.com`. Browser+CORS preflight OPTIONS requests are answered by rclone without+authentication as browsers never send credentials with them. All other+requests, including other OPTIONS requests, must be authenticated when+authentication is enabled.-Default is IP address on which rc is running.+The `rclone gui` command sets this automatically to the origin the GUI+is served from if it is not set explicitly.++Setting this to `*` allows any origin which is a security risk on an+authenticated server.++Default is off (no CORS).### --rc-web-fetch-url@@ -22306,6 +23664,9 @@ If this is set then no authorisation will be required on the server touse these methods. The alternative is to use `--rc-user` and`--rc-pass` and use these credentials in the request.+Without authorisation or this flag, POST request bodies are limited to+64 KiB.+Default Off.### --rc-baseurl@@ -22386,6 +23747,30 @@ Rather thanrclone rc operations/list --json '{"fs": "/tmp", "remote": "test", "opt": {"showHash": true}}'```+### Paths on Windows++Prefer forward slashes (`/`) in Windows paths passed to the rc API, for+example `C:/path1/path2`. If you use backslashes (`\`) in a JSON string,+escape each one as `\\`. These JSON objects contain equivalent Windows+paths:++```json+{"remote": "C:/path1/path2"}+```++```json+{"remote": "C:\\path1\\path2"}+```++This applies to JSON request bodies and to JSON-encoded object parameters,+such as `parameters` in `config/create`, even when that JSON is sent in a+query string or form data. An unescaped backslash can cause a JSON parsing+error or be interpreted as an escape sequence, such as `\t` for a tab.++String parameters sent directly in a query string or form data, without+JSON encoding, do not need JSON escaping. Use your HTTP client's URL+encoding for query strings and form data.+## Special parametersThe rc interface supports some special parameters which apply to@@ -22554,6 +23939,16 @@ Or flat:If you wish to check the `_config` assignment has worked properly thencalling `options/local` will show what the value got set to.+Setting `LogLevel` sets the log level for the logs made by the rc call+only, for example to debug a single call.++```json+"_config":{"LogLevel": "DEBUG"}+```++Note that not all of rclone's logs support this yet. In particular the+backends log at the global log level.+### Setting filter flags with _filterIf you wish to set filters for the duration of an rc call only then@@ -23262,6 +24657,7 @@ Returns the following values:"totalTransfers": total number of transfers in the group,"transferTime" : total time spent on running jobs,"transfers": number of transferred files,+ "updatedDirs": number of directories updated (modtime or metadata set),"transferring": an array of currently active file transfers:[{@@ -23527,8 +24923,6 @@ Results:- runningIds - array of integer job ids that are running- finishedIds - array of integer job ids that are finished-**Authentication is not required for this call.**-### job/status: Reads the status of the job ID {#job-status}Parameters:@@ -23549,8 +24943,6 @@ Results:- output - output of the job as would have been returned if called synchronously- progress - output of the progress related to the underlying job-**Authentication is not required for this call.**-### job/stop: Stop the running job {#job-stop}Parameters:@@ -23880,6 +25272,25 @@ This command does not have a command line equivalent so use this instead:rclone rc --loopback operations/fsinfo fs=remote:+### operations/getfile: Get a single file into a JSON response. {#operations-getfile}++This takes the following parameters:++- fs - a remote name string e.g. "drive:path/to/file" or "drive:path/to/dir"+- remote - (optional) a path within that remote e.g. "file.txt"+- offset - (optional) start reading at offset N (or from end if negative) (default 0)+- count - (optional) only read N bytes (default -1)+- head - (optional) only read the first N bytes (default 0)+- tail - (optional) only read the last N bytes (default 0)+- maxSize - (optional) maximum size limit in bytes (can only lower the built-in maximum) (default 2097152 / 2MiB)+- base64 - (optional) if true, returns the file contents base64-encoded (default false)++Returns:++- result - contents of the file as string or base64-encoded string++For streaming file content over HTTP with Range support, use the --rc-serve flag instead.+### operations/hashsum: Produces a hashsum file for all the objects in the path. {#operations-hashsum}Produces a hash file for all the objects in the path using the hash@@ -23954,6 +25365,26 @@ Example:See the [hashsum](https://rclone.org/commands/rclone_hashsum/) command for more information on the above.+### operations/index: Write static directory listings into a remote {#operations-index}++This takes the following parameters:++- fs - a remote name string e.g. "drive:"+- opt - a dictionary of options to control the indexing (optional)+ - output - list of listings to write in each directory as "NAME=FORMAT" (default ["index.html=html"])+ - rules - which directories get listings, e.g. {"ExcludeRule": ["/private/**"]}+ - maxDepth - only make listings this many directories deep (default -1 for no limit)+ - linkIndex - link to "dir/NAME" rather than "dir/"+ - dirTime - how to work out directory times: "newest", "dir" or "none"+ - noModTime - don't show modification times+ - rewrite - write every listing even if it is unchanged+ - changed - list of changed paths for a partial run, directories ending in "/"+ - changedFrom - list of files of changed paths, one per line+ - changedCombined - list of files in the sync --combined report format+ - changedMaxDirs - do a full run if a partial one would list more directories than this++See the [index](https://rclone.org/commands/rclone_index/) command for more information on the above.+### operations/list: List the given remote and path in JSON format {#operations-list}This takes the following parameters:@@ -24173,85 +25604,6 @@ And this sets NOTICE level logs (normal without -v)rclone rc options/set --json '{"main": {"LogLevel": "NOTICE"}}'-### pluginsctl/addPlugin: Add a plugin using url {#pluginsctl-addPlugin}--Used for adding a plugin to the webgui.--This takes the following parameters:--- url - http url of the github repo where the plugin is hosted (http://github.com/rclone/rclone-webui-react).--Example:-- rclone rc pluginsctl/addPlugin--### pluginsctl/getPluginsForType: Get plugins with type criteria {#pluginsctl-getPluginsForType}--This shows all possible plugins by a mime type.--This takes the following parameters:--- type - supported mime type by a loaded plugin e.g. (video/mp4, audio/mp3).-- pluginType - filter plugins based on their type e.g. (DASHBOARD, FILE_HANDLER, TERMINAL).--Returns:--- loadedPlugins - list of current production plugins.-- testPlugins - list of temporarily loaded development plugins, usually running on a different server.--Example:-- rclone rc pluginsctl/getPluginsForType type=video/mp4--### pluginsctl/listPlugins: Get the list of currently loaded plugins {#pluginsctl-listPlugins}--This allows you to get the currently enabled plugins and their details.--This takes no parameters and returns:--- loadedPlugins - list of current production plugins.-- testPlugins - list of temporarily loaded development plugins, usually running on a different server.--E.g.-- rclone rc pluginsctl/listPlugins--### pluginsctl/listTestPlugins: Show currently loaded test plugins {#pluginsctl-listTestPlugins}--Allows listing of test plugins with the rclone.test set to true in package.json of the plugin.--This takes no parameters and returns:--- loadedTestPlugins - list of currently available test plugins.--E.g.-- rclone rc pluginsctl/listTestPlugins--### pluginsctl/removePlugin: Remove a loaded plugin {#pluginsctl-removePlugin}--This allows you to remove a plugin using it's name.--This takes parameters:--- name - name of the plugin in the format `author`/`plugin_name`.--E.g.-- rclone rc pluginsctl/removePlugin name=rclone/video-plugin--### pluginsctl/removeTestPlugin: Remove a test plugin {#pluginsctl-removeTestPlugin}--This allows you to remove a plugin using it's name.--This takes the following parameters:--- name - name of the plugin in the format `author`/`plugin_name`.--Example:-- rclone rc pluginsctl/removeTestPlugin name=rclone/rclone-webui-react-### rc/error: This returns an error {#rc-error}This returns an error with the input as part of its error string.@@ -24426,6 +25778,8 @@ This takes the following parameters:- path1 (required) - (string) a remote directory string e.g. `drive:path1`- path2 (required) - (string) a remote directory string e.g. `drive:path2`- dryRun - (bool) dry-run mode+- maxDelete - (int) Safety check on maximum percentage of deleted files+allowed. If exceeded, the bisync run will abort. (default: 50%)- backupDir1 - (string) --backup-dir for Path1. Must be a non-overlapping path onthe same remote.- backupDir2 - (string) --backup-dir for Path2. Must be a non-overlapping path on@@ -24455,6 +25809,8 @@ unavailable. (warning: may be slow and use lots of data!)using with --verbose- ignoreListingChecksum - (bool) Do not use checksums for listings (add --ignore-checksum to additionally skip post-copy checksum checks)+- maxDeleteRenamesAware - (bool) Exclude tracked renames from the --max-delete+safety check (requires --track-renames)- maxLock - (Duration) Consider lock files older than this to be expired(default: 0 (never expire)) (minimum: 2m)- noCleanup - (bool) Retain working files (useful for troubleshooting and@@ -24488,7 +25844,27 @@ This takes the following parameters:- srcFs - a remote name string e.g. "drive:src" for the source- dstFs - a remote name string e.g. "drive:dst" for the destination- createEmptySrcDirs - create empty src directories on destination if set+- combined - make a combined report of changes (default false)+- missingOnSrc - report all files missing from the source (default false)+- missingOnDst - report all files missing from the destination (default false)+- match - report all matching files (default false)+- differ - report all non-matching files (default false)+- error - report all files with errors (hashing or reading) (default false)+- destAfter - report all files that exist on the destination post-copy (default false)+Returns:++- combined - array of strings of combined report of changes+- missingOnSrc - array of strings of all files missing from the source+- missingOnDst - array of strings of all files missing from the destination+- match - array of strings of all matching files+- differ - array of strings of all non-matching files+- error - array of strings of all files with errors (hashing or reading)+- destAfter - array of strings of all files that exist on the destination post-copy++Each report is only returned if its parameter is set to true. If the+operation fails the reports are only available if it was run with+`_async`, as part of the job output.See the [copy](https://rclone.org/commands/rclone_copy/) command for more information on the above.@@ -24500,7 +25876,27 @@ This takes the following parameters:- dstFs - a remote name string e.g. "drive:dst" for the destination- createEmptySrcDirs - create empty src directories on destination if set- deleteEmptySrcDirs - delete empty src directories if set+- combined - make a combined report of changes (default false)+- missingOnSrc - report all files missing from the source (default false)+- missingOnDst - report all files missing from the destination (default false)+- match - report all matching files (default false)+- differ - report all non-matching files (default false)+- error - report all files with errors (hashing or reading) (default false)+- destAfter - report all files that exist on the destination post-move (default false)++Returns:++- combined - array of strings of combined report of changes+- missingOnSrc - array of strings of all files missing from the source+- missingOnDst - array of strings of all files missing from the destination+- match - array of strings of all matching files+- differ - array of strings of all non-matching files+- error - array of strings of all files with errors (hashing or reading)+- destAfter - array of strings of all files that exist on the destination post-move+Each report is only returned if its parameter is set to true. If the+operation fails the reports are only available if it was run with+`_async`, as part of the job output.See the [move](https://rclone.org/commands/rclone_move/) command for more information on the above.@@ -24511,7 +25907,27 @@ This takes the following parameters:- srcFs - a remote name string e.g. "drive:src" for the source- dstFs - a remote name string e.g. "drive:dst" for the destination- createEmptySrcDirs - create empty src directories on destination if set+- combined - make a combined report of changes (default false)+- missingOnSrc - report all files missing from the source (default false)+- missingOnDst - report all files missing from the destination (default false)+- match - report all matching files (default false)+- differ - report all non-matching files (default false)+- error - report all files with errors (hashing or reading) (default false)+- destAfter - report all files that exist on the destination post-sync (default false)+Returns:++- combined - array of strings of combined report of changes+- missingOnSrc - array of strings of all files missing from the source+- missingOnDst - array of strings of all files missing from the destination+- match - array of strings of all matching files+- differ - array of strings of all non-matching files+- error - array of strings of all files with errors (hashing or reading)+- destAfter - array of strings of all files that exist on the destination post-sync++Each report is only returned if its parameter is set to true. If the+operation fails the reports are only available if it was run with+`_async`, as part of the job output.See the [sync](https://rclone.org/commands/rclone_sync/) command for more information on the above.@@ -24751,9 +26167,11 @@ The keys in the error response are:### CORS-The sever implements basic CORS support and allows all origins for that.-The response to a preflight OPTIONS request will echo the requested-"Access-Control-Request-Headers" back.+The server implements basic CORS support when `--rc-allow-origin` is+set. Browser CORS preflight OPTIONS requests are then answered by+rclone without authentication with the `Access-Control-*` headers.+Without `--rc-allow-origin` no CORS headers are sent and a preflight+request is authenticated like any other request.### Using POST with URL parameters only@@ -24866,7 +26284,7 @@ profiling tools on the same port.The profiling endpoints follow the same authentication rules as therest of the rc: they are only served if authentication has been set up(`--rc-user`/`--rc-pass` or `--rc-htpasswd`) or the-[`--rc-no-auth`](#--rc-no-auth) flag is in use. For debugging on the+[`--rc-no-auth`](#rc-no-auth) flag is in use. For debugging on thedefault localhost port the easiest thing is to use `--rc --rc-no-auth`(but see the warning about using `--rc-no-auth` on a non-loopbackbind).@@ -25580,6 +26998,7 @@ Flags used for check commands.Flags for general networking and HTTP stuff.```+ --allow-https-downgrade Follow redirects from HTTPS to HTTP (insecure)--bind string Local address to bind to for outgoing connections, IPv4, IPv6 or name--bwlimit BwTimetable Bandwidth limit in KiB/s, or use suffix B|K|M|G|T|P or a full timetable--bwlimit-file BwTimetable Bandwidth limit per file in KiB/s, or use suffix B|K|M|G|T|P or a full timetable@@ -25603,7 +27022,7 @@ Flags for general networking and HTTP stuff.--tpslimit float Limit HTTP transactions per second to this--tpslimit-burst int Max burst of transactions for --tpslimit (default 1)--use-cookies Enable session cookiejar- --user-agent string Set the user-agent to a specified string (default "rclone/v1.75.0")+ --user-agent string Set the user-agent to a specified string (default "rclone/v1.76.0")```@@ -25784,11 +27203,6 @@ Flags to control the Remote Control API.--rc-template string User-specified template--rc-user string User name for authentication--rc-user-from-header string User name from a defined HTTP header- --rc-web-fetch-url string URL to fetch the releases for webgui (default "https://api.github.com/repos/rclone/rclone-webui-react/releases/latest")- --rc-web-gui Launch WebGUI on localhost- --rc-web-gui-force-update Force update to latest version of web gui- --rc-web-gui-no-open-browser Don't open the browser automatically- --rc-web-gui-update Check and update to latest version of web gui```@@ -25850,8 +27264,10 @@ Backend-only flags (these can be set in the config file also).--azureblob-encoding Encoding The encoding for the backend (default Slash,BackSlash,Del,Ctl,RightPeriod,InvalidUtf8)--azureblob-endpoint string Endpoint for the service--azureblob-env-auth Read credentials from runtime (environment variables, CLI or MSI)+ --azureblob-hns Tristate Set if the storage account has a hierarchical namespace (default unset)--azureblob-key string Storage Account Shared Key--azureblob-list-chunk int Size of blob list (default 5000)+ --azureblob-list-parallelism int Number of parallel shards to list a directory with--azureblob-msi-client-id string Object ID of the user-assigned MSI to use, if any--azureblob-msi-mi-res-id string Azure resource ID of the user-assigned MSI to use, if any--azureblob-msi-object-id string Object ID of the user-assigned MSI to use, if any@@ -25864,6 +27280,7 @@ Backend-only flags (these can be set in the config file also).--azureblob-tenant string ID of the service principal's tenant. Also called its directory ID--azureblob-upload-concurrency int Concurrency for multipart uploads (default 16)--azureblob-upload-cutoff string Cutoff for switching to chunked upload (<= 256 MiB) (deprecated)+ --azureblob-use-arrow-list Use the Apache Arrow listing format--azureblob-use-az Use Azure CLI tool az for authentication--azureblob-use-copy-blob Whether to use the Copy Blob API when copying to the same storage account (default true)--azureblob-use-emulator Uses local storage emulator if provided as 'true'@@ -25996,6 +27413,12 @@ Backend-only flags (these can be set in the config file also).--doi-doi string The DOI or the doi.org URL--doi-doi-resolver-api-url string The URL of the DOI resolver API to use--doi-provider string DOI provider+ --dosya-api-key string Your API Key, get it from https://dosya.dev/profile under "API keys"+ --dosya-api-url string API base URL (default "https://api.dosya.dev")+ --dosya-description string Description of the remote+ --dosya-encoding Encoding The encoding for the backend (default Slash,LtGt,DoubleQuote,SingleQuote,BackSlash,Del,Ctl,LeftSpace,RightSpace,InvalidUtf8,Dot)+ --dosya-upload-concurrency int Concurrency for multipart uploads (default 4)+ --dosya-workspace-id string Your workspace ID--drime-access-token string API Access token--drime-chunk-size SizeSuffix Chunk size to use for uploading (default 5Mi)--drime-description string Description of the remote@@ -26218,6 +27641,8 @@ Backend-only flags (these can be set in the config file also).--http-headers CommaSepList Set HTTP headers for all transactions--http-no-escape Do not escape URL metacharacters in path names--http-no-head Don't use HEAD requests+ --http-no-head-size SizeSuffix Set the reported file size for --http-no-head (default off)+ --http-no-head-time Time Set a default modification time for --http-no-head (default off)--http-no-slash Set this if the site doesn't end directories with /--http-url string URL of HTTP host to connect to--huaweidrive-auth-url string Auth server URL@@ -26277,6 +27702,17 @@ Backend-only flags (these can be set in the config file also).--jottacloud-token-url string Token server url--jottacloud-trashed-only Only show files that are in the trash--jottacloud-upload-resume-limit SizeSuffix Files bigger than this can be resumed if the upload fail's (default 10Mi)+ --kdrive-auth-url string Auth server URL+ --kdrive-chunk-size SizeSuffix Files above this size are uploaded in chunks of this size (default 20Mi)+ --kdrive-client-id string OAuth Client Id+ --kdrive-description string Description of the remote+ --kdrive-drive-id string ID of the kDrive to use+ --kdrive-encoding Encoding The encoding for the backend (default Slash,Del,Ctl,LeftSpace,RightSpace,InvalidUtf8,Dot)+ --kdrive-endpoint string The API endpoint to use+ --kdrive-root-folder-id string The folder to use as root+ --kdrive-token string OAuth Access Token as a JSON blob+ --kdrive-token-url string Token server url+ --kdrive-upload-concurrency int Concurrency for chunked uploads (default 4)--koofr-description string Description of the remote--koofr-encoding Encoding The encoding for the backend (default Slash,BackSlash,Del,Ctl,InvalidUtf8,Dot)--koofr-endpoint string The Koofr API endpoint to use@@ -26362,6 +27798,7 @@ Backend-only flags (these can be set in the config file also).--onedrive-root-folder-id string ID of the root folder--onedrive-server-side-across-configs Deprecated: use --server-side-across-configs instead--onedrive-tenant string ID of the service principal's tenant. Also called its directory ID+ --onedrive-tenant-api-version string The SharePoint API version to use with tenant_url (default "v2.0")--onedrive-tenant-url string The tenant URL for non-admin OneDrive access--onedrive-token string OAuth Access Token as a JSON blob--onedrive-token-url string Token server url@@ -26376,7 +27813,7 @@ Backend-only flags (these can be set in the config file also).--oos-decompress If set this will decompress gzip encoded objects--oos-description string Description of the remote--oos-disable-checksum Don't store MD5 checksum with object metadata- --oos-encoding Encoding The encoding for the backend (default Slash,InvalidUtf8,Dot)+ --oos-encoding Encoding The encoding for the backend (default Slash,CrLf,InvalidUtf8,Dot)--oos-endpoint string Endpoint for Object storage API--oos-leave-parts-on-error If true avoid calling abort upload on a failure, leaving all successfully uploaded parts for manual recovery--oos-max-upload-parts int Maximum number of parts in a multipart upload (default 10000)@@ -26571,7 +28008,7 @@ Backend-only flags (these can be set in the config file also).--sftp-disable-concurrent-reads If set don't use concurrent reads--sftp-disable-concurrent-writes If set don't use concurrent writes--sftp-disable-hashcheck Disable the execution of SSH commands to determine if remote file hashing is available- --sftp-encoding Encoding The encoding for the backend (default Slash,Del,Ctl,Dot)+ --sftp-encoding Encoding The encoding for the backend (default Slash,Dot)--sftp-hashes CommaSepList Comma separated list of supported checksum types--sftp-host string SSH host to connect to--sftp-host-key-algorithms SpaceSepList Space separated list of host key algorithms, ordered by preference@@ -26586,6 +28023,7 @@ Backend-only flags (these can be set in the config file also).--sftp-known-hosts-file string Optional path to known_hosts file--sftp-macs SpaceSepList Space separated list of MACs (message authentication code) algorithms, ordered by preference--sftp-md5sum-command string The command used to read MD5 hashes+ --sftp-multithread-upload Set this to use multi-thread (multi-connection) uploads--sftp-pass string SSH password, leave blank to use ssh-agent (obscured)--sftp-path-override string Override path used by SSH shell commands--sftp-pin-host-key Pin the server host key on first connection (Trust On First Use)@@ -26648,7 +28086,8 @@ Backend-only flags (these can be set in the config file also).--smb-port int SMB port number (default 445)--smb-spn string Service principal name--smb-use-kerberos Use Kerberos authentication- --smb-user string SMB username (default "$USER")+ --smb-user string SMB username+ --smb-workstation string Workstation name sent in NTLM authentication--storj-access-grant string Access grant--storj-api-key string API key--storj-description string Description of the remote@@ -26723,6 +28162,7 @@ Backend-only flags (these can be set in the config file also).--webdav-url string URL of http host to connect to--webdav-user string User name--webdav-vendor string Name of the WebDAV site/service/software you are using+ --yandex-app-folder Use the application folder as the root--yandex-auth-url string Auth server URL--yandex-client-credentials Use client credentials OAuth flow--yandex-client-id string OAuth Client Id@@ -27468,6 +28908,7 @@ Optional Flags:--force Bypass --max-delete safety check and run the sync. Consider using with --verbose-h, --help help for bisync--ignore-listing-checksum Do not use checksums for listings (add --ignore-checksum to additionally skip post-copy checksum checks)+ --max-delete-renames-aware Exclude tracked renames from the --max-delete safety check (requires --track-renames)--max-lock Duration Consider lock files older than this to be expired (default: 0 (never expire)) (minimum: 2m) (default 0s)--no-cleanup Retain working files (useful for troubleshooting and testing).--no-slow-hash Ignore listing checksums only on backends where they are slow@@ -27540,7 +28981,7 @@ will (by default) be overwritten by the Path1 filesystem version.(Note that this is [NOT entirely symmetrical](https://github.com/rclone/rclone/issues/5681#issuecomment-938761815),and more symmetrical options can be specified with the[`--resync-mode`](#resync-mode) flag.)-Carefully evaluate deltas using [--dry-run](https://rclone.org/flags/#non-backend-flags).+Carefully evaluate deltas using [--dry-run](https://rclone.org/docs/#n-dry-run).For a resync run, one of the paths may be empty (no files in the path tree).The resync run should result in files on both paths, else a normal non-resync@@ -27680,7 +29121,7 @@ Time stamps and file contents for `RCLONE_TEST` files are not important, justthe names and locations. If you have symbolic links in your sync tree it isrecommended to place `RCLONE_TEST` files in the linked-to directory tree toprotect against bisync assuming a bunch of deleted files if the linked-to tree-should not be accessible. See also the [--check-filename](--check-filename) flag.+should not be accessible. See also the [--check-filename](#check-filename) flag.### --check-filename@@ -27857,6 +29298,23 @@ to bypass the check.Also see the [all files changed](#all-files-changed) check.+### --max-delete-renames-aware++When `--track-renames` is in use, `--max-delete-renames-aware` can exclude+files that are guaranteed to match as tracked renames from the deletion count.+If the initial deletion count exceeds `--max-delete`, bisync performs a rename+matching preflight before starting either sync direction. The run still aborts+if the remaining unmatched deletions exceed the limit.++The preflight uses the selected [`--track-renames-strategy`](https://rclone.org/docs/#track-renames-strategy)+and the attributes stored in the normal bisync listings. A hash strategy forces+both listings to use a common hash and is incompatible with+`--ignore-listing-checksum`. The option only exempts matches when the destination+supports server-side moves or copies and the selected strategy is supported by both paths.+It requires `--track-renames` and is disabled by default because hashing may add+listing time. `--force` bypasses both the safety check and this preflight.+The preflight is also disabled during `--resync`, when rename tracking is unavailable.+### --filters-file {#filters-file}By using rclone filter features you can exclude file types or directory@@ -28055,7 +29513,7 @@ other, if there were changes during or after the run. This is normal, and anydifferences will be detected and synced on the next run.For a robust integrity check of the current state of the remotes (as opposed to-just their listing snapshots), consider using [`check`](commands/rclone_check/)+just their listing snapshots), consider using [`check`](https://rclone.org/commands/rclone_check/)(or [`cryptcheck`](https://rclone.org/commands/rclone_cryptcheck/), if at least one path is a`crypt` remote) instead of `--check-sync`, keeping in mind that differences areexpected if files changed during or after your last bisync run.@@ -28289,7 +29747,7 @@ conflicts, which would previously have resulted in unnecessary renames andduplicates. Now, when bisync comes to a file that it wants to rename (because itis new/changed on both sides), it first checks whether the Path1 and Path2versions are currently *identical* (using the same underlying function as-[`check`](commands/rclone_check/).) If bisync concludes that the files are+[`check`](https://rclone.org/commands/rclone_check/).) If bisync concludes that the files areidentical, it will skip them and move on. Otherwise, it will create renamedduplicates, as before. This behavior also [improves the experience of renamingdirectories](https://forum.rclone.org/t/bisync-bugs-and-feature-requests/37636#:~:text=Renamed%20directories),@@ -28413,14 +29871,20 @@ The following backends have known issues that need more investigation:- `TestHuaweiDrive` (`huaweidrive`)- [`TestBisyncRemoteLocal/ext_paths`](https://pub.rclone.org/integration-tests/current/huaweidrive-cmd.bisync-TestHuaweiDrive-1.txt)- [`TestBisyncRemoteLocal/extended_filenames`](https://pub.rclone.org/integration-tests/current/huaweidrive-cmd.bisync-TestHuaweiDrive-1.txt)- - [`TestBisyncRemoteLocal/normalization`](https://pub.rclone.org/integration-tests/current/huaweidrive-cmd.bisync-TestHuaweiDrive-1.txt)- [`TestBisyncLocalRemote/ext_paths`](https://pub.rclone.org/integration-tests/current/huaweidrive-cmd.bisync-TestHuaweiDrive-1.txt)- [`TestBisyncLocalRemote/extended_filenames`](https://pub.rclone.org/integration-tests/current/huaweidrive-cmd.bisync-TestHuaweiDrive-1.txt)- - [4 more](https://pub.rclone.org/integration-tests/current/)+ - [`TestBisyncLocalRemote/normalization`](https://pub.rclone.org/integration-tests/current/huaweidrive-cmd.bisync-TestHuaweiDrive-1.txt)+ - [3 more](https://pub.rclone.org/integration-tests/current/)- `TestPcloud` (`pcloud`)- - [`TestBisyncRemoteRemote/check_access`](https://pub.rclone.org/integration-tests/current/pcloud-cmd.bisync-TestPcloud-1.txt)- - [`TestBisyncRemoteRemote/rmdirs`](https://pub.rclone.org/integration-tests/current/pcloud-cmd.bisync-TestPcloud-1.txt)-- Updated: 2026-07-31-010017+ - [`TestBisyncRemoteRemote/extended_filenames`](https://pub.rclone.org/integration-tests/current/pcloud-cmd.bisync-TestPcloud-1.txt)+- `TestYandex` (`yandex`)+ - [`TestBisyncRemoteLocal/all_changed`](https://pub.rclone.org/integration-tests/current/yandex-cmd.bisync-TestYandex-1.txt)+ - [`TestBisyncRemoteLocal/backupdir`](https://pub.rclone.org/integration-tests/current/yandex-cmd.bisync-TestYandex-1.txt)+ - [`TestBisyncRemoteLocal/basic`](https://pub.rclone.org/integration-tests/current/yandex-cmd.bisync-TestYandex-1.txt)+ - [`TestBisyncRemoteLocal/changes`](https://pub.rclone.org/integration-tests/current/yandex-cmd.bisync-TestYandex-1.txt)+ - [`TestBisyncRemoteLocal/check_access`](https://pub.rclone.org/integration-tests/current/yandex-cmd.bisync-TestYandex-1.txt)+ - [94 more](https://pub.rclone.org/integration-tests/current/)+- Updated: 2026-11-06-010024<!--- end list_failures - DO NOT EDIT THIS SECTION - use make commanddocs --->The following backends either have not been tested recently or have known issues@@ -28438,6 +29902,8 @@ that are deemed unfixable for the time being:- `TestFilesCom` (`filescom`)- `TestImageKit` (`imagekit`)- `TestJottacloud` (`jottacloud`)+- `TestKdrive` (`kdrive`)+ - `TestBisyncRemoteRemote/normalization`- `TestLinkbox` (`linkbox`)- `TestMailru` (`mailru`)- `TestMega` (`mega`)@@ -28450,7 +29916,8 @@ that are deemed unfixable for the time being:- `TestQuatrix` (`quatrix`)- `TestS3GCS` (`s3`)- `TestBisyncRemoteRemote/extended_filenames`-- `TestS3Rclone` (`s3`)+ - `TestBisyncRemoteLocal/extended_filenames`+ - `TestBisyncLocalRemote/extended_filenames`- `TestSFTPRsyncNet` (`sftp`)- `TestStorj` (`storj`)- `TestWebdavNextcloud` (`webdav`)@@ -28529,10 +29996,10 @@ is to rename it to the same name on both sides. (As of `rclone v1.64`,a `--resync` is no longer required after doing so, as bisync will automaticallydetect that Path1 and Path2 are in agreement.)-Note that although the flag --track-renames ensures that renamed/moved files won't-be deleted and uploaded again, they are still counted as deleted files for purposes-of the --max-delete flag (as this check happens before the rename detection-operation). See [this issue](https://github.com/rclone/rclone/issues/8685).+By default, renamed or moved files are still counted as deleted files for+purposes of `--max-delete`, because the safety check happens before rename+detection. Use `--max-delete-renames-aware` together with `--track-renames` to+run a rename preflight and exclude guaranteed matches from this count.### `--fast-list` used by default@@ -29272,108 +30739,9 @@ about *Unison* and synchronization in general.## Changelog-### `v1.74.2`--- Fixed an issue causing `--conflict-loser pathname` to produce unexpected-behavior if using a non-default `--conflict-resolve` value.--### `v1.74`--- Added several missing `rc` parameters.-- Optional `rc` parameters are now truly optional.-- `rc` output now provides more structured information.--### `v1.71`--- `bisync` is now officially released from beta.--- Fixed an issue causing hash type to be set incorrectly if `path2` had slow-hashes and `--no-slow-hash` or `--slow-hash-sync-only` was in use.--- Fixed an issue causing errors when running concurrent bisync runs through the `rc`.--### `v1.69.1`--- Fixed an issue causing listings to not capture concurrent modifications under-certain conditions--### `v1.68`--- Fixed an issue affecting backends that round modtimes to a lower precision.--### `v1.67`--- Added integration tests against all backends.--### `v1.66`--- Copies and deletes are now handled in one operation instead of two--- `--track-renames` and `--backup-dir` are now supported-- Partial uploads known issue on `local`/`ftp`/`sftp` has been resolved (unless-using `--inplace`)-- Final listings are now generated from sync results, to avoid needing to re-list-- Bisync is now much more resilient to changes that happen during a bisync run,-and far less prone to critical errors / undetected changes-- Bisync is now capable of rolling a file listing back in cases of uncertainty,-essentially marking the file as needing to be rechecked next time.-- A few basic terminal colors are now supported, controllable with-[`--color`](https://rclone.org/docs/#color) (`AUTO`|`NEVER`|`ALWAYS`)-- Initial listing snapshots of Path1 and Path2 are now generated concurrently,-using the same "march" infrastructure as `check` and `sync`,-for performance improvements and less-[risk of error](https://forum.rclone.org/t/bisync-bugs-and-feature-requests/37636#:~:text=4.%20Listings%20should%20alternate%20between%20paths%20to%20minimize%20errors).-- Fixed handling of unicode normalization and case insensitivity, support for-[`--fix-case`](https://rclone.org/docs/#fix-case), [`--ignore-case-sync`](/docs/#ignore-case-sync),-[`--no-unicode-normalization`](https://rclone.org/docs/#no-unicode-normalization)-- `--resync` is now much more efficient (especially for users of `--create-empty-src-dirs`)-- Google Docs (and other files of unknown size) are now supported (with the same-options as in `sync`)-- Equality checks before a sync conflict rename now fall back to `cryptcheck`-(when possible) or `--download`,-instead of of `--size-only`, when `check` is not available.-- Bisync no longer fails to find the correct listing file when configs are-overridden with backend-specific flags.-- Bisync now fully supports comparing based on any combination of size, modtime,-and checksum, lifting the prior restriction on backends without modtime support.-- Bisync now supports a "Graceful Shutdown" mode to cleanly cancel a run early-without requiring `--resync`.-- New `--recover` flag allows robust recovery in the event of interruptions,-without requiring `--resync`.-- A new `--max-lock` setting allows lock files to automatically renew and expire,-for better automatic recovery when a run is interrupted.-- Bisync now supports auto-resolving sync conflicts and customizing rename-behavior with new [`--conflict-resolve`](#conflict-resolve),-[`--conflict-loser`](#conflict-loser), and-[`--conflict-suffix`](#conflict-suffix) flags.-- A new [`--resync-mode`](#resync-mode) flag allows more control over which-version of a file gets kept during a `--resync`.-- Bisync now supports [`--retries`](https://rclone.org/docs/#retries-int) and [`--retries-sleep`](/docs/#retries-sleep-time)-(when [`--resilient`](#resilient) is set.)--### `v1.64`--- Fixed an [issue](https://forum.rclone.org/t/bisync-bugs-and-feature-requests/37636#:~:text=1.%20Dry%20runs%20are%20not%20completely%20dry)-causing dry runs to inadvertently commit filter changes--- Fixed an [issue](https://forum.rclone.org/t/bisync-bugs-and-feature-requests/37636#:~:text=2.%20%2D%2Dresync%20deletes%20data%2C%20contrary%20to%20docs)-causing `--resync` to erroneously delete empty folders and duplicate files-unique to Path2-- `--check-access` is now enforced during `--resync`, preventing data loss in-[certain user error scenarios](https://forum.rclone.org/t/bisync-bugs-and-feature-requests/37636#:~:text=%2D%2Dcheck%2Daccess%20doesn%27t%20always%20fail%20when%20it%20should)-- Fixed an [issue](https://forum.rclone.org/t/bisync-bugs-and-feature-requests/37636#:~:text=5.%20Bisync%20reads%20files%20in%20excluded%20directories%20during%20delete%20operations)-causing bisync to consider more files than necessary due to overbroad filters-during delete operations-- [Improved detection of false positive change conflicts](https://forum.rclone.org/t/bisync-bugs-and-feature-requests/37636#:~:text=1.%20Identical%20files%20should%20be%20left%20alone%2C%20even%20if%20new/newer/changed%20on%20both%20sides)-(identical files are now left alone instead of renamed)-- Added [support for `--create-empty-src-dirs`](https://forum.rclone.org/t/bisync-bugs-and-feature-requests/37636#:~:text=3.%20Bisync%20should%20create/delete%20empty%20directories%20as%20sync%20does%2C%20when%20%2D%2Dcreate%2Dempty%2Dsrc%2Ddirs%20is%20passed)-- Added experimental `--resilient` mode to allow-[recovery from self-correctable errors](https://forum.rclone.org/t/bisync-bugs-and-feature-requests/37636#:~:text=2.%20Bisync%20should%20be%20more%20resilient%20to%20self%2Dcorrectable%20errors)-- Added [new `--ignore-listing-checksum` flag](https://forum.rclone.org/t/bisync-bugs-and-feature-requests/37636#:~:text=6.%20%2D%2Dignore%2Dchecksum%20should%20be%20split%20into%20two%20flags%20for%20separate%20purposes)-to distinguish from `--ignore-checksum`-- [Performance improvements](https://forum.rclone.org/t/bisync-bugs-and-feature-requests/37636#:~:text=6.%20Deletes%20take%20several%20times%20longer%20than%20copies)-for large remotes-- Documentation and testing improvements+Changes to bisync are listed in the [rclone changelog](https://rclone.org/changelog/).+The changelog this page kept until `v1.76` is+[archived on GitHub](https://github.com/rclone/rclone/blob/0891ce30606c726990c98d70812d31b555782791/docs/content/bisync.md?plain=1#L1934-L2061).# Release signing@@ -29500,19 +30868,23 @@ Repeat for `MD5SUMS` and `SHA1SUMS` if desired.Now that we know the signatures on the hashes are OK we can verify thebinaries match the hashes, completing the verification.+Extract the signed data with:++```console+$ gpg --output SHA256SUMS.signed --decrypt SHA256SUMS+```++Then check the binaries:+```console-$ sha256sum -c SHA256SUMS 2>&1 | grep OK+$ sha256sum -c --ignore-missing SHA256SUMS.signedrclone-v1.63.1-windows-amd64.zip: OK```Or do the check with rclone```console-$ rclone hashsum sha256 -C SHA256SUMS rclone-v1.63.1-windows-amd64.zip-2023/09/11 10:53:58 NOTICE: SHA256SUMS: improperly formatted checksum line 0-2023/09/11 10:53:58 NOTICE: SHA256SUMS: improperly formatted checksum line 1-2023/09/11 10:53:58 NOTICE: SHA256SUMS: improperly formatted checksum line 49-2023/09/11 10:53:58 NOTICE: SHA256SUMS: 4 warning(s) suppressed...+$ rclone hashsum sha256 -C SHA256SUMS.signed rclone-v1.63.1-windows-amd64.zip= rclone-v1.63.1-windows-amd64.zip2023/09/11 10:53:58 NOTICE: Local file system at /tmp/check: 0 differences found2023/09/11 10:53:58 NOTICE: Local file system at /tmp/check: 1 matching files@@ -29902,8 +31274,10 @@ The S3 backend can be used with a number of different providers:- Alibaba Cloud (Aliyun) Object Storage System (OSS)- Ceph- China Mobile Ecloud Elastic Object Storage (EOS)+- Clever Cloud Cellar- Cloudflare R2- Arvan Cloud Object Storage (AOS)+- Autonomys Auto Drive- Bizfly Cloud Simple Storage- Cubbit DS3- DigitalOcean Spaces@@ -29918,6 +31292,7 @@ The S3 backend can be used with a number of different providers:- IBM COS S3- IDrive e2- Impossible Cloud+- Infomaniak Swiss Backup- Intercolo Object Storage- IONOS Cloud- Leviia Object Storage@@ -29926,6 +31301,7 @@ The S3 backend can be used with a number of different providers:- Magalu Object Storage- MEGA S4 Object Storage- Minio+- Neon Object Storage- Outscale- OVHcloud- Petabox@@ -29934,6 +31310,7 @@ The S3 backend can be used with a number of different providers:- Rabata Cloud Storage- RackCorp Object Storage- Rclone Serve S3+- RelAix.Storage- Scaleway- Scality (RING / ARTESCA)- Seagate Lyve Cloud@@ -30790,6 +32167,9 @@ tries to access data from the glacier storage class you will see an error like b2017/09/11 19:07:43 Failed to sync: failed to open source object: Object in GLACIER, restore first: path/to/file```+The message shows the object's actual storage class, for example+`DEEP_ARCHIVE`.+In this case you need to [restore](http://docs.aws.amazon.com/AmazonS3/latest/user-guide/restore-archived-objects.html)the object(s) in question before accessing object contents.The [restore](#restore) section below shows how to do this with rclone.@@ -30853,7 +32233,7 @@ mode from COMPLIANCE to GOVERNANCE while preserving the original retention date:<!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/s3/s3.go and run make backenddocs to verify --> <!-- markdownlint-disable-line line-length -->### Standard options-Here are the Standard options specific to s3 (Amazon S3 Compliant Storage Providers including AWS, Alibaba, ArvanCloud, BizflyCloud, Ceph, ChinaMobile, Cloudflare, Cubbit, DigitalOcean, Dreamhost, Exaba, Fastly, FileLu, FlashBlade, GCS, HCP, Hetzner, HuaweiOBS, IBMCOS, IDrive, ImpossibleCloud, Intercolo, IONOS, Leviia, Liara, Linode, LyveCloud, Magalu, Mega, Minio, Netease, Outscale, OVHcloud, Petabox, Qiniu, Rabata, RackCorp, Rclone, Scaleway, Scality, SeaweedFS, Selectel, Servercore, SpectraLogic, Storj, Synology, TencentCOS, US3, Wasabi, Zadara, Zata, ZeroServices, Other).+Here are the Standard options specific to s3 (Amazon S3 Compliant Storage Providers including AWS, Alibaba, ArvanCloud, Autonomys, BizflyCloud, Cellar, Ceph, ChinaMobile, Cloudflare, Cubbit, DigitalOcean, Dreamhost, Exaba, Fastly, FileLu, FlashBlade, GCS, HCP, Hetzner, HuaweiOBS, IBMCOS, IDrive, ImpossibleCloud, Infomaniak, Intercolo, IONOS, Leviia, Liara, Linode, LyveCloud, Magalu, Mega, Minio, Neon, Netease, Outscale, OVHcloud, Petabox, Qiniu, Rabata, RackCorp, Rclone, RelAix, Scaleway, Scality, SeaweedFS, Selectel, Servercore, SpectraLogic, Storj, Synology, TencentCOS, US3, Wasabi, Zadara, Zata, ZeroServices, Other).#### --s3-provider@@ -30872,8 +32252,12 @@ Properties:- Alibaba Cloud Object Storage System (OSS) formerly Aliyun- "ArvanCloud"- Arvan Cloud Object Storage (AOS)+ - "Autonomys"+ - Autonomys Auto Drive- "BizflyCloud"- Bizfly Cloud Simple Storage+ - "Cellar"+ - Clever Cloud Cellar Object Storage- "Ceph"- Ceph Object Storage- "ChinaMobile"@@ -30908,6 +32292,8 @@ Properties:- IDrive e2- "ImpossibleCloud"- Impossible Cloud Object Storage+ - "Infomaniak"+ - Infomaniak Swiss Backup- "Intercolo"- Intercolo Object Storage- "IONOS"@@ -30926,6 +32312,8 @@ Properties:- MEGA S4 Object Storage- "Minio"- Minio Object Storage+ - "Neon"+ - Neon Object Storage- "Netease"- Netease Object Storage (NOS)- "Outscale"@@ -30942,6 +32330,8 @@ Properties:- RackCorp Object Storage- "Rclone"- Rclone S3 Server+ - "RelAix"+ - RelAix.Storage- "Scaleway"- Scaleway Object Storage- "Scality"@@ -31027,7 +32417,7 @@ Properties:- Config: region- Env Var: RCLONE_S3_REGION-- Provider: AWS,BizflyCloud,Ceph,Cloudflare,Cubbit,DigitalOcean,Dreamhost,Exaba,Fastly,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IBMCOS,ImpossibleCloud,Intercolo,IONOS,Leviia,LyveCloud,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,Rabata,RackCorp,Scaleway,Scality,SeaweedFS,Selectel,Servercore,Synology,Wasabi,Zadara,Zata,ZeroServices,Other+- Provider: AWS,BizflyCloud,Ceph,Cloudflare,Cubbit,DigitalOcean,Dreamhost,Exaba,Fastly,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IBMCOS,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,LyveCloud,Minio,Neon,Netease,Outscale,OVHcloud,Petabox,Qiniu,Rabata,RackCorp,Scaleway,Scality,SeaweedFS,Selectel,Servercore,Synology,Wasabi,Zadara,Zata,ZeroServices,Other- Type: string- Required: false- Examples:@@ -31162,6 +32552,9 @@ Properties:- "eu-central"- EU Central- Provider: Fastly+ - "eu-central-1"+ - EU Central 1+ - Provider: Fastly- "eu-south-1"- EU South 1- Provider: Fastly@@ -31213,50 +32606,104 @@ Properties:- "nbg1"- Nuremberg- Provider: Hetzner+ - "af-north-1"+ - AF-Cairo+ - Provider: HuaweiOBS- "af-south-1"- AF-Johannesburg- Provider: HuaweiOBS+ - "ap-southeast-1"+ - CN-Hong Kong+ - Provider: HuaweiOBS- "ap-southeast-2"- AP-Bangkok- Provider: HuaweiOBS- "ap-southeast-3"- AP-Singapore- Provider: HuaweiOBS- - "cn-east-3"- - CN East-Shanghai1+ - "ap-southeast-4"+ - AP-Jakarta+ - Provider: HuaweiOBS+ - "ap-southeast-5"+ - AP-Manila- Provider: HuaweiOBS- "cn-east-2"- CN East-Shanghai2- Provider: HuaweiOBS+ - "cn-east-3"+ - CN East-Shanghai1+ - Provider: HuaweiOBS+ - "cn-east-4"+ - CN East-Qingdao+ - Provider: HuaweiOBS+ - "cn-east-5"+ - CN East-Suzhou+ - Provider: HuaweiOBS- "cn-north-1"- CN North-Beijing1- Provider: HuaweiOBS+ - "cn-north-2"+ - CN North-Beijing2+ - Provider: HuaweiOBS- "cn-north-4"- CN North-Beijing4- Provider: HuaweiOBS+ - "cn-north-9"+ - CN North-Ulanqab+ - Provider: HuaweiOBS+ - "cn-north-11"+ - CN North-Hebei+ - Provider: HuaweiOBS+ - "cn-north-12"+ - CN North-Hohhot+ - Provider: HuaweiOBS- "cn-south-1"- CN South-Guangzhou- Provider: HuaweiOBS- - "ap-southeast-1"- - CN-Hong Kong+ - "cn-south-2"+ - CN South-Shenzhen- Provider: HuaweiOBS- - "sa-argentina-1"- - LA-Buenos Aires1+ - "cn-south-4"+ - CN South-Guangzhou-InvitationOnly- Provider: HuaweiOBS- - "sa-peru-1"- - LA-Lima1+ - "cn-southwest-2"+ - CN Southwest-Guiyang1+ - Provider: HuaweiOBS+ - "cn-southwest-3"+ - CN Southwest-Chongqing+ - Provider: HuaweiOBS+ - "eu-west-101"+ - EU-Dublin+ - Provider: HuaweiOBS+ - "la-north-2"+ - LA-Mexico City2+ - Provider: HuaweiOBS+ - "la-south-2"+ - LA-Santiago+ - Provider: HuaweiOBS+ - "me-east-1"+ - ME-Riyadh- Provider: HuaweiOBS- "na-mexico-1"- LA-Mexico City1- Provider: HuaweiOBS- - "sa-chile-1"- - LA-Santiago2+ - "ru-northwest-2"+ - RU-Moscow2+ - Provider: HuaweiOBS+ - "sa-argentina-1"+ - LA-Buenos Aires1- Provider: HuaweiOBS- "sa-brazil-1"- LA-Sao Paulo1- Provider: HuaweiOBS- - "ru-northwest-2"- - RU-Moscow2+ - "sa-chile-1"+ - LA-Santiago2+ - Provider: HuaweiOBS+ - "sa-peru-1"+ - LA-Lima1+ - Provider: HuaweiOBS+ - "tr-west-1"+ - TR-Istanbul- Provider: HuaweiOBS- "eu-central-2"- Frankfurt, Germany@@ -31279,6 +32726,10 @@ Properties:- "us-east-1"- New York, USA- Provider: ImpossibleCloud+ - "us-east-1"+ - Default region for compatibility.+ - All data is stored in Switzerland.+ - Provider: Infomaniak- "de-fra"- Frankfurt, Germany- Provider: Intercolo@@ -31300,6 +32751,18 @@ Properties:- "us-central-1"- Lenexa, USA- Provider: IONOS+ - "us-east-1"+ - US East (N. Virginia)+ - Provider: Neon,Petabox,Rabata+ - "us-east-2"+ - US East (Ohio)+ - Provider: Neon+ - "eu-central-1"+ - Europe (Frankfurt)+ - Provider: Neon,Petabox+ - "ap-southeast-1"+ - Asia Pacific (Singapore)+ - Provider: Neon,Petabox- "eu-west-2"- Paris, France- Provider: Outscale@@ -31357,15 +32820,6 @@ Properties:- "rbx-archive"- Roubaix, France (Cold Archive)- Provider: OVHcloud- - "us-east-1"- - US East (N. Virginia)- - Provider: Petabox,Rabata- - "eu-central-1"- - Europe (Frankfurt)- - Provider: Petabox- - "ap-southeast-1"- - Asia Pacific (Singapore)- - Provider: Petabox- "me-south-1"- Middle East (Bahrain)- Provider: Petabox@@ -31473,6 +32927,9 @@ Properties:- "pl-waw"- Warsaw, Poland- Provider: Scaleway+ - "it-mil"+ - Milan, Italy+ - Provider: Scaleway- "ru-1"- St. Petersburg- Provider: Selectel,Servercore@@ -31539,7 +32996,7 @@ Properties:- Config: endpoint- Env Var: RCLONE_S3_ENDPOINT-- Provider: AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,ChinaMobile,Cloudflare,Cubbit,DigitalOcean,Dreamhost,Exaba,Fastly,FileLu,FlashBlade,GCS,HCP,Hetzner,HuaweiOBS,IBMCOS,ImpossibleCloud,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Mega,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,Rabata,RackCorp,Rclone,Scaleway,Scality,SeaweedFS,Selectel,Servercore,SpectraLogic,Storj,Synology,TencentCOS,US3,Wasabi,Zadara,Zata,ZeroServices,Other+- Provider: AWS,Alibaba,ArvanCloud,Autonomys,BizflyCloud,Cellar,Ceph,ChinaMobile,Cloudflare,Cubbit,DigitalOcean,Dreamhost,Exaba,Fastly,FileLu,FlashBlade,GCS,HCP,Hetzner,HuaweiOBS,IBMCOS,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Mega,Minio,Neon,Netease,Outscale,OVHcloud,Petabox,Qiniu,Rabata,RackCorp,Rclone,RelAix,Scaleway,Scality,SeaweedFS,Selectel,Servercore,SpectraLogic,Storj,Synology,TencentCOS,US3,Wasabi,Zadara,Zata,ZeroServices,Other- Type: string- Required: false- Examples:@@ -31625,12 +33082,18 @@ Properties:- "s3.ir-tbz-sh1.arvanstorage.ir"- Tabriz Iran (Shahriar)- Provider: ArvanCloud+ - "s3.auto-drive.autonomys.xyz"+ - Auto Drive mainnet S3 endpoint+ - Provider: Autonomys- "hn.ss.bfcplatform.vn"- Hanoi endpoint- Provider: BizflyCloud- "hcm.ss.bfcplatform.vn"- Ho Chi Minh endpoint- Provider: BizflyCloud+ - "cellar-c2.services.clever-cloud.com"+ - Paris, France+ - Provider: Cellar- "eos-wuxi-1.cmecloud.cn"- The default endpoint - a good choice if you are unsure.- East China (Suzhou)@@ -31767,6 +33230,9 @@ Properties:- "eu-central.object.fastlystorage.app"- EU Central- Provider: Fastly+ - "eu-central-1.object.fastlystorage.app"+ - EU Central 1+ - Provider: Fastly- "eu-south-1.object.fastlystorage.app"- EU South 1- Provider: Fastly@@ -31821,50 +33287,104 @@ Properties:- "nbg1.your-objectstorage.com"- Nuremberg- Provider: Hetzner+ - "obs.af-north-1.myhuaweicloud.com"+ - AF-Cairo+ - Provider: HuaweiOBS- "obs.af-south-1.myhuaweicloud.com"- AF-Johannesburg- Provider: HuaweiOBS+ - "obs.ap-southeast-1.myhuaweicloud.com"+ - CN-Hong Kong+ - Provider: HuaweiOBS- "obs.ap-southeast-2.myhuaweicloud.com"- AP-Bangkok- Provider: HuaweiOBS- "obs.ap-southeast-3.myhuaweicloud.com"- AP-Singapore- Provider: HuaweiOBS- - "obs.cn-east-3.myhuaweicloud.com"- - CN East-Shanghai1+ - "obs.ap-southeast-4.myhuaweicloud.com"+ - AP-Jakarta+ - Provider: HuaweiOBS+ - "obs.ap-southeast-5.myhuaweicloud.com"+ - AP-Manila- Provider: HuaweiOBS- "obs.cn-east-2.myhuaweicloud.com"- CN East-Shanghai2- Provider: HuaweiOBS+ - "obs.cn-east-3.myhuaweicloud.com"+ - CN East-Shanghai1+ - Provider: HuaweiOBS+ - "obs.cn-east-4.myhuaweicloud.com"+ - CN East-Qingdao+ - Provider: HuaweiOBS+ - "obs.cn-east-5.myhuaweicloud.com"+ - CN East-Suzhou+ - Provider: HuaweiOBS- "obs.cn-north-1.myhuaweicloud.com"- CN North-Beijing1- Provider: HuaweiOBS+ - "obs.cn-north-2.myhuaweicloud.com"+ - CN North-Beijing2+ - Provider: HuaweiOBS- "obs.cn-north-4.myhuaweicloud.com"- CN North-Beijing4- Provider: HuaweiOBS+ - "obs.cn-north-9.myhuaweicloud.com"+ - CN North-Ulanqab+ - Provider: HuaweiOBS+ - "obs.cn-north-11.myhuaweicloud.com"+ - CN North-Hebei+ - Provider: HuaweiOBS+ - "obs.cn-north-12.myhuaweicloud.com"+ - CN North-Hohhot+ - Provider: HuaweiOBS- "obs.cn-south-1.myhuaweicloud.com"- CN South-Guangzhou- Provider: HuaweiOBS- - "obs.ap-southeast-1.myhuaweicloud.com"- - CN-Hong Kong+ - "obs.cn-south-2.myhuaweicloud.com"+ - CN South-Shenzhen- Provider: HuaweiOBS- - "obs.sa-argentina-1.myhuaweicloud.com"- - LA-Buenos Aires1+ - "obs.cn-south-4.myhuaweicloud.com"+ - CN South-Guangzhou-InvitationOnly- Provider: HuaweiOBS- - "obs.sa-peru-1.myhuaweicloud.com"- - LA-Lima1+ - "obs.cn-southwest-2.myhuaweicloud.com"+ - CN Southwest-Guiyang1+ - Provider: HuaweiOBS+ - "obs.cn-southwest-3.myhuaweicloud.com"+ - CN Southwest-Chongqing+ - Provider: HuaweiOBS+ - "obs.eu-west-101.myhuaweicloud.eu"+ - EU-Dublin+ - Provider: HuaweiOBS+ - "obs.la-north-2.myhuaweicloud.com"+ - LA-Mexico City2+ - Provider: HuaweiOBS+ - "obs.la-south-2.myhuaweicloud.com"+ - LA-Santiago+ - Provider: HuaweiOBS+ - "obs.me-east-1.myhuaweicloud.com"+ - ME-Riyadh- Provider: HuaweiOBS- "obs.na-mexico-1.myhuaweicloud.com"- LA-Mexico City1- Provider: HuaweiOBS- - "obs.sa-chile-1.myhuaweicloud.com"- - LA-Santiago2+ - "obs.ru-northwest-2.myhuaweicloud.com"+ - RU-Moscow2+ - Provider: HuaweiOBS+ - "obs.sa-argentina-1.myhuaweicloud.com"+ - LA-Buenos Aires1- Provider: HuaweiOBS- "obs.sa-brazil-1.myhuaweicloud.com"- LA-Sao Paulo1- Provider: HuaweiOBS- - "obs.ru-northwest-2.myhuaweicloud.com"- - RU-Moscow2+ - "obs.sa-chile-1.myhuaweicloud.com"+ - LA-Santiago2+ - Provider: HuaweiOBS+ - "obs.sa-peru-1.myhuaweicloud.com"+ - LA-Lima1+ - Provider: HuaweiOBS+ - "obs.tr-west-1.myhuaweicloud.com"+ - TR-Istanbul- Provider: HuaweiOBS- "s3.us.cloud-object-storage.appdomain.cloud"- US Cross Region Endpoint@@ -32073,6 +33593,18 @@ Properties:- "us-east-1.storage.impossibleapi.net"- New York, USA- Provider: ImpossibleCloud+ - "s3.swiss-backup01.infomaniak.com"+ - Swiss Backup Location 01+ - Provider: Infomaniak+ - "s3.swiss-backup02.infomaniak.com"+ - Swiss Backup Location 02+ - Provider: Infomaniak+ - "s3.swiss-backup03.infomaniak.com"+ - Swiss Backup Location 03+ - Provider: Infomaniak+ - "s3.swiss-backup04.infomaniak.com"+ - Swiss Backup Location 04+ - Provider: Infomaniak- "de-fra.i3storage.com"- Frankfurt, Germany- Provider: Intercolo@@ -32177,38 +33709,47 @@ Properties:- "br-ne1.magaluobjects.com"- Fortaleza, CE (BR), br-ne1- Provider: Magalu- - "s3.eu-amsterdam.megas4.com"- - Mega S4 Amsterdam+ - "s3.eu-luxembourg-1.megas4.com"+ - Mega S4 Luxembourg 1+ - Provider: Mega+ - "s3.eu-luxembourg-2.megas4.com"+ - Mega S4 Luxembourg 2- Provider: Mega- - "s3.eu-luxembourg.megas4.com"- - Mega S4 Luxembourg+ - "s3.eu-amsterdam-1.megas4.com"+ - Mega S4 Amsterdam 1- Provider: Mega- - "s3.eu-paris.megas4.com"- - Mega S4 Paris+ - "s3.eu-amsterdam-2.megas4.com"+ - Mega S4 Amsterdam 2- Provider: Mega- - "s3.eu-barcelona.megas4.com"- - Mega S4 Barcelona+ - "s3.eu-paris-1.megas4.com"+ - Mega S4 Paris 1- Provider: Mega- - "s3.ca-montreal.megas4.com"- - Mega S4 Montreal+ - "s3.eu-paris-2.megas4.com"+ - Mega S4 Paris 2- Provider: Mega- - "s3.ca-vancouver.megas4.com"- - Mega S4 Vancouver+ - "s3.eu-barcelona-1.megas4.com"+ - Mega S4 Barcelona 1- Provider: Mega- - "s3.ap-tokyo.megas4.com"- - Mega S4 Tokyo+ - "s3.eu-barcelona-2.megas4.com"+ - Mega S4 Barcelona 2- Provider: Mega- - "s3.eu-central-1.s4.mega.io"- - Mega S4 eu-central-1 (Amsterdam, legacy)+ - "s3.ca-montreal-1.megas4.com"+ - Mega S4 Montreal 1- Provider: Mega- - "s3.eu-central-2.s4.mega.io"- - Mega S4 eu-central-2 (Bettembourg, legacy)+ - "s3.ca-montreal-2.megas4.com"+ - Mega S4 Montreal 2- Provider: Mega- - "s3.ca-central-1.s4.mega.io"- - Mega S4 ca-central-1 (Montreal, legacy)+ - "s3.ca-vancouver-1.megas4.com"+ - Mega S4 Vancouver 1- Provider: Mega- - "s3.ca-west-1.s4.mega.io"- - Mega S4 ca-west-1 (Vancouver, legacy)+ - "s3.ca-vancouver-2.megas4.com"+ - Mega S4 Vancouver 2+ - Provider: Mega+ - "s3.ap-tokyo-1.megas4.com"+ - Mega S4 Tokyo 1+ - Provider: Mega+ - "s3.ap-tokyo-2.megas4.com"+ - Mega S4 Tokyo 2- Provider: Mega- "oos.eu-west-2.outscale.com"- Outscale EU West 2 (Paris)@@ -32375,6 +33916,12 @@ Properties:- "nz.s3.rackcorp.com"- Auckland (New Zealand) Endpoint- Provider: RackCorp+ - "s3-gw1.relaix.net"+ - Aachen-1, Germany+ - Provider: RelAix+ - "s3-gw2.relaix.net"+ - Aachen-2, Germany+ - Provider: RelAix- "s3.nl-ams.scw.cloud"- Amsterdam Endpoint- Provider: Scaleway@@ -32384,6 +33931,9 @@ Properties:- "s3.pl-waw.scw.cloud"- Warsaw Endpoint- Provider: Scaleway+ - "s3.it-mil.scw.cloud"+ - Milan Endpoint+ - Provider: Scaleway- "localhost:8333"- SeaweedFS S3 localhost- Provider: SeaweedFS@@ -32620,7 +34170,7 @@ Properties:- Config: location_constraint- Env Var: RCLONE_S3_LOCATION_CONSTRAINT-- Provider: AWS,ArvanCloud,Ceph,ChinaMobile,DigitalOcean,Dreamhost,Exaba,GCS,HCP,Hetzner,IBMCOS,ImpossibleCloud,LyveCloud,Minio,Netease,Qiniu,Rabata,RackCorp,Scality,SeaweedFS,Synology,Wasabi,Zata,ZeroServices,Other+- Provider: AWS,ArvanCloud,Ceph,ChinaMobile,DigitalOcean,Dreamhost,Exaba,GCS,HCP,Hetzner,IBMCOS,ImpossibleCloud,Infomaniak,LyveCloud,Minio,Netease,Qiniu,Rabata,RackCorp,Scality,SeaweedFS,Synology,Wasabi,Zata,ZeroServices,Other- Type: string- Required: false- Examples:@@ -33001,36 +34551,36 @@ Properties:- Config: acl- Env Var: RCLONE_S3_ACL-- Provider: AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IBMCOS,IDrive,ImpossibleCloud,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,Scaleway,Scality,SeaweedFS,TencentCOS,US3,Wasabi,Zata,ZeroServices,Other+- Provider: AWS,Alibaba,ArvanCloud,BizflyCloud,Cellar,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IBMCOS,IDrive,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,RelAix,Scaleway,Scality,SeaweedFS,TencentCOS,US3,Wasabi,Zata,ZeroServices,Other- Type: string- Required: false- Examples:- "private"- Owner gets FULL_CONTROL.- No one else has access rights (default).- - Provider: AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,Scaleway,Scality,SeaweedFS,US3,Wasabi,Zata,ZeroServices,Other+ - Provider: AWS,Alibaba,ArvanCloud,BizflyCloud,Cellar,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,RelAix,Scaleway,Scality,SeaweedFS,US3,Wasabi,Zata,ZeroServices,Other- "public-read"- Owner gets FULL_CONTROL.- The AllUsers group gets READ access.- - Provider: AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,Scaleway,Scality,SeaweedFS,TencentCOS,US3,Wasabi,Zata,ZeroServices,Other+ - Provider: AWS,Alibaba,ArvanCloud,BizflyCloud,Cellar,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,RelAix,Scaleway,Scality,SeaweedFS,TencentCOS,US3,Wasabi,Zata,ZeroServices,Other- "public-read-write"- Owner gets FULL_CONTROL.- The AllUsers group gets READ and WRITE access.- Granting this on a bucket is generally not recommended.- - Provider: AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,Scaleway,Scality,SeaweedFS,TencentCOS,Wasabi,Zata,ZeroServices,Other+ - Provider: AWS,Alibaba,ArvanCloud,BizflyCloud,Cellar,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,RelAix,Scaleway,Scality,SeaweedFS,TencentCOS,Wasabi,Zata,ZeroServices,Other- "authenticated-read"- Owner gets FULL_CONTROL.- The AuthenticatedUsers group gets READ access.- - Provider: AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,Scaleway,Scality,SeaweedFS,TencentCOS,Wasabi,Zata,ZeroServices,Other+ - Provider: AWS,Alibaba,ArvanCloud,BizflyCloud,Cellar,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,RelAix,Scaleway,Scality,SeaweedFS,TencentCOS,Wasabi,Zata,ZeroServices,Other- "bucket-owner-read"- Object owner gets FULL_CONTROL.- Bucket owner gets READ access.- If you specify this canned ACL when creating a bucket, Amazon S3 ignores it.- - Provider: AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,Scaleway,Scality,SeaweedFS,TencentCOS,Wasabi,Zata,ZeroServices,Other+ - Provider: AWS,Alibaba,ArvanCloud,BizflyCloud,Cellar,Ceph,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,RelAix,Scaleway,Scality,SeaweedFS,TencentCOS,Wasabi,Zata,ZeroServices,Other- "bucket-owner-full-control"- Both the object owner and the bucket owner get FULL_CONTROL over the object.- If you specify this canned ACL when creating a bucket, Amazon S3 ignores it.- - Provider: AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,Scaleway,Scality,SeaweedFS,TencentCOS,Wasabi,Zata,ZeroServices,Other+ - Provider: AWS,Alibaba,ArvanCloud,BizflyCloud,Cellar,Ceph,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,RelAix,Scaleway,Scality,SeaweedFS,TencentCOS,Wasabi,Zata,ZeroServices,Other- "private"- Owner gets FULL_CONTROL.- No one else has access rights (default).@@ -33224,7 +34774,7 @@ Properties:### Advanced options-Here are the Advanced options specific to s3 (Amazon S3 Compliant Storage Providers including AWS, Alibaba, ArvanCloud, BizflyCloud, Ceph, ChinaMobile, Cloudflare, Cubbit, DigitalOcean, Dreamhost, Exaba, Fastly, FileLu, FlashBlade, GCS, HCP, Hetzner, HuaweiOBS, IBMCOS, IDrive, ImpossibleCloud, Intercolo, IONOS, Leviia, Liara, Linode, LyveCloud, Magalu, Mega, Minio, Netease, Outscale, OVHcloud, Petabox, Qiniu, Rabata, RackCorp, Rclone, Scaleway, Scality, SeaweedFS, Selectel, Servercore, SpectraLogic, Storj, Synology, TencentCOS, US3, Wasabi, Zadara, Zata, ZeroServices, Other).+Here are the Advanced options specific to s3 (Amazon S3 Compliant Storage Providers including AWS, Alibaba, ArvanCloud, Autonomys, BizflyCloud, Cellar, Ceph, ChinaMobile, Cloudflare, Cubbit, DigitalOcean, Dreamhost, Exaba, Fastly, FileLu, FlashBlade, GCS, HCP, Hetzner, HuaweiOBS, IBMCOS, IDrive, ImpossibleCloud, Infomaniak, Intercolo, IONOS, Leviia, Liara, Linode, LyveCloud, Magalu, Mega, Minio, Neon, Netease, Outscale, OVHcloud, Petabox, Qiniu, Rabata, RackCorp, Rclone, RelAix, Scaleway, Scality, SeaweedFS, Selectel, Servercore, SpectraLogic, Storj, Synology, TencentCOS, US3, Wasabi, Zadara, Zata, ZeroServices, Other).#### --s3-bucket-acl@@ -33243,7 +34793,7 @@ Properties:- Config: bucket_acl- Env Var: RCLONE_S3_BUCKET_ACL-- Provider: AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IBMCOS,IDrive,ImpossibleCloud,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Mega,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,Scaleway,SeaweedFS,Servercore,TencentCOS,US3,Wasabi,Zata,ZeroServices,Other+- Provider: AWS,Alibaba,ArvanCloud,BizflyCloud,Cellar,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IBMCOS,IDrive,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Mega,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,RelAix,Scaleway,SeaweedFS,Servercore,TencentCOS,US3,Wasabi,Zata,ZeroServices,Other- Type: string- Required: false- Examples:@@ -33281,7 +34831,7 @@ Properties:- Config: sse_customer_algorithm- Env Var: RCLONE_S3_SSE_CUSTOMER_ALGORITHM-- Provider: AWS,Ceph,ChinaMobile,Minio+- Provider: AWS,Cellar,Ceph,ChinaMobile,Minio- Type: string- Required: false- Examples:@@ -33294,13 +34844,13 @@ Properties:To use SSE-C you may provide the secret encryption key used to encrypt/decrypt your data.-Alternatively you can provide --sse-customer-key-base64.+Alternatively you can provide --s3-sse-customer-key-base64.Properties:- Config: sse_customer_key- Env Var: RCLONE_S3_SSE_CUSTOMER_KEY-- Provider: AWS,Ceph,ChinaMobile,Minio+- Provider: AWS,Cellar,Ceph,ChinaMobile,Minio- Type: string- Required: false- Examples:@@ -33311,13 +34861,13 @@ Properties:If using SSE-C you must provide the secret encryption key encoded in base64 format to encrypt/decrypt your data.-Alternatively you can provide --sse-customer-key.+Alternatively you can provide --s3-sse-customer-key.Properties:- Config: sse_customer_key_base64- Env Var: RCLONE_S3_SSE_CUSTOMER_KEY_BASE64-- Provider: AWS,Ceph,ChinaMobile,Minio+- Provider: AWS,Cellar,Ceph,ChinaMobile,Minio- Type: string- Required: false- Examples:@@ -33335,7 +34885,7 @@ Properties:- Config: sse_customer_key_md5- Env Var: RCLONE_S3_SSE_CUSTOMER_KEY_MD5-- Provider: AWS,Ceph,ChinaMobile,Minio+- Provider: AWS,Cellar,Ceph,ChinaMobile,Minio- Type: string- Required: false- Examples:@@ -34425,6 +35975,47 @@ info on how to pass options and arguments.These can be run on a running backend using the rc command[backend/command](https://rclone.org/rc/#backend-command).+### link++Generate a signed link with response header overrides.++```console+rclone backend link remote: [options] [<arguments>+]+```++This command generates a signed download link for one file, with optional+overrides for the HTTP response headers. Pass the file path as a separate+argument, relative to the remote path.++Usage examples:++```console+rclone backend link s3:bucket path/to/file -o expire=1h+rclone backend link s3:bucket path/to/file -o response-expires="Thu, 01 Jan 1970 00:00:00 GMT"+rclone backend link s3:bucket path/to/file -o response-content-disposition='attachment; filename="download.txt"'+```++The link expires after 7 days by default. Use `-o expire=1h` to change+its lifetime. Durations must be at least 1 second; values above 7 days are+limited to 7 days, as with `rclone link`.++The `response-expires` option sets the HTTP Expires response header,+not the lifetime of the signed link. It must be an HTTP date, such as+`Thu, 01 Jan 1970 00:00:00 GMT`.++The overrides are included in the signature and must not be changed in the+returned URL. They do not modify the object's stored metadata.++Options:++- "expire": How long the link will be valid (default 7d, maximum 7d).+- "response-cache-control": Set the Cache-Control response header.+- "response-content-disposition": Set the Content-Disposition response header.+- "response-content-encoding": Set the Content-Encoding response header.+- "response-content-language": Set the Content-Language response header.+- "response-content-type": Set the Content-Type response header.+- "response-expires": Set the Expires response header to an HTTP date.+### restoreRestore objects from GLACIER or INTELLIGENT-TIERING archive tier.@@ -34725,15 +36316,15 @@ section above.From rclone v1.69 [Directory Buckets](https://docs.aws.amazon.com/AmazonS3/latest/userguide/directory-buckets-overview.html)are supported.-You will need to set the `directory_buckets = true` config parameter-or use `--s3-directory-buckets`.+You will need to set the `directory_bucket = true` config parameter+or use `--s3-directory-bucket`.Note that rclone cannot yet:- Create directory buckets- List directory buckets-See [the --s3-directory-buckets flag](#s3-directory-buckets) for more info+See [the --s3-directory-bucket flag](#s3-directory-bucket) for more info### AWS Snowball Edge@@ -34984,6 +36575,97 @@ server_side_encryption =storage_class =```+### Autonomys Auto Drive {#autonomys}++[Auto Drive](https://ai3.storage/) is permanent storage on the Autonomys+Network, with an S3-compatible API. Deleting an object hides it, but its+data and every earlier version are kept. Buckets are created by the first+upload into them.++Use your Auto Drive API key as the `access_key_id`, and any value as the+`secret_access_key`.++Here is an example of making a configuration. First run:++```console+rclone config+```++This will guide you through an interactive setup process.++```text+No remotes found, make a new one?+n) New remote+s) Set configuration password+q) Quit config+n/s/q> n++Enter name for new remote.+name> autodrive++Option Storage.+Type of storage to configure.+Storage> s3++Option provider.+Choose your S3 provider.+provider> Autonomys++Option env_auth.+Get AWS credentials from runtime (environment variables or EC2/ECS meta data if no env vars).+Only applies if access_key_id and secret_access_key is blank.+env_auth> false++Option access_key_id.+AWS Access Key ID.+access_key_id> YOUR_API_KEY++Option secret_access_key.+AWS Secret Access Key (password).+secret_access_key> ANY_VALUE++Option endpoint.+Endpoint for S3 API.+Choose a number from below, or type in your own value.+ 1 / Auto Drive mainnet S3 endpoint+ \ (s3.auto-drive.autonomys.xyz)+endpoint> 1++Option bucket_object_lock_enabled.+Enable Object Lock when creating new buckets.+Enter a boolean value (true or false). Press Enter for the default (false).+bucket_object_lock_enabled>++Edit advanced config?+y) Yes+n) No (default)+y/n> n++Configuration complete.+Options:+- type: s3+- provider: Autonomys+- access_key_id: YOUR_API_KEY+- secret_access_key: ANY_VALUE+- endpoint: s3.auto-drive.autonomys.xyz+Keep this "autodrive" remote?+y) Yes this is OK (default)+e) Edit this remote+d) Delete this remote+y/e/d> y+```++The resulting configuration file should look like:++```ini+[autodrive]+type = s3+provider = Autonomys+access_key_id = YOUR_API_KEY+secret_access_key = ANY_VALUE+endpoint = s3.auto-drive.autonomys.xyz+```+### BizflyCloud {#bizflycloud}[Bizfly Cloud Simple Storage](https://bizflycloud.vn/simple-storage) is an@@ -35045,6 +36727,12 @@ parameter `--s3-upload-cutoff 0` or put this in the config file as`upload_cutoff 0` to work around a bug which causes uploading of smallfiles to fail.+Rclone defaults `sign_accept_encoding` to false for the Ceph provider.+That avoids `SignatureDoesNotMatch` errors when a reverse proxy (for+example Cloudflare) rewrites the `Accept-Encoding` request header. You+can override this with `--s3-sign-accept-encoding` / `sign_accept_encoding`+if needed.+Note also that Ceph sometimes puts `/` in the passwords it givesusers. If you read the secret access key using the command line toolsyou will get a JSON blob with the `/` escaped as `\/`. Make sure you@@ -35072,7 +36760,7 @@ use the secret key as `xxxxxx/xxxx` it will work fine.### China Mobile Ecloud Elastic Object Storage (EOS) {#china-mobile-ecloud-eos}-Here is an example of making an [China Mobile Ecloud Elastic Object Storage (EOS)](https:///ecloud.10086.cn/home/product-introduction/eos/)+Here is an example of making an [China Mobile Ecloud Elastic Object Storage (EOS)](https://ecloud.10086.cn/home/product-introduction/eos/)configuration. First run:```console@@ -35322,6 +37010,117 @@ d) Delete this remotey/e/d> y```+### Clever Cloud Cellar {#clever-cloud-cellar}++[Cellar](https://www.clever.cloud/product/cellar-object-storage/) is the+S3-compatible object storage service of+[Clever Cloud](https://www.clever.cloud/), hosted in France. Credentials and the+endpoint are shown in the Cellar add-on dashboard, or in the+`CELLAR_ADDON_KEY_ID`, `CELLAR_ADDON_KEY_SECRET` and `CELLAR_ADDON_HOST`+environment variables of applications linked to it.++Cellar does not apply Object Lock headers sent when initiating a multipart+upload. When setting Object Lock retention or legal hold options, also set+`--s3-object-lock-set-after-upload`: the lock is then applied with separate API+calls after the upload, so the object is not protected by these settings until+those calls succeed.++Cellar supports server-side encryption with customer-provided keys (SSE-C), set+with `sse_customer_algorithm` and `sse_customer_key_base64`. It rejects+server-side copies of SSE-C objects unless they are done as multipart copies, so+use `--no-update-modtime` to avoid modification time updates, and expect+server-side moves and renames of encrypted objects smaller than+`--s3-copy-cutoff` to fail. See the+[Cellar SSE-C documentation](https://www.clever.cloud/developers/doc/deploy/storage/cellar/#server-side-encryption-with-customer-keys-sse-c).++First run:++```console+rclone config+```++```text+This will guide you through an interactive setup process:++n) New remote+n/s/q> n++Enter name for new remote.+name> cellar++Option Storage.+Type of storage to configure.+Choose a number from below, or type in your own value.+[snip]+XX / Amazon S3 Compliant Storage Providers including AWS, ..., Cellar, ... and others+ \ (s3)+Storage> s3++Option provider.+Choose your S3 provider.+[snip]+XX / Clever Cloud Cellar Object Storage+ \ (Cellar)+provider> Cellar++Option env_auth.+Get AWS credentials from runtime (environment variables or EC2/ECS meta data if no env vars).+ 1 / Enter AWS credentials in the next step.+ \ (false)+ 2 / Get AWS credentials from the environment (env vars or IAM).+ \ (true)+env_auth> 1++Option access_key_id.+AWS Access Key ID.+access_key_id> ACCESS_KEY++Option secret_access_key.+AWS Secret Access Key (password).+secret_access_key> SECRET_KEY++Option endpoint.+Endpoint for S3 API.+ 1 / Paris, France+ \ (cellar-c2.services.clever-cloud.com)+endpoint> 1++Option acl.+Canned ACL used when creating buckets and storing or copying objects.+[snip]+acl> 1++Option bucket_object_lock_enabled.+Enable Object Lock when creating new buckets.+bucket_object_lock_enabled>++Edit advanced config?+y/n> n++Configuration complete.+Options:+- type: s3+- provider: Cellar+- access_key_id: ACCESS_KEY+- secret_access_key: SECRET_KEY+- endpoint: cellar-c2.services.clever-cloud.com+- acl: private+Keep this "cellar" remote?+y/e/d> y+```++This will leave the config file looking like this.++```ini+[cellar]+type = s3+provider = Cellar+access_key_id = ACCESS_KEY+secret_access_key = SECRET_KEY+endpoint = cellar-c2.services.clever-cloud.com+acl = private+```+### Cloudflare R2 {#cloudflare-r2}[Cloudflare R2](https://blog.cloudflare.com/r2-open-beta/) Storage@@ -36664,6 +38463,114 @@ endpoint = eu-central-2.storage.impossibleapi.netacl = private```+### Infomaniak Swiss Backup {#infomaniak}++[Infomaniak Swiss Backup](https://www.infomaniak.com/en/swiss-backup/cloud-storage) is a cloud backup space in Switzerland that you can use however you want with the script or software of your choice that is compatible with Swift, S3 or SFTP.+++#### Prerequisites++Create and setup your Swiss Backup product following [this documentation](https://www.infomaniak.com/en/support/faq/2546/create-an-s3-location-on-swiss-backup). Memorize the generated `Endpoint`, `Access Key`, and `Secret Key`:++#### Configuration++To configure access to an S3 location on Swiss Backup, follow the steps below:++1. Run `rclone config` and select `n` for a new remote.++ ```sh+ rclone config+ No remotes found, make a new one?+ n) New remote+ s) Set configuration password+ q) Quit config+ n/s/q> n+ ```++2. Give the name of the configuration. For example, name it 'swissbackup'.++ ```sh+ name> swissbackup+ ```++3. Select `s3` storage.++ ```sh+ Type of storage to configure.+ XX / Amazon S3 Compliant Storage Providers including AWS, ... - Tier 1+ \ (s3)+ Storage> s3+ ```++4. Select `Infomaniak` provider.++ ```sh+ Choose your S3 provider.+ xx / Infomaniak Swiss Backup+ \ (Infomaniak)+ provider> Infomaniak+ ```++5. Enter your Access Key and Secret Key. _You can generate new ones by clicking 'Generate new keys' on your Swiss Backup product dashboard._++ ```sh+ Option env_auth.+ 1 / Enter AWS credentials in the next step+ \ (false)+ 2 / Get AWS credentials from the environment (env vars or IAM)+ \ (true)+ env_auth> 1++ AWS Access Key ID.+ access_key_id> {YOUR-ACCESS-ID}++ AWS Secret Access Key (password)+ secret_access_key> {YOUR-ACCESS-SECRET}+ ```++6. Select `us-east-1` region for compatibility.++ ```sh+ Region to connect to.+ / Default region for compatibility.+ 1 | All data is stored in Switzerland.+ \ (us-east-1)+ region> 1+ ```++7. Select the endpoint corresponding to the one generated with your Access and Secret Keys.++ ```sh+ Endpoint for S3 API.+ 1 / Swiss Backup Location 01+ \ (s3.swiss-backup01.infomaniak.com)+ 2 / Swiss Backup Location 02+ \ (s3.swiss-backup02.infomaniak.com)+ 3 / Swiss Backup Location 03+ \ (s3.swiss-backup03.infomaniak.com)+ 4 / Swiss Backup Location 04+ \ (s3.swiss-backup04.infomaniak.com)+ endpoint> 1+ ```++8. Adjust other parameters and advanced config at your convenience. Leave blank if not sure.++9. You should obtain something like this.++ ```sh+ Configuration complete.+ Options:+ - type: s3+ - provider: Infomaniak+ - access_key_id: xxxxxxxxxx+ - secret_access_key: xxxxxxxxxx+ - region: us-east-1+ - endpoint: s3.swiss-backup01.infomaniak.com+ ```++10. Press `y` to validate. Then `q` to quit.++### Intercolo Object Storage {#intercolo}[Intercolo Object Storage](https://intercolo.de/object-storage) offers@@ -37363,6 +39270,10 @@ secret_access_key = SECRET_ACCESS_KEYendpoint = eu-central-1.linodeobjects.com```+Linode Object Storage is Ceph-backed. Rclone defaults+`sign_accept_encoding` to false for this provider so SigV4 requests do+not fail with `SignatureDoesNotMatch` (see the [Ceph](#ceph) section).+### Magalu {#magalu}Here is an example of making a [Magalu Object Storage](https://magalu.cloud/object-storage/)@@ -37663,6 +39574,157 @@ So once set up, for example, to copy files into a bucketrclone copy /path/to/files minio:bucket```+### Neon Object Storage {#neon}++[Neon Object Storage](https://neon.com/docs/storage/overview) is S3+compatible object storage in a Neon project. Each branch of the project+has its own copy of the buckets and its own S3 endpoint. The endpoint+looks like `https://<branch_id>.storage.<cell>.<region>.aws.neon.tech`.++In the Neon Console, click **Connect**, open the **Storage** tab and+choose **Parameters only**. Use `AWS_ACCESS_KEY_ID`,+`AWS_SECRET_ACCESS_KEY`, `AWS_ENDPOINT_URL_S3` and `AWS_REGION` as+`access_key_id`, `secret_access_key`, `endpoint` and `region`. To make+a separate credential, use **Settings** > **Credentials** or run+`neon credentials create --scope storage:read --scope storage:write`.++A credential works on the branch it was made on and on that branch's+child branches. To use a child branch, change only `endpoint`.++Neon also has these limits (see the+[S3 compatibility docs](https://neon.com/docs/storage/s3-compatibility)):++- The maximum object size is 5 GiB for a single upload and 10 GiB for+ a multipart upload.+- Neon rejects object keys containing control characters with+ `400 InvalidURI`. To store files with these names, add `Ctl,Del` to+ the encoding, for example `encoding = Slash,InvalidUtf8,Dot,Ctl,Del`.+- Neon doesn't support ACLs over the S3 API. Leave `acl` and+ `bucket_acl` empty and set bucket access (`private` or `public_read`)+ in Neon.+- Neon stores versioning and lifecycle configurations but doesn't+ enforce them.+- Neon may throttle heavy use with `503 SlowDown` errors. rclone retries+ these.++Here is an example of making a Neon Object Storage configuration. First run:++```console+rclone config+```++This will guide you through an interactive setup process.++```text+No remotes found, make a new one?+n) New remote+s) Set configuration password+q) Quit config+n/s/q> n++Enter name for new remote.+name> neon++Option Storage.+Type of storage to configure.+Choose a number from below, or type in your own value.+[snip]+XX / Amazon S3 Compliant Storage Providers including AWS, ... Neon, ...+ \ (s3)+[snip]+Storage> s3++Option provider.+Choose your S3 provider.+Choose a number from below, or type in your own value.+Press Enter to leave empty.+[snip]+XX / Neon Object Storage+ \ (Neon)+[snip]+provider> Neon++Option env_auth.+Get AWS credentials from runtime (environment variables or EC2/ECS meta data if no env vars).+Only applies if access_key_id and secret_access_key is blank.+Choose a number from below, or type in your own boolean value (true or false).+Press Enter for the default (false).+ 1 / Enter AWS credentials in the next step.+ \ (false)+ 2 / Get AWS credentials from the environment (env vars or IAM).+ \ (true)+env_auth>++Option access_key_id.+AWS Access Key ID.+Leave blank for anonymous access or runtime credentials.+Enter a value. Press Enter to leave empty.+access_key_id> NEON_ACCESS_KEY_ID++Option secret_access_key.+AWS Secret Access Key (password).+Leave blank for anonymous access or runtime credentials.+Enter a value. Press Enter to leave empty.+secret_access_key> NEON_SECRET_ACCESS_KEY++Option region.+Region to connect to.+Leave blank if you are using an S3 clone and you don't have a region.+Choose a number from below, or type in your own value.+Press Enter to leave empty.+ 1 / US East (N. Virginia)+ \ (us-east-1)+ 2 / US East (Ohio)+ \ (us-east-2)+ 3 / Europe (Frankfurt)+ \ (eu-central-1)+ 4 / Asia Pacific (Singapore)+ \ (ap-southeast-1)+region> us-east-2++Option endpoint.+Endpoint for S3 API.+Required when using an S3 clone.+Enter a value. Press Enter to leave empty.+endpoint> https://<branch_id>.storage.<cell>.<region>.aws.neon.tech++Option bucket_object_lock_enabled.+Enable Object Lock when creating new buckets.+Enter a boolean value (true or false). Press Enter for the default (false).+bucket_object_lock_enabled>++Edit advanced config?+y) Yes+n) No (default)+y/n> n++Configuration complete.+Options:+- type: s3+- provider: Neon+- access_key_id: NEON_ACCESS_KEY_ID+- secret_access_key: NEON_SECRET_ACCESS_KEY+- region: us-east-2+- endpoint: https://<branch_id>.storage.<cell>.<region>.aws.neon.tech+Keep this "neon" remote?+y) Yes this is OK (default)+e) Edit this remote+d) Delete this remote+y/e/d> y+```++This will leave the config file looking like this.++```ini+[neon]+type = s3+provider = Neon+access_key_id = NEON_ACCESS_KEY_ID+secret_access_key = NEON_SECRET_ACCESS_KEY+region = us-east-2+endpoint = https://<branch_id>.storage.<cell>.<region>.aws.neon.tech+```+### Netease NOSFor Netease NOS configure as per the configurator `rclone config`@@ -38807,6 +40869,134 @@ use_multipart_uploads = falseNote that setting `use_multipart_uploads = false` is to work around[a bug](https://rclone.org/commands/rclone_serve_s3/#bugs) which will be fixed in due course.+### RelAix.Storage {#relaix}++RelAix.Storage is S3 compatible object storage from+[RelAix Networks](https://www.relaix.net/), operated in its own data+centres in Aachen, Germany.+There are two locations, Aachen-1 and Aachen-2, each with its own endpoint.+Buckets and data stay at the location of the endpoint you use.++You get an access key and a secret key from RelAix Networks. Then run:++```console+rclone config+```++This will guide you through an interactive setup process.++```text+No remotes found, make a new one?+n) New remote+s) Set configuration password+q) Quit config+n/s/q> n++Enter name for new remote.+name> my-relaix++Option Storage.+Type of storage to configure.+Choose a number from below, or type in your own value.+[snip]+XX / Amazon S3 Compliant Storage Providers including AWS, ...+ \ (s3)+[snip]+Storage> s3++Option provider.+Choose your S3 provider.+Choose a number from below, or type in your own value.+Press Enter to leave empty.+[snip]+XX / RelAix.Storage+ \ (RelAix)+[snip]+provider> RelAix++Option env_auth.+Get AWS credentials from runtime (environment variables or EC2/ECS meta data if no env vars).+Only applies if access_key_id and secret_access_key is blank.+Choose a number from below, or type in your own boolean value (true or false).+Press Enter for the default (false).+ 1 / Enter AWS credentials in the next step.+ \ (false)+ 2 / Get AWS credentials from the environment (env vars or IAM).+ \ (true)+env_auth>++Option access_key_id.+AWS Access Key ID.+Leave blank for anonymous access or runtime credentials.+Enter a value. Press Enter to leave empty.+access_key_id> ACCESS_KEY++Option secret_access_key.+AWS Secret Access Key (password).+Leave blank for anonymous access or runtime credentials.+Enter a value. Press Enter to leave empty.+secret_access_key> SECRET_KEY++Option endpoint.+Endpoint for S3 API.+Required when using an S3 clone.+Choose a number from below, or type in your own value.+Press Enter to leave empty.+ 1 / Aachen-1, Germany+ \ (s3-gw1.relaix.net)+ 2 / Aachen-2, Germany+ \ (s3-gw2.relaix.net)+endpoint> 1++Option acl.+Canned ACL used when creating buckets and storing or copying objects.+[snip]+acl>++Option bucket_object_lock_enabled.+Enable Object Lock when creating new buckets.+Enter a boolean value (true or false). Press Enter for the default (false).+bucket_object_lock_enabled>++Edit advanced config?+y) Yes+n) No (default)+y/n> n++Configuration complete.+Options:+- type: s3+- provider: RelAix+- access_key_id: ACCESS_KEY+- secret_access_key: SECRET_KEY+- endpoint: s3-gw1.relaix.net+Keep this "my-relaix" remote?+y) Yes this is OK (default)+e) Edit this remote+d) Delete this remote+y/e/d> y+```++This will leave the config file looking like this.++```ini+[my-relaix]+type = s3+provider = RelAix+access_key_id = ACCESS_KEY+secret_access_key = SECRET_KEY+endpoint = s3-gw1.relaix.net+```++Object Lock can only be enabled when a bucket is created. To create buckets+with Object Lock, set `bucket_object_lock_enabled = true` before running+`rclone mkdir`.++To lock objects on upload with `object_lock_mode`, also set+`object_lock_set_after_upload = true`. RelAix.Storage does not yet apply+Object Lock headers sent with multipart uploads, so rclone needs to set the+lock in a separate call after the upload.+### Scaleway[Scaleway](https://www.scaleway.com/object-storage/) The Object Storage platform@@ -40708,6 +42898,25 @@ Rclone works fastest with large squashfs block sizes. For example:mksquashfs 100files 100files.sqfs -comp zstd -b 1M```+## Symlinks++Symlinks stored in an archive are skipped unless the `-l/--links` flag+is supplied, as they are on the other backends. With the flag they are+listed with a `.rclonelink` suffix and read back as the path they point+at, the same convention the [local](https://rclone.org/local/#symlinks-junction-points)+backend uses, so they can be copied back as symlinks:++```+$ rclone lsl --links :archive:/tmp/test.zip+ 5 2026-09-28 14:16:59.000000000 src/file.txt+ 8 2026-09-28 14:16:59.000000000 src/link.txt.rclonelink+$ rclone cat --links :archive:/tmp/test.zip/src/link.txt.rclonelink+file.txt+$ rclone copy --links :archive:/tmp/test.zip /tmp/out+$ ls -l /tmp/out/src/link.txt+lrwxr-xr-x 1 user user 8 Sep 28 14:16 /tmp/out/src/link.txt -> file.txt+```+## LimitationsFiles in the archive backend are read only. It isn't possible to@@ -41362,6 +43571,11 @@ If the custom endpoint rewrites the requests for authentication,e.g., in Cloudflare Workers, this header needs to be handled properly.Leave blank if you want to use the endpoint provided by Backblaze.+Rclone will not follow a redirect from this endpoint which downgrades+HTTPS to HTTP, and will not send the "Authorization" header or the+SSE-C key headers once a redirect has left the host and port of this+endpoint.+The URL provided here SHOULD have the protocol and SHOULD NOT havea trailing slash or specify the /file/bucket subpath as rclone willrequest files with "{download_url}/file/{bucket_name}/{path}".@@ -41458,7 +43672,7 @@ Properties:To use SSE-C, you may provide the secret encryption key encoded in a UTF-8 compatible string to encrypt/decrypt your data-Alternatively you can provide --sse-customer-key-base64.+Alternatively you can provide --b2-sse-customer-key-base64.Properties:@@ -41474,7 +43688,7 @@ Properties:To use SSE-C, you may provide the secret encryption key encoded in Base64 format to encrypt/decrypt your data-Alternatively you can provide --sse-customer-key.+Alternatively you can provide --b2-sse-customer-key.Properties:@@ -41770,9 +43984,10 @@ have just set.### Invalid refresh token-According to the [box docs](https://developer.box.com/v2.0/docs/oauth-20#section-6-using-the-access-and-refresh-tokens):+According to the [box docs](https://developer.box.com/guides/authentication/tokens/refresh/):-> Each refresh_token is valid for one use in 60 days.+> A Refresh Token is valid for 60 days and can be used to obtain a new Access+> Token and Refresh Token only once.This means that if you@@ -44801,6 +47016,15 @@ This means that- filenames with the same name will encrypt the same- filenames which start the same won't have a common prefix+A version string of the form `-vYYYY-MM-DD-HHMMSS-NNN` on the end of a+file name (as added by `--b2-versions` / `--s3-versions`) is left in+plain text so that versioned files can be found. Directory names are+encrypted in full. Rclone before v1.76 left such a suffix in plain+text on directory names too, so a directory named like this created by+an older rclone will appear in listings with a warning but can't be+opened or removed until renamed on the underlying remote to the name+given in the warning.+This uses a 32 byte key (256 bits) and a 16 byte (128 bits) IV both ofwhich are derived from the user password.@@ -45423,6 +47647,287 @@ It doesn't return anything.<!-- autogenerated options stop -->+# dosya.dev++[dosya.dev](https://dosya.dev) is a cloud file storage, sharing, and+synchronization platform with workspace-based multi-tenancy, file+versioning, and team collaboration features.++## Configuration++Here is an example of how to make a remote called `remote`. First run:++```console+rclone config+```++This will guide you through an interactive setup process:++```text+No remotes found, make a new one?+n) New remote+s) Set configuration password+q) Quit config+n/s/q> n++Enter name for new remote.+name> remote++Option Storage.+Type of storage to configure.+Choose a number from below, or type in your own value.+XX / dosya.dev+ \ (dosya)+Storage> dosya++Option api_key.+Your API Key, get it from https://dosya.dev/profile under "API keys".+Enter a value.+api_key> dos_YOURAPIKEY++Option workspace_id.+Your workspace ID.++Leave this blank to use the one workspace the API key can reach, which+is what a key restricted to a single workspace does. If the key can+reach several, rclone will refuse and list their IDs for you to choose+from.+Enter a value. Press Enter to leave empty.+workspace_id>++Edit advanced config?+y) Yes+n) No (default)+y/n> n++Configuration complete.+Options:+- type: dosya+- api_key: dos_YOURAPIKEY+Keep this "remote" remote?+y) Yes this is OK (default)+e) Edit this remote+d) Delete this remote+y/e/d> y+```++Once configured you can then use `rclone` like this (replace `remote`+with the name you gave your remote):++List directories and files in the top level of your dosya workspace++```console+rclone lsf remote:+```++To copy a local directory to a dosya directory called backup++```console+rclone copy /home/source remote:backup+```++The dosya backend uploads large files in parts and stores them as single+files that remain accessible through the web app and other clients. If you+previously used a WebDAV remote with a chunker overlay for large uploads,+you can use this backend without the overlay.++### Getting your API key++1. Log in to [dosya.dev](https://dosya.dev)+2. Go to [Profile](https://dosya.dev/profile) > **API keys**+3. Create a new API key+4. Copy the key (starts with `dos_`)++A key can be restricted to a single workspace when you create it, which+is the simplest thing to do for rclone: `workspace_id` can then be left+out of the config entirely.++### Workspaces++Leave `workspace_id` blank and rclone asks which workspaces the key can+reach. A key restricted to one workspace, and an account that has only+one, both resolve without any configuration. If the key can reach+several, rclone refuses and lists their IDs so you can set the one you+want:++```text+this API key can reach 2 workspaces, so workspace_id must be set to one of:+ ws_AAAAAAAA (Work)+ ws_BBBBBBBB (Personal)+```++To access several workspaces at once, create a separate remote for each:++```console+rclone config create work dosya api_key=dos_XXX workspace_id=ws_WORK+rclone config create personal dosya api_key=dos_XXX workspace_id=ws_PERSONAL+```++Then use them independently:++```console+rclone ls work:+rclone copy work:docs/ personal:backup/+```++### Modification times and hashes++dosya.dev supports modification times with a resolution of 1 second. The+modification time is sent with the upload, so a copied or synced file keeps+its original timestamp. Modification times cannot be changed after upload+without re-uploading the file.++dosya.dev supports SHA-256 checksums. The server computes the hash of a+single-shot (small file) upload from the bytes it receives, so `rclone check`+and hash-based sync work without downloading. Large files uploaded in multiple+parts have no server-side hash (the parts are never re-read to hash the whole+object), the same limitation the S3 backend has for multipart objects; those+objects report an empty hash.++### Restricted filename characters++In addition to the [default restricted characters set](https://rclone.org/overview/#restricted-characters)+the following characters are also replaced:++| Character | Value | Replacement |+| --------- |:-----:|:-----------:|+| " | 0x22 | " |+| ' | 0x27 | ' |+| < | 0x3C | < |+| > | 0x3E | > |+| \ | 0x5C | \ |++Invalid UTF-8 bytes will also be [replaced](https://rclone.org/overview/#invalid-utf8),+as they can't be used in JSON strings.++### Public Links++dosya.dev supports `rclone link` to create public share links to+files. You can use the `--expire` flag to specify how long the link+should be valid.++```console+rclone link remote:path/to/file.txt+rclone link --expire 7d remote:path/to/file.txt+```++<!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/dosya/dosya.go and run make backenddocs to verify --> <!-- markdownlint-disable-line line-length -->+### Standard options++Here are the Standard options specific to dosya (dosya.dev).++#### --dosya-api-key++Your API Key, get it from https://dosya.dev/profile under "API keys".++Properties:++- Config: api_key+- Env Var: RCLONE_DOSYA_API_KEY+- Type: string+- Required: true++#### --dosya-workspace-id++Your workspace ID.++Leave this blank to use the one workspace the API key can reach, which+is what a key restricted to a single workspace does. If the key can+reach several, rclone will refuse and list their IDs for you to choose+from.++Properties:++- Config: workspace_id+- Env Var: RCLONE_DOSYA_WORKSPACE_ID+- Type: string+- Required: false++### Advanced options++Here are the Advanced options specific to dosya (dosya.dev).++#### --dosya-api-url++API base URL.++Properties:++- Config: api_url+- Env Var: RCLONE_DOSYA_API_URL+- Type: string+- Default: "https://api.dosya.dev"++#### --dosya-upload-concurrency++Concurrency for multipart uploads.++This is the number of parts of the same file that are uploaded+concurrently. The first part is always uploaded on its own, because+the server creates the upload when it receives it.++Each part in flight is held in memory, so memory use grows with+this value times the part size the server chooses.++If you are uploading large files over high-speed links and these+uploads do not fully utilize your bandwidth, then increasing this+may help to speed up the transfers.++Properties:++- Config: upload_concurrency+- Env Var: RCLONE_DOSYA_UPLOAD_CONCURRENCY+- Type: int+- Default: 4++#### --dosya-encoding++The encoding for the backend.++See the [encoding section in the overview](https://rclone.org/overview/#encoding) for more info.++Properties:++- Config: encoding+- Env Var: RCLONE_DOSYA_ENCODING+- Type: Encoding+- Default: Slash,LtGt,DoubleQuote,SingleQuote,BackSlash,Del,Ctl,LeftSpace,RightSpace,InvalidUtf8,Dot++#### --dosya-description++Description of the remote.++Properties:++- Config: description+- Env Var: RCLONE_DOSYA_DESCRIPTION+- Type: string+- Required: false++<!-- autogenerated options stop -->++## Limitations++dosya.dev only supports filenames up to 255 characters in length.++Empty files (0 bytes) cannot be uploaded to dosya.dev.++Upload requires knowing the file size in advance, so streaming+uploads of unknown-size files are not supported.++Each workspace limits the number of uploads a user can have in flight+at once (5 by default, configurable in the workspace settings). When+`--transfers` is set higher than that limit, rclone waits for a free+upload slot rather than failing the transfer, retrying every 5 seconds+for up to 10 minutes.++### Deleting files and directories++rclone deletes files and directories permanently, in a single request+(`DELETE ?permanent=1`), bypassing the workspace trash. Removing a+directory also asks the server whether it is empty (`?if_empty=1`)+rather than listing it first.+# Drime[Drime](https://drime.cloud/) is a cloud storage and transfer service focused@@ -47358,6 +49863,34 @@ excess files in the directory.rclone sync --interactive /home/local/directory remote:dir```+### Restricted filename characters++In addition to the [default restricted characters set](https://rclone.org/overview/#restricted-characters)+the following characters are also replaced:++| Character | Value | Replacement |+| --------- |:-----:|:-----------:|+| \ | 0x5C | \ |++File names can also not end with the following characters.+These only get replaced if they are the last character in the name:++| Character | Value | Replacement |+| --------- |:-----:|:-----------:|+| SP | 0x20 | ␠ |+| LF | 0x0A | ␊ |+| CR | 0x0D | ␍ |+| HT | 0x09 | ␉ |+| VT | 0x0B | ␋ |++Invalid UTF-8 bytes will also be [replaced](https://rclone.org/overview/#invalid-utf8).++Files.com treats the replacements `/` and `\` as if they were `/`+and `\`, and the names `.` and `..` as if they were `.` and `..`,+and rejects them with `Invalid path`. This means that files with `/`+or `\` in their names, or which are called `.` or `..`, can't be+stored on Files.com.+### HashesIn December 2024 `files.com` started [supporting more checksums](https://www.files.com/blog/2024/11/01/new-modern-checksum-options-now-available-with-opt).@@ -47569,7 +50102,7 @@ string as a password, although it is common to use either the password"anonymous" or "guest". Some servers require the use of a valid e-mailaddress as password.-Using [on-the-fly](#backend-path-to-dir) or+Using [on-the-fly](https://rclone.org/docs/#backend-path-to-dir) or[connection string](https://rclone.org/docs/#connection-strings) remotes makes it easy to accesssuch servers, without requiring any configuration in advance. The followingare examples of that:@@ -50290,7 +52823,7 @@ recommended usage is using the flag form --drive-size-as-quota whendoing rclone ls/lsl/lsf/lsjson/etc only.If you do use this flag for syncing (not recommended) then you will-need to use --ignore size also.+need to use --ignore-size also.Properties:@@ -51174,7 +53707,7 @@ includingYou should now see the three scopes on your Data access page. Now press saveat the bottom!-6. After adding scopes, click Audience+6. After adding scopes, click Audience.Scroll down and click "+ Add users". Add yourself as a test user and press save.7. Go to Overview on the left panel, click "Create OAuth client". Choose@@ -51188,6 +53721,15 @@ Scroll down and click "+ Add users". Add yourself as a test user and press save.9. Go to "Audience" and then click "PUBLISH APP" button and confirm.Add yourself as a test user if you haven't already.+ (If the "PUBLISH APP" button is greyed out, Google now requires a+ homepage URL and a privacy policy URL before publishing, even for a+ personal single-user app. Go to "Branding" in the left panel, fill in+ "Application home page" and "Application privacy policy link" - a free+ GitHub Pages site is fine for these if you don't have a domain of your+ own - then "Authorized domains" with the domain you used, and click+ Save. Return to "Audience" and the "PUBLISH APP" button should now be+ available.)+10. Provide the noted client ID and client secret to rclone.11. Run the web-based authorization flow from within `rclone config`, by answering@@ -52406,9 +54948,18 @@ Before accessing Drive service, developers need to create a server application f#### Enable Drive Application Switch-1. Log in to the AppGallery Connect website and click "My Projects".-2. Find your project in the project list, select the application that needs to access Drive in the project, and click "API Management".-3. In the "API Management" area of the "Project Settings" page, click the switch on the right side of Drive Kit to ensure the switch is in the open state.+1. Log in to AppGallery Connect with your Huawei account.+2. Create a project or select an existing project.+3. Add a web application to the project.+4. Open "Project settings" for the application and copy the OAuth 2.0 Client+ ID and Client Secret from the "General" tab.+5. Open "API Management" for the project and enable "Cloud Drive".+6. Add `http://127.0.0.1:53682/` as a redirect URI for the OAuth 2.0 Client+ ID.++The older Huawei Developer Console may show some of the same credential+information, but AppGallery Connect is the place to create the project,+application, API permission, and redirect URI used by rclone.#### Sign Huawei Cloud Drive Privacy Notice and User Agreement@@ -52420,23 +54971,12 @@ Applications accessing Drive should use authentication credential Access Token.Key steps summary:-1. Log into the [Huawei Developer Console](https://developer.huawei.com/consumer/en/console) with your Huawei account.--2. Create a new project or select an existing project.--3. Go to "Manage APIs" and enable the "Drive Kit" API.--4. Click on "Credentials" in the left side panel.--5. Click "Create Credentials" and choose "OAuth client ID".--6. Choose "Web application" as the application type.--7. Add `http://localhost:53682/` to the "Authorized redirect URIs".--8. Click "Create" and note down the Client ID and Client Secret.--9. Provide the noted client ID and client secret to rclone during configuration.+1. Create or select a project in AppGallery Connect.+2. Add a web application to the project.+3. Enable Cloud Drive in API Management.+4. Configure `http://127.0.0.1:53682/` as the OAuth redirect URI.+5. Copy the OAuth 2.0 Client ID and Client Secret from the application.+6. Provide the client ID and client secret to rclone during configuration.### Scopes@@ -53549,11 +56089,13 @@ it will be resolved relative to this url, according to the URL standard. Thismeans with remote url `https://beta.rclone.org/branch` and path `fix`, theresolved URL will be `https://beta.rclone.org/branch/fix`, while with path`/fix` the resolved URL will be `https://beta.rclone.org/fix` as the absolute-path is resolved from the root of the domain.+path is resolved from the root of the domain. The path can't change the scheme,+host or user of the configured url, so a path such as `//other.example.com/fix`+is rejected.If the path following the `remote:` ends with `/` it will be assumed to pointto a directory. If the path does not end with `/`, then a HEAD request is sent-and the response used to decide if it it is treated as a file or a directory+and the response used to decide if it is treated as a file or a directory(run with `-vv` to see details). When [--http-no-head](#http-no-head) isspecified, a path without ending `/` is always assumed to be a file. If rcloneincorrectly assumes the path is a file, the solution is to specify the path with@@ -53739,6 +56281,13 @@ For example, to set a Cookie use 'Cookie,name=value', or '"Cookie","name=value"'You can set multiple headers, e.g. '"Cookie","name=value","Authorization","xxx"'.+The headers are only sent to the host in the configured URL. If the+server redirects to another host (including a subdomain or a different+port) the headers are not sent to it, or to any further hop in that+redirect chain. When headers are set, a redirect from https to http is+refused as it would send them in cleartext, even with+--allow-https-downgrade.+Properties:- Config: headers@@ -53792,6 +56341,42 @@ Properties:- Type: bool- Default: false+#### --http-no-head-size++Set the reported file size for --http-no-head.++When HEAD requests are disabled, rclone normally reports an unknown file size.+Set this option to a non-negative value to report that size for every file,+including 0. Leave it unset (the default, shown as "off") to keep reporting+an unknown size.++The configured size is authoritative. If it does not match the actual file+size, transfers may fail their size check and retry unless --ignore-size is+used. It can also make multi-thread transfers request incorrect ranges and+make mounted files unreadable past the reported size.++Properties:++- Config: no_head_size+- Env Var: RCLONE_HTTP_NO_HEAD_SIZE+- Type: SizeSuffix+- Default: off++#### --http-no-head-time++Set a default modification time for --http-no-head.++When HEAD requests are disabled, rclone normally reports an unset modification+time. Set this option to report the specified time for every file instead.+Changing the configured time later makes every file appear modified to a sync.++Properties:++- Config: no_head_time+- Env Var: RCLONE_HTTP_NO_HEAD_TIME+- Type: Time+- Default: off+#### --http-descriptionDescription of the remote.@@ -55601,6 +58186,334 @@ Jottacloud exhibits some inconsistent behaviours regarding deleted files andfolders which may cause Copy, Move and DirMove operations to previouslydeleted paths to fail. Emptying the trash should help in such cases.+# kDrive++Paths are specified as `remote:path`++Paths may be as deep as required, e.g. `remote:directory/subdirectory`.++## Configuration++The initial setup for kDrive involves getting an OAuth2 token which you+need to do in your browser. `rclone config` walks you through it.++kDrive uses the PKCE OAuth2 flow (RFC 7636), which means no+`client_secret` is required. The `client_id` is optional and defaults+to rclone's built-in application credentials.++Here is an example of how to make a remote called `remote`. First run:++```sh+rclone config+```++This will guide you through an interactive setup process:++```text+No remotes found, make a new one?+n) New remote+s) Set configuration password+q) Quit config+n/s/q> n+name> remote+Type of storage to configure.+Choose a number from below, or type in your own value+[snip]+XX / Infomaniak kDrive+ \ "kdrive"+[snip]+Storage> kdrive++Option client_id.+OAuth Client Id.+Leave blank normally.+Enter a value. Press Enter to leave empty.+client_id>++Option drive_id.+ID of the kDrive to use.+When showing a folder on kdrive, you can find the drive_id here:+https://ksuite.infomaniak.com/{account_id}/kdrive/app/drive/{drive_id}/files/...+Enter a value. Press Enter to leave empty.+drive_id> 0654321++Use web browser to automatically authenticate rclone with remote?+ * Say Y if the machine running rclone has a web browser you can use+ * Say N if running rclone on a (remote) machine without web browser access+If not sure try Y. If Y failed, try N.+y/n> y++Waiting for code...+Got code++Configuration complete.+Options:+- type: kdrive+- drive_id: 0654321++Keep this "remote" remote?+y) Yes this is OK+e) Edit this remote+d) Delete this remote+y/e/d> y+```++Once configured you can then use `rclone` like this (replace `remote` with the name you gave your remote):++List directories in top level of your rclone++```sh+rclone lsd remote:+```++**Note** that by default the top level directory of a private kDrive+storage is your "Private" folder. When used with a Pro account, a second+folder "Common documents" is also available, which can be used as the+root by setting `root_folder_id` to `common`.+*These top-level folders cannot be modified*.++List all the files in your kDrive++```sh+rclone ls remote:+```++To copy a local directory to a kDrive directory called backup++```sh+rclone copy /home/source remote:backup+```++### Modification times and hashes++kDrive allows modification times to be set on objects accurate to 1+second. These will be used to detect whether objects need syncing or+not.++kDrive supports the XXH3 checksum hash, so you can use the `--checksum`+flag. XXH3 is only available for files uploaded as a single chunk:+files uploaded in chunks (by rclone when larger than+`--kdrive-chunk-size`, or by the kDrive apps which chunk files at their+own sizes) have a "nested" hash which can't be compared with a plain+XXH3, and rclone reports no hash for those files.++### Restricted filename characters++In addition to the [default restricted characters set](https://rclone.org/overview/#restricted-characters)+the following characters are also replaced:++| Character | Value | Replacement |+| ---------- |:-----:|:-----------:|+| LeftSpace | 0x20 | ␠ |+| RightSpace | 0x20 | ␠ |++Invalid UTF-8 bytes will also be [replaced](https://rclone.org/overview/#invalid-utf8),+as they can't be used in JSON strings.++### Overwriting files++When rclone overwrites an existing file, kDrive keeps the replaced+content as a new version of the file. Old versions count against the+drive quota, so repeatedly overwriting large files grows the storage+usage. Manage old versions from the kDrive web app.++### Uploading files of unknown size++kDrive can't upload files of unknown size, so `rclone rcat` and+streaming uploads are not supported.++### Deleting files++Deleted files will be moved to the trash. Your subscription level+will determine how long items stay in the trash. `rclone cleanup` can+be used to empty the trash.++### Root folder ID++You can set the `root_folder_id` for rclone. This is the directory+(identified by its numeric `Folder ID`) that rclone considers to be the+root of your kDrive drive.++Normally you will leave this blank and rclone will use your private+directory. You can also use one of the convenience shortcuts:++- `private`: your user private directory (default)+- `common`: the kDrive common directory, shared among users++Or the numeric `Folder ID` of any directory in your drive. The root of+the drive only contains the `Private` and `Common documents` folders,+so a numeric ID is needed to use any other directory as root. In order+to find the `Folder ID`, list the parent directory of the folder you+wish rclone to use with `rclone lsf`; the first column of the output is+the folder ID:++Example:+```+$ rclone lsf --dirs-only -Fip --csv remote:Documents+456,Sub folder/+789,Photos/+```++So if the folder you want rclone to use is "Photos/", then use the+returned id from ```rclone lsf``` command (ex. `789`) as the+`root_folder_id` variable value in the config file.++<!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/kdrive/kdrive.go and run make backenddocs to verify --> <!-- markdownlint-disable-line line-length -->+### Standard options++Here are the Standard options specific to kdrive (Infomaniak kDrive).++#### --kdrive-client-id++OAuth Client Id.++Leave blank normally.++Properties:++- Config: client_id+- Env Var: RCLONE_KDRIVE_CLIENT_ID+- Type: string+- Required: false++#### --kdrive-drive-id++ID of the kDrive to use.++When showing a folder on kdrive, you can find the drive_id here:+https://ksuite.infomaniak.com/{account_id}/kdrive/app/drive/{drive_id}/files/...++Properties:++- Config: drive_id+- Env Var: RCLONE_KDRIVE_DRIVE_ID+- Type: string+- Required: false++### Advanced options++Here are the Advanced options specific to kdrive (Infomaniak kDrive).++#### --kdrive-token++OAuth Access Token as a JSON blob.++Properties:++- Config: token+- Env Var: RCLONE_KDRIVE_TOKEN+- Type: string+- Required: false++#### --kdrive-auth-url++Auth server URL.++Leave blank to use the provider defaults.++Properties:++- Config: auth_url+- Env Var: RCLONE_KDRIVE_AUTH_URL+- Type: string+- Required: false++#### --kdrive-token-url++Token server url.++Leave blank to use the provider defaults.++Properties:++- Config: token_url+- Env Var: RCLONE_KDRIVE_TOKEN_URL+- Type: string+- Required: false++#### --kdrive-encoding++The encoding for the backend.++See the [encoding section in the overview](https://rclone.org/overview/#encoding) for more info.++Properties:++- Config: encoding+- Env Var: RCLONE_KDRIVE_ENCODING+- Type: Encoding+- Default: Slash,Del,Ctl,LeftSpace,RightSpace,InvalidUtf8,Dot++#### --kdrive-root-folder-id++The folder to use as root.++You may use either one of the convenience shortcuts [private or common]+or an explicit folder ID.++- private: your user private directory+- common: the kDrive common directory, shared among users++Properties:++- Config: root_folder_id+- Env Var: RCLONE_KDRIVE_ROOT_FOLDER_ID+- Type: string+- Required: false++#### --kdrive-chunk-size++Files above this size are uploaded in chunks of this size.++The kDrive API accepts chunks between 1 MB and 1 GB, up to 10,000+chunks per file and a total file size of 1 TiB. The chunk size is+raised automatically for very large files to stay within the 10,000+chunks limit.++Properties:++- Config: chunk_size+- Env Var: RCLONE_KDRIVE_CHUNK_SIZE+- Type: SizeSuffix+- Default: 20Mi++#### --kdrive-upload-concurrency++Concurrency for chunked uploads.++Properties:++- Config: upload_concurrency+- Env Var: RCLONE_KDRIVE_UPLOAD_CONCURRENCY+- Type: int+- Default: 4++#### --kdrive-endpoint++The API endpoint to use.++Leave blank normally. There is no reason to change the endpoint except for internal use or beta-testing.++Properties:++- Config: endpoint+- Env Var: RCLONE_KDRIVE_ENDPOINT+- Type: string+- Required: false++#### --kdrive-description++Description of the remote.++Properties:++- Config: description+- Env Var: RCLONE_KDRIVE_DESCRIPTION+- Type: string+- Required: false++<!-- autogenerated options stop -->+# KoofrPaths are specified as `remote:path`@@ -56125,6 +59038,15 @@ Properties:Invalid UTF-8 bytes will also be [replaced](https://rclone.org/overview/#invalid-utf8),as they can't be used in JSON strings.+Linkbox converts file extensions to lower case on upload, so a file+uploaded as `file.TXT` is stored as `file.txt`.++Linkbox renames an uploaded file to `name(1)` if its name is the start+of the name of a file already in the same directory, for example when+uploading `apple` to a directory containing `applebanana`. rclone then+can't find the file it uploaded and the upload fails with `object not+found`.+# Mail.ru Cloud[Mail.ru Cloud](https://cloud.mail.ru/) is a cloud storage provided by a@@ -57479,6 +60401,24 @@ when `--metadata` is enabled (or when using `--metadata-set` / `--metadata-mappeNotes:- Rclone ignores reserved `x-ms-*` keys (except `x-ms-tags`) for user metadata.+### Hierarchical namespace++Storage accounts with hierarchical namespace enabled (Azure Data Lake+Storage Gen2) have real directories. On these accounts rclone supports+empty directories, creating them with `rclone mkdir` and removing+them with `rclone rmdir`, whatever the `--azureblob-directory-markers`+setting.++Note that deleting the last file in a directory leaves the directory+behind. Use `rclone rmdirs` or `--delete-empty-src-dirs` to remove+empty directories.++Rclone reads whether the storage account has a hierarchical namespace+when it first uses the remote and saves it as the `hns` option in the+config file. Set `--azureblob-hns` to override this.++Blob tags (`x-ms-tags` metadata) aren't supported on these accounts.+### PerformanceWhen uploading large files, increasing the value of@@ -57710,13 +60650,13 @@ be explicitly specified using exactly one of the `msi_object_id`,`msi_client_id`, or `msi_mi_res_id` parameters.If none of `msi_object_id`, `msi_client_id`, or `msi_mi_res_id` is-set, this is is equivalent to using `env_auth`.+set, this is equivalent to using `env_auth`.#### Federated Identity CredentialsIf these variables are set, rclone will authenticate with federated identity.-- `tenant`: tenant ID of of the storage+- `tenant`: tenant ID of the storage- `client_id`: client ID of the application the user will authenticate to storage- `msi_client_id`: managed identity client ID of the application the user willauthenticate to@@ -58257,6 +61197,52 @@ Properties:- Type: int- Default: 5000+#### --azureblob-use-arrow-list++Use the Apache Arrow listing format.++If set, directory listings are fetched using the ListBlobs Apache+Arrow response format instead of XML. Arrow responses are smaller and+much cheaper to parse, making listings of large containers several+times faster. Combine with "list_parallelism" for the biggest gains.++"Blob Listing with Apache Arrow" is in public preview at Microsoft and+is only supported on flat namespace accounts. On accounts with a+hierarchical namespace (ADLS Gen2), or where the feature is otherwise+unavailable, the server returns XML and the listing transparently+falls back to the normal XML path (logged at debug level).++Properties:++- Config: use_arrow_list+- Env Var: RCLONE_AZUREBLOB_USE_ARROW_LIST+- Type: bool+- Default: false++#### --azureblob-list-parallelism++Number of parallel shards to list a directory with.++If set greater than 1, the blob name keyspace of each directory is+split into ranges which are listed concurrently using the Arrow+startFrom/endBefore range parameters. This can dramatically speed up+listing containers with millions of objects, for both recursive+(ListR) and single directory listings. Speed keeps improving up to a+parallelism of around 30.++This has no effect unless "use_arrow_list" is also set, as Arrow is the+only listing path that supports server-side name ranges. If the+account does not support range listing (e.g. it has a hierarchical+namespace) the listing falls back to sequential. The default of 0 (or+1) lists sequentially.++Properties:++- Config: list_parallelism+- Env Var: RCLONE_AZUREBLOB_LIST_PARALLELISM+- Type: int+- Default: 0+#### --azureblob-access-tierAccess tier of blob: hot, cool, cold or archive.@@ -58360,6 +61346,9 @@ creates an empty object ending with "/", to persist the folder.This object also has the metadata "hdi_isfolder = true" to conform tothe Microsoft standard.++This is not needed on storage accounts with a hierarchical namespace+(see the hns option) as these always support empty directories.Properties:@@ -58369,6 +61358,24 @@ Properties:- Type: bool- Default: false+#### --azureblob-hns++Set if the storage account has a hierarchical namespace.++Storage accounts with hierarchical namespace enabled (Azure Data Lake+Storage Gen2) have real directories which can be empty.++If this is unset then rclone will read it from the storage account+and save it in the config file.+++Properties:++- Config: hns+- Env Var: RCLONE_AZUREBLOB_HNS+- Type: Tristate+- Default: unset+#### --azureblob-no-check-containerIf set, don't attempt to check the container exists or create it.@@ -58828,13 +61835,13 @@ be explicitly specified using exactly one of the `msi_object_id`,`msi_client_id`, or `msi_mi_res_id` parameters.If none of `msi_object_id`, `msi_client_id`, or `msi_mi_res_id` is-set, this is is equivalent to using `env_auth`.+set, this is equivalent to using `env_auth`.#### Federated Identity CredentialsIf these variables are set, rclone will authenticate with federated identity.-- `tenant`: tenant ID of of the storage+- `tenant`: tenant ID of the storage- `client_id`: client ID of the application the user will authenticate to storage- `msi_client_id`: managed identity client ID of the application the user willauthenticate to@@ -59239,7 +62246,7 @@ Properties:Upload chunk size.Note that this is stored in memory and there may be up to-"--transfers" * "--azurefile-upload-concurrency" chunks stored at once+"--transfers" * "--azurefiles-upload-concurrency" chunks stored at oncein memory.Properties:@@ -59261,7 +62268,7 @@ links and these uploads do not fully utilize your bandwidth, thenincreasing this may help to speed up the transfers.Note that chunks are stored in memory and there may be up to-"--transfers" * "--azurefile-upload-concurrency" chunks stored at once+"--transfers" * "--azurefiles-upload-concurrency" chunks stored at oncein memory.Properties:@@ -59343,6 +62350,10 @@ Paths are specified as `remote:path`Paths may be as deep as required, e.g. `remote:directory/subdirectory`.+**Note**: OneDrive for Business and SharePoint users need rclone+v1.76 or later from April 1, 2027 - see+[Pre-authenticated URLs](#pre-authenticated-urls).+## ConfigurationThe initial setup for OneDrive involves getting a token from@@ -59575,7 +62586,7 @@ OneDrive files. Take care to safeguard these credentials.If you do not have admin access to your organization's OneDrive forBusiness, you can still connect by manually providing the SharePointtenant URL and drive ID. This works by overriding the base API URL-from the standard Microsoft Graph endpoint to the SharePoint v2.0+from the standard Microsoft Graph endpoint to the SharePoint APIendpoint.#### Steps to manually obtain credentials@@ -59589,6 +62600,13 @@ endpoint.".driveAccessToken": "access_token={access_token}"```+ In some guest access scenarios SharePoint may also expose a matching+ v2.1 URL and token pair:+ ```json+ ".driveUrlV21": "{tenant_url}/v2.1/drives/{drive_id}",+ ".driveAccessTokenV21": "access_token={access_token}"+ ```+#### Rclone configurationUse the extracted values to configure your remote:@@ -59601,6 +62619,14 @@ tenant_url = {tenant_url}drive_type = business```+If you use a v2.1 token, set the matching SharePoint API version:++```ini+tenant_api_version = v2.1+```++If `tenant_api_version` is omitted, rclone defaults to `v2.0`.+Since the exact expiry time cannot be determined from web traffic,set the expiry to a future date. Note that the token will eventuallyexpire and you will need to repeat the process to obtain a new one.@@ -59644,10 +62670,11 @@ This can be useful with `rclone mount` and [rclone rc vfs/refreshrecursive=true](https://rclone.org/rc/#vfs-refresh)) to very quickly fill the mount withinformation about all the files.-The API used for the recursive listing (`ListR`) only supports listing-from the root of the drive. This will become increasingly inefficient-the further away you get from the root as rclone will have to discard-files outside of the directory you are using.+Rclone asks for the delta listing of the directory being listed. Some+drives only support delta listings from the root of the drive, in+which case rclone lists the whole drive and discards the files outside+of the directory you are using, which is increasingly inefficient the+further away from the root you get.Some commands (like `rclone lsf -R`) will use `ListR` by default - youcan turn this off with `--disable ListR` if you need to.@@ -59841,7 +62868,7 @@ Properties:The tenant URL for non-admin OneDrive access.-Set this to your SharePoint tenant URL to use the SharePoint v2.0 API+Set this to your SharePoint tenant URL to use the SharePoint APIendpoint instead of the standard Microsoft Graph API. This allowsaccessing business OneDrive without admin consent.@@ -59858,6 +62885,22 @@ Properties:- Type: string- Required: false+#### --onedrive-tenant-api-version++The SharePoint API version to use with tenant_url.++Set this to the SharePoint API version matching the browser-extracted+access token. For example, use v2.1 with a driveAccessTokenV21 token.++This only applies when tenant_url is set.++Properties:++- Config: tenant_api_version+- Env Var: RCLONE_ONEDRIVE_TENANT_API_VERSION+- Type: string+- Default: "v2.0"+#### --onedrive-chunk-sizeChunk size to upload files with - must be multiple of 320k (327,680 bytes).@@ -59984,7 +63027,7 @@ Properties:Remove all versions on modifying operations.-Onedrive for business creates versions when rclone uploads new files+Onedrive creates versions when rclone uploads new filesoverwriting an existing one and when it sets the modification time.These versions take up space out of the quota.@@ -59992,9 +63035,6 @@ These versions take up space out of the quota.This flag checks for versions after file upload and settingmodification time and removes all but the last version.-**NB** Onedrive personal can't currently delete versions so don't use-this flag there.-Properties:@@ -60010,8 +63050,7 @@ Permanently delete files on removal.Normally files will get sent to the recycle bin on deletion. Settingthis flag causes them to be permanently deleted. Use with care.-OneDrive personal accounts do not support the permanentDelete API,-it only applies to OneDrive for Business and SharePoint document libraries.+This works with OneDrive for Business, SharePoint document libraries, and OneDrive personal accounts, including free accounts.Properties:@@ -60057,12 +63096,17 @@ Properties:- Creates a read-write link to the item.- "embed"- Creates an embeddable link to the item.+ - Only available in OneDrive personal.#### --onedrive-link-passwordSet the password for links created by the link command.-At the time of writing this only works with OneDrive personal paid accounts.+At the time of writing this works with OneDrive for Business and+OneDrive personal paid accounts.++OneDrive personal free accounts can't set a password or an expiry time+(with --expire) on links.Properties:@@ -60161,21 +63205,12 @@ Setting this flag speeds up these things greatly:rclone size onedrive:rclone rc vfs/refresh recursive=true-**However** the delta listing API **only** works at the root of the-drive. If you use it not at the root then it recurses from the root-and discards all the data that is not under the directory you asked-for. So it will be correct but may not be very efficient.--This is why this flag is not set as the default.--As a rule of thumb if nearly all of your data is under rclone's root-directory (the `root/directory` in `onedrive:root/directory`) then-using this flag will be a big performance win. If your data is-mostly not under the root then using this flag will be a big-performance loss.--It is recommended if you are mounting your onedrive at the root-(or near the root when using crypt) and using rclone `rc vfs/refresh`.+Rclone asks for the delta listing of the directory being listed. If+the drive only supports delta listings at the root of the drive (as+some older OneDrive for Business and SharePoint drives do) then+rclone lists from the root and discards all the data that is not+under the directory you asked for. So it will be correct but may not+be very efficient.Properties:@@ -60373,7 +63408,7 @@ Here are the possible system metadata items for the onedrive backend.| Name | Help | Type | Example | Read Only ||------|------|------|---------|-----------|-| btime | Time of file birth (creation) with S accuracy (mS for OneDrive Personal). | RFC 3339 | 2006-01-02T15:04:05Z | N |+| btime | Time of file birth (creation) with S accuracy. | RFC 3339 | 2006-01-02T15:04:05Z | N || content-type | The MIME type of the file. | string | text/plain | **Y** || created-by-display-name | Display name of the user that created the item. | string | John Doe | **Y** || created-by-id | ID of the user that created the item. | string | 48d31887-5fad-4d73-a9f5-3c356e68a038 | **Y** |@@ -60382,14 +63417,14 @@ Here are the possible system metadata items for the onedrive backend.| last-modified-by-display-name | Display name of the user that last modified the item. | string | John Doe | **Y** || last-modified-by-id | ID of the user that last modified the item. | string | 48d31887-5fad-4d73-a9f5-3c356e68a038 | **Y** || malware-detected | Whether OneDrive has detected that the item contains malware. | boolean | true | **Y** |-| mtime | Time of last modification with S accuracy (mS for OneDrive Personal). | RFC 3339 | 2006-01-02T15:04:05Z | N |+| mtime | Time of last modification with S accuracy. | RFC 3339 | 2006-01-02T15:04:05Z | N || package-type | If present, indicates that this item is a package instead of a folder or file. Packages are treated like files in some contexts and folders in others. | string | oneNote | **Y** || permissions | Permissions in a JSON dump of OneDrive format. Enable with --onedrive-metadata-permissions. Properties: id, grantedTo, grantedToIdentities, invitation, inheritedFrom, link, roles, shareId | JSON | {} | N || shared-by-id | ID of the user that shared the item (if shared). | string | 48d31887-5fad-4d73-a9f5-3c356e68a038 | **Y** || shared-owner-id | ID of the owner of the shared item (if shared). | string | 48d31887-5fad-4d73-a9f5-3c356e68a038 | **Y** || shared-scope | If shared, indicates the scope of how the item is shared: anonymous, organization, or users. | string | users | **Y** |-| shared-time | Time when the item was shared, with S accuracy (mS for OneDrive Personal). | RFC 3339 | 2006-01-02T15:04:05Z | **Y** |-| utime | Time of upload with S accuracy (mS for OneDrive Personal). | RFC 3339 | 2006-01-02T15:04:05Z | **Y** |+| shared-time | Time when the item was shared, with S accuracy. | RFC 3339 | 2006-01-02T15:04:05Z | **Y** |+| utime | Time of upload with S accuracy. | RFC 3339 | 2006-01-02T15:04:05Z | **Y** |See the [metadata](https://rclone.org/docs/#metadata) docs for more info.@@ -60487,11 +63522,6 @@ Or you can set the `no_versions` parameter to `true` and rclone willremove versions after operations which create new versions. This takesextra transactions so only enable it if you need it.-**Note** At the time of writing Onedrive Personal creates versions-(but not for setting the modification time) but the API for removing-them returns "API not found" so cleanup and `no_versions` should not-be used on Onedrive Personal.-### Disabling versioningStarting October 2018, users will no longer be able to@@ -60546,10 +63576,37 @@ rclone cleanup --interactive remote:path/subdir # interactively remove all old vrclone cleanup remote:path/subdir # unconditionally remove all old version for path/subdir```-**NB** Onedrive personal can't currently delete versions-## Troubleshooting+### Pre-authenticated URLs++Microsoft is [retiring pre-authenticated (tempauth) URLs](https://devblogs.microsoft.com/microsoft365dev/pre-authenticated-tempauth-urls-are-retiring-move-to-microsoft-entra-id-tokens-before-april-1-2027/)+in OneDrive for Business and SharePoint from April 1, 2027.++**Upgrade to rclone v1.76 or later before April 1, 2027.** Earlier+versions rely on pre-authenticated URLs, so after that date their+downloads, chunked uploads and server-side copies will fail with+OneDrive for Business and SharePoint. Rclone v1.76 and later keep+working with the existing configuration, apart from the client+credentials flow with a client secret (see below).++Rclone uses these URLs when downloading files, when uploading files+bigger than `--onedrive-upload-cutoff` in chunks and when doing+server-side copies. When SharePoint returns them without+pre-authentication rclone authenticates them with a SharePoint token+which it gets using the existing login, so no reconfiguration is+needed. OneDrive Personal is not affected.++If rclone can't get a SharePoint token, or SharePoint refuses it, you+will see the error `SharePoint refused to authenticate a request`. In+this case rclone uploads files smaller than 250 MiB in a single+request instead and copies files by downloading and uploading them,+but bigger files can't be uploaded and downloads may fail.++SharePoint does not accept tokens from the client credentials flow+which were got with a client secret, so this affects+`client_credentials` users.+### Excessive throttling or blocked on SharePointIf you experience excessive throttling or is being blocked on SharePoint then@@ -60611,7 +63668,7 @@ This means that rclone can't use the OneDrive for Business API with your accountYou can't do much about it, maybe write an email to your admins.However, there are other ways to interact with your OneDrive account. Have a look-at the WebDAV backend: <https://rclone.org/webdav/#sharepoint>+at the WebDAV backend: <https://rclone.org/webdav/#sharepoint-online>### invalid\_grant (AADSTS50076)@@ -60651,6 +63708,27 @@ Shared with me files is not supported by rclone")")+Listing such a shortcut may still fail with++```text+invalidRequest: The provided drive id appears to be malformed, or does not represent a valid drive.+```++as the drive it points to can't always be read with the drive ID the+API supplies. `Personal Vault` is a shortcut of the same kind and+fails to list in the same way, though it reports+`invalidResourceId: ObjectHandle is Invalid`.++A normal listing reports the error for that folder and carries on, so+the rest of the drive is still listed, and excluding the folder (for+example with `--exclude "/Personal Vault/**"`) avoids the error.++A recursive listing with `--fast-list` stops at the first error, so+listing the root of a drive which has one of these shortcuts in it+will fail. Filters don't help here as the folder is listed by the+backend before the filters are applied - use `--disable ListR` (or+don't use `--fast-list`) on such a drive.+### Live Photos uploaded from iOS (small video clips in .heic files)The iOS OneDrive app introduced [upload and storage](https://techcommunity.microsoft.com/t5/microsoft-onedrive-blog/live-photos-come-to-onedrive/ba-p/1953452)@@ -61523,7 +64601,7 @@ Properties:- Config: encoding- Env Var: RCLONE_OOS_ENCODING- Type: Encoding-- Default: Slash,InvalidUtf8,Dot+- Default: Slash,CrLf,InvalidUtf8,Dot#### --oos-leave-parts-on-error@@ -62510,7 +65588,7 @@ Before you can use rclone with Sia, you will need to have a running copy ofnetwork (e.g. a NAS). Please follow the [Get started](https://sia.tech/get-started)guide and install one.-rclone interacts with Sia network by talking to the Sia daemon via [HTTP API](https://sia.tech/docs/)+rclone interacts with Sia network by talking to the Sia daemon via [HTTP API](https://docs.sia.tech/)which is usually available on port *9980*. By default you will run the daemonlocally on the same computer so it's safe to leave the API password blank(the API URL will be `http://127.0.0.1:9980` making external access impossible).@@ -62720,6 +65798,7 @@ Commercial implementations of that being:- [Oracle Cloud Storage](https://docs.oracle.com/en-us/iaas/integration/doc/configure-object-storage.html)- [Blomp Cloud Storage](https://www.blomp.com/cloud-storage/)- [IBM Bluemix Cloud ObjectStorage Swift](https://console.bluemix.net/docs/infrastructure/objectstorage-swift/index.html)+- [Infomaniak Swiss Backup](https://www.infomaniak.com/en/swiss-backup/cloud-storage)Paths are specified as `remote:container` (or `remote:` for the `lsd`command.) You may put subdirectories in too, e.g. `remote:container/path/to/dir`.@@ -62856,7 +65935,7 @@ rclone sync --interactive /home/local/directory remote:container### Configuration from an OpenStack credentials file-An OpenStack credentials file typically looks something something+An OpenStack credentials file typically looks somethinglike this (without the comments)```sh@@ -63032,6 +66111,14 @@ Properties:- OVH- "https://authenticate.ain.net"- Blomp Cloud Storage+ - "https://swiss-backup01.infomaniak.com/identity/v3"+ - Infomaniak Swiss Backup Location 01+ - "https://swiss-backup02.infomaniak.com/identity/v3"+ - Infomaniak Swiss Backup Location 02+ - "https://swiss-backup03.infomaniak.com/identity/v3"+ - Infomaniak Swiss Backup Location 03+ - "https://swiss-backup04.infomaniak.com/identity/v3"+ - Infomaniak Swiss Backup Location 04#### --swift-user-id@@ -63414,7 +66501,7 @@ Request" error rather than a more sensible error when theauthentication fails for Swift.So this most likely means your username / password is wrong. You can-investigate further with the `--dump-bodies` flag.+investigate further with the `--dump bodies` flag.This may also be caused by specifying the region when you shouldn'thave (e.g. OVH).@@ -64006,6 +67093,8 @@ Properties:Files bigger than this will be cached on disk to calculate hash if required.+Files of unknown size are always cached on disk.+Properties:- Config: hash_memory_limit@@ -65788,6 +68877,12 @@ the following characters are also replaced:Invalid UTF-8 bytes will also be [replaced](https://rclone.org/overview/#invalid-utf8),as they can't be used in JSON strings.+Seafile stores file and directory names in Unicode NFC form whatever form+they are uploaded in, so rclone normalizes names to NFC before sending them.+Listings therefore always show NFC names, even for files created with+decomposed (NFD) names as macOS produces them. Library names are not+normalized as Seafile keeps them exactly as entered.+### Seafile and rclone linkRclone supports generating share links for non-encrypted libraries only.@@ -66386,7 +69481,7 @@ correct, and support all features.The shell type auto-detection logic, described above, means thatby default rclone will try to run a shell command the first timea new sftp remote is accessed. If you configure a sftp remote-without a config file, e.g. an [on the fly](https://rclone.org/docs/#backend-path-to-dir])+without a config file, e.g. an [on the fly](https://rclone.org/docs/#backend-path-to-dir)remote, rclone will have nowhere to store the result, and itwill re-run the command on every access. To avoid this you shouldexplicitly set the `shell_type` option to the correct value,@@ -66466,6 +69561,24 @@ upload (for example, certain configurations of ProFTPd with mod_sftp). If youare using one of these servers, you can set the option `set_modtime = false` inyour RClone backend configuration to disable this behaviour.+### Restricted filename characters++SFTP servers on Unix-like systems can store any character in a file name+except NUL and `/`, so by default only these are replaced:++| Character | Value | Replacement |+| --------- |:-----:|:-----------:|+| NUL | 0x00 | ␀ |+| / | 0x2F | / |++If the server runs on Windows, which can't store many more characters,+add the Windows restricted characters with `--sftp-encoding`, as in the+[FTP example](https://rclone.org/overview/#encoding-example-ftp):++```text+--sftp-encoding "Slash,LtGt,DoubleQuote,Colon,Question,Asterisk,Pipe,BackSlash,Ctl,RightSpace,RightPeriod,InvalidUtf8,Dot"+```+### About commandThe `about` command returns the total space, free space, and used@@ -66846,7 +69959,7 @@ Properties:- Config: encoding- Env Var: RCLONE_SFTP_ENCODING- Type: Encoding-- Default: Slash,Del,Ctl,Dot+- Default: Slash,Dot#### --sftp-set-modtime@@ -67095,6 +70208,34 @@ Properties:- Type: bool- Default: false+#### --sftp-multithread-upload++Set this to use multi-thread (multi-connection) uploads.++Normally rclone uploads a single large file over one SFTP connection.+With this set rclone uploads large files over several SFTP connections+at once (one per chunk), which greatly improves throughput on high+latency links, the same way multi-thread downloads already work. Each+connection writes a different, non-overlapping byte range of the file,+so this requires a server that allows several handles to the same file+with writes at arbitrary offsets, which OpenSSH does.++This is off by default because many SFTP servers only accept sequential+writes (for example some object storage backed SFTP gateways) and would+fail large uploads with "truncate failed" or "invalid offset" style+errors.++This is ignored (uploads stay single-threaded) when+--sftp-disable-concurrent-writes or --sftp-connections is in use.+++Properties:++- Config: multithread_upload+- Env Var: RCLONE_SFTP_MULTITHREAD_UPLOAD+- Type: bool+- Default: false+#### --sftp-idle-timeoutMax time before closing idle connections.@@ -67370,8 +70511,8 @@ Properties:On some SFTP servers (e.g. Synology) the paths are differentfor SSH and SFTP so the hashes can't be calculated properly.-You can either use [`--sftp-path-override`](#--sftp-path-override)-or [`disable_hashcheck`](#--sftp-disable-hashcheck).+You can either use [`--sftp-path-override`](#sftp-path-override)+or [`disable_hashcheck`](#sftp-disable-hashcheck).The only ssh agent supported under Windows is Putty's pageant.@@ -67380,13 +70521,13 @@ default, due to security concerns. This can be re-enabled on aper-connection basis by setting the `use_insecure_cipher` setting inthe configuration file to `true`. Further details on the insecurity ofthis cipher can be found-[in this paper](http://www.isg.rhul.ac.uk/~kp/SandPfinal.pdf).+[in this paper](https://conferences.computer.org/sp/pdfs/sp/2009/oakland2009-01.pdf).SFTP isn't supported under plan9 until [thisissue](https://github.com/pkg/sftp/issues/156) is fixed.Note that since SFTP isn't HTTP based the following flags don't work-with it: `--dump-headers`, `--dump-bodies`, `--dump-auth`.+with it: `--dump headers`, `--dump bodies`, `--dump auth`.Note that `--timeout` and `--contimeout` are both supported.@@ -67796,12 +70937,14 @@ Properties:SMB username.+Leave blank to use the current user name.+Properties:- Config: user- Env Var: RCLONE_SMB_USER- Type: string-- Default: "$USER"+- Required: false#### --smb-port@@ -67878,6 +71021,23 @@ Properties:Here are the Advanced options specific to smb (SMB / CIFS).+#### --smb-workstation++Workstation name sent in NTLM authentication.++The NetBIOS name sent as the client workstation in the NTLM AUTHENTICATE+message. Some servers (e.g. domain-joined Windows servers with per-account+workstation restrictions or NTLM allow-lists) require this to match a+specific name. Leave blank to send no workstation name (the default).+++Properties:++- Config: workstation+- Env Var: RCLONE_SMB_WORKSTATION+- Type: string+- Required: false+#### --smb-idle-timeoutMax time before closing idle connections.@@ -68074,7 +71234,7 @@ Side by side comparison with more details:To make a new Storj configuration you need one of the following:- Access Grant that someone else shared with you.-- [API Key](https://documentation.storj.io/getting-started/uploading-your-first-object/create-an-api-key)+- [API Key](https://storj.dev/learn/concepts/access/access-grants/api-key)of a Storj project you are a member of.Here is an example of how to make a remote called `remote`. First run:@@ -69482,7 +72642,9 @@ Likewise plain WebDAV does not support hashes, however when used withFastmail Files, ownCloud or Nextcloud rclone will support SHA1 and MD5 hashes.Depending on the exact version of ownCloud or Nextcloud hashes mayappear on all objects, or only on objects which had a hash uploaded-with them.+with them. With Nextcloud, rclone asks the server to calculate the SHA1+of uploads which had no hash to send, such as streamed uploads, and+after setting the modification time, which discards the stored hash.<!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/webdav/webdav.go and run make backenddocs to verify --> <!-- markdownlint-disable-line line-length -->### Standard options@@ -69610,6 +72772,13 @@ For example, to set a Cookie use 'Cookie,name=value', or '"Cookie","name=value"'You can set multiple headers, e.g. '"Cookie","name=value","Authorization","xxx"'.+The headers are only sent to the host in the configured URL. If the+server redirects to another host (including a subdomain or a different+port) the headers are not sent to it, or to any further hop in that+redirect chain. The exception is the directory listing and file+download requests when auth_redirect is set, as that keeps all+credentials across redirects.+Properties:@@ -69745,7 +72914,7 @@ ownCloud supports modified times using the `X-OC-Mtime` header.This is configured in an identical way to ownCloud. Note thatNextcloud initially did not support streaming of files (`rcat`) whereas-ownCloud did, but [this](https://github.com/nextcloud/nextcloud-snap/issues/365)+ownCloud did, but [this](https://github.com/nextcloud-snap/nextcloud-snap/issues/365)seems to be fixed as of 2020-11-27 (tested with rclone v1.53.1 and NextcloudServer v19).@@ -69848,7 +73017,7 @@ Rclone uses the "Last Modified" datetime property to compare your documents:## RcloneUse this option if you are hosting remotes over WebDAV provided by rclone.-Read [rclone serve webdav](commands/rclone_serve_webdav/) for more details.+Read [rclone serve webdav](https://rclone.org/commands/rclone_serve_webdav/) for more details.rclone serve supports modified times using the `X-OC-Mtime` header.@@ -70051,6 +73220,16 @@ are replaced.Invalid UTF-8 bytes will also be [replaced](https://rclone.org/overview/#invalid-utf8),as they can't be used in JSON strings.+### Application folder++If you registered your own OAuth application with Yandex and only granted+it the "Application Folder" permission (`cloud_api:disk.app_folder`)+instead of full disk access, then you must set the `app_folder` option to+`true`. Yandex Disk requires paths to be addressed with an `app:/` prefix+instead of the usual `disk:/` prefix when using this restricted scope, and+rclone will return a 403 Forbidden error on every request if this option+isn't set correctly for the token in use.+<!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/yandex/yandex.go and run make backenddocs to verify --> <!-- markdownlint-disable-line line-length -->### Standard options@@ -70194,6 +73373,26 @@ Properties:- Type: Duration- Default: 0s+#### --yandex-app-folder++Use the application folder as the root.++If you registered your own OAuth application with Yandex with only+the "Application Folder" permission (cloud_api:disk.app_folder)+instead of full disk access, then rclone needs to address paths with+the "app:/" prefix instead of the usual "disk:/" prefix, otherwise+all requests fail with a 403 Forbidden error.++Set this to true if your OAuth token only has access to the+application folder.++Properties:++- Config: app_folder+- Env Var: RCLONE_YANDEX_APP_FOLDER+- Type: bool+- Default: false+#### --yandex-descriptionDescription of the remote.@@ -70871,6 +74070,9 @@ $ rclone -L ls /tmp/a6 b/one```+Symlinks which point to a parent directory would loop forever, so+rclone reports them as errors and skips them.+#### --local-links, --links, -lNormally rclone will ignore symlinks or junction points (which behave@@ -70927,6 +74129,9 @@ $ tree /tmp/b└── file2 -> /home/user/file3```+A `.rclonelink` file whose contents are too long to be a path (more+than 128 KiB) is refused rather than turned into a symlink.+However, if copied back without '-l'```console@@ -71235,6 +74440,13 @@ deduplicated by default (i.e. having two identical files does not consume morestorage than having just one.) However, for use cases where data redundancy ispreferable, --local-no-clone can be used to disable cloning and force "deep" copies.+Cloning also preserves macOS-specific metadata such as extended attributes,+which a normal copy does not, so this metadata is lost when cloning is disabled.++If the destination file already exists, cloning replaces it with a new file,+even with --inplace, so hard links to it are broken. Use --local-no-clone to+update existing files in place.+Currently, cloning is only supported when using APFS on macOS (support for otherplatforms may be added in the future.)@@ -71324,10 +74536,10 @@ Properties:Make out-of-space errors fatal during transfers.-When enabled, an ENOSPC error during a write returns a fatal error so-that rclone aborts rather than retrying the operation. Useful for-backup scripts that should halt loudly on a full disk rather than spin-retrying.+When enabled, an out of space error while writing, creating a directory,+or creating a file returns a fatal error so that rclone aborts rather+than retrying the operation. Useful for backup scripts that should halt+loudly on a full disk rather than spin retrying.Properties:@@ -71425,8 +74637,10 @@ Metadata is supported on files and directories.When restoring metadata with `--metadata` rclone applies the"mode", "uid" and "gid" from the source. These come from the sourceremote which may not be trusted, so restoring metadata as root from an-untrusted source can change file ownership and is not recommended. The-setuid, setgid and sticky bits are not restored by default - see the+untrusted source can change file ownership and is not recommended. A+user other than root keeps its own uid and sets the gid only to a group+it is a member of, as rsync does. The setuid, setgid and sticky bits are+not restored by default - see the`--local-metadata-restore-special-bits` flag.Here are the possible system metadata items for the local backend.@@ -71482,6 +74696,644 @@ Options:# Changelog+## v1.76.0 - 2026-11-06++[See commits](https://github.com/rclone/rclone/compare/v1.75.0...v1.76.0)++- New backends+ - [Dosya](https://rclone.org/dosya/) (Kaya)+ - [Infomaniak kDrive](https://rclone.org/kdrive/) (Guillaume MALLERET)+- New S3 Providers+ - [Autonomys Auto Drive](https://rclone.org/s3/#autonomys) (Jim Counter)+ - [Clever Cloud Cellar](https://rclone.org/s3/#clever-cloud-cellar) (David Legrand)+ - [Infomaniak Swiss Backup](https://rclone.org/s3/#infomaniak) (Louis)+ - [Neon Object Storage](https://rclone.org/s3/#neon) (Rishi Raj Jain)+ - [RelAix.Storage](https://rclone.org/s3/#relaix) (mygrexit)+- New commands+ - [index](https://rclone.org/commands/rclone_index/): Write static directory listings into a remote (Nick Craig-Wood)+- Security+ - build+ - Disable Multipath TCP listeners by default to mitigate CVE-2026-98432 (Nick Craig-Wood)+ - Fix multiple CVEs including CVE-2026-111943 by upgrading to go1.27.3 (Nick Craig-Wood)+ - Update `golang.org/x/net` to v0.62.0 to fix multiple CVEs (Nick Craig-Wood)+ - Update `github.com/andybalholm/brotli` to v1.2.7 to fix CVE-2026-111497 (Nick Craig-Wood)+ - fshttp: Refuse HTTPS to HTTP redirects unless `--allow-https-downgrade` GHSA-m97j-7x65-wh93 CVE-PENDING (Nick Craig-Wood)+ - lib/http: Stop `--htpasswd` and `--user` revealing which usernames exist through timing GHSA-hwj7-wp9g-wrv9 CVE-PENDING (Nick Craig-Wood)+ - local+ - Fix user metadata escaping the root through a planted symlink GHSA-r5wv-vqqh-2r9c CVE-PENDING (Nick Craig-Wood)+ - Fix MkdirMetadata finding directories outside the root through a planted symlink (Nick Craig-Wood)+ - Fix Remove and Rmdir deleting outside the root through a planted symlink (Nick Craig-Wood)+ - Fix DirMove creating the destination parent through a planted symlink (Nick Craig-Wood)+ - netstorage: Fix API auth headers leaking to other hosts on redirect GHSA-m97j-7x65-wh93 CVE-PENDING (Nick Craig-Wood)+ - rc+ - Limit request size when the rc server has no authentication GHSA-f8p4-8g4j-q3m2 CVE-PENDING (Nick Craig-Wood)+ - Reject unauthorised and unknown calls before reading the request GHSA-f8p4-8g4j-q3m2 CVE-PENDING (Nick Craig-Wood)+ - seafile: Fix HTTPS to HTTP redirects leaking credentials GHSA-m97j-7x65-wh93 CVE-PENDING (Nick Craig-Wood)+ - serve: Make auth proxy revocation take effect within 5 minutes GHSA-vmgx-g9c5-669p CVE-PENDING (Nick Craig-Wood)+ - serve sftp+ - Fix auth proxy letting a public key be used as a password GHSA-66qv-66mg-9h74 CVE-PENDING (Nick Craig-Wood)+ - Fix auth proxy running for public keys the client can't use GHSA-66qv-66mg-9h74 CVE-PENDING (Nick Craig-Wood)+ - swift+ - Fix auth token leaking to other hosts or over plaintext on redirect GHSA-m97j-7x65-wh93 CVE-PENDING (Nick Craig-Wood)+ - Fix password being resent to another host on auth redirect GHSA-m97j-7x65-wh93 CVE-PENDING (Nick Craig-Wood)+ - ulozto: Fix auth tokens leaking on redirects from download links GHSA-3xpx-hjg2-vrjh CVE-PENDING (Nick Craig-Wood)+- New Features+ - accounting+ - Stop `--max-transfer` overshooting the limit (Nick Craig-Wood)+ - Add "Updated dirs" count of directory modtime/metadata updates to the stats (Nick Craig-Wood)+ - batcher: Log when committing uploads has finished (phatlc)+ - bisync+ - Make `--max-delete` aware of `--track-renames` (Morax)+ - Fall back to `--download` when a checksum is missing (nielash)+ - Reject `--max-delete` above 100 (nielash)+ - build+ - Update to go1.27 and make go1.26 the minimum required version (Nick Craig-Wood)+ - Update all dependencies (Nick Craig-Wood)+ - cmd: Complete enum flag values from their choices (Demiao Huang)+ - config: Only run `--password-command` once when using `--daemon` (phatlc)+ - doc fixes (ashwinibailangadi, ferrumclaudepilgrim, frodleTian, FTCHD, gomitrah, Nick Craig-Wood, qodelicious, Satvik Mishra, ZainnQureshii)+ - docker: Make container user UID and GID configurable via build ARGs (Vaibhav Mashal)+ - filter+ - Speed up large `--include-from` lists of directories (kaii9)+ - Make `[!...]` in globs a negated character class (kevin9327)+ - Make `[^...]` in globs exclude `/` and document it (kevin9327)+ - fs+ - Reject sizes too large for SizeSuffix instead of silently disabling the limit (KBS)+ - Stop writing config for on the fly and environment variable remotes into the config file (Nick Craig-Wood)+ - Reject durations like 300y that overflow instead of wrapping (kevin9327)+ - Accept multi-unit relative times like 1d12h in Time options (kevin)+ - Reject NaN sizes and CountSuffix values that overflow (kevin9327)+ - Reject `--bwlimit` times without a colon or with signed digits (kevin9327)+ - Reject NaN durations instead of wrapping (Soujas K)+ - Add MIME type for JPEG XL (.jxl) files (Seth Westphal)+ - gui+ - Take the GUI out of beta (Nick Craig-Wood)+ - Update embedded release to 1.2.1 (Nick Craig-Wood)+ - hash: Add CRC-32C (Castagnoli) support (Dirk Petersen)+ - index: Add partial runs which only re-index changed directories (Nick Craig-Wood)+ - lib/http+ - Return 416 for ranges `--rc-serve` and serve restic can't satisfy (Rohit Behera)+ - Replace `github.com/abbot/go-http-auth` with built in htpasswd support (Nick Craig-Wood)+ - Accept the Basic auth scheme in any case (kevin)+ - lib/proxy: Send the decoded user and password in Proxy-Authorization (kevin)+ - lib/transform: Make titlecase capitalise words instead of upper-casing them (kevin)+ - ncdu: Show progress and the item count when deleting a selection (Lev Devaev)+ - operations+ - Check checksums in rcat with known size (phatlc)+ - Speed up checksum by hashing files in parallel (Teddy Tennant)+ - Fall back to a single stream when a multi-thread source ignores ranges (Morax)+ - Show directory operations in the `--progress` display (Nick Craig-Wood)+ - Add `operations/getfile` remote control API endpoint (Hakan İSMAİL)+ - Make `--immutable` work with copyto, moveto and single file copy (phatlc)+ - Respect `--backup-dir` in deletefile (Cao Yuhang)+ - rc+ - Deprecate `--rc-web-gui` flag (Nick Craig-Wood)+ - Add sync report parameters to `sync/sync`, `sync/copy` and `sync/move` (Nick Craig-Wood)+ - rest: Reject negative sizes in HTTP response headers (Yousef Issa)+ - serve http+ - Sort directory listings in the browser (Nick Craig-Wood)+ - Redesign the directory listing page (Nick Craig-Wood)+ - serve s3+ - Report the object's modtime as Last-Modified after a PUT (kevin9327)+ - Support changing object tags (Nick Craig-Wood)+ - Quote the ETags of objects in listings (Nick Craig-Wood)+ - Support listing with delimiters other than "/" (Nick Craig-Wood)+ - Support directory markers and refuse file/directory conflicts (Nick Craig-Wood)+ - Make the ETag of multipart uploads agree with HEAD and support partNumber (Nick Craig-Wood)+ - Make conditional writes atomic against other writes through the server (Nick Craig-Wood)+ - Make the ETag returned on upload agree with HEAD for any `--etag-hash` (Nick Craig-Wood)+ - List empty directories as directory markers (Nick Craig-Wood)+ - Allow copying directory markers (Nick Craig-Wood)+ - Remove the directories made for a failed upload (Nick Craig-Wood)+ - Keep the MD5 ETag of objects on remotes without MD5 (Nick Craig-Wood)+ - Don't read directories needlessly when listing with a delimiter (Nick Craig-Wood)+ - Don't compute the MD5 of multipart uploads unless the ETag is MD5 (Nick Craig-Wood)+ - List the directory markers put for directories holding other keys (Nick Craig-Wood)+ - Let a failed CompleteMultipartUpload be retried (Nick Craig-Wood)+ - List objects with the storage class they were put with (Nick Craig-Wood)+ - Add `--directory-markers` to list every directory as a marker (Nick Craig-Wood)+ - Remove a directory with its last key after its marker is deleted (Nick Craig-Wood)+ - Make `--no-cleanup` a deprecated alias of `--directory-markers` (Nick Craig-Wood)+ - Document which S3 features are supported and the design goals (Nick Craig-Wood)+ - serve webdav+ - Sort directory listings in the browser (Nick Craig-Wood)+ - Redesign the directory listing page (Nick Craig-Wood)+ - size: Include source in JSON output (Jiacheng Xu)+ - sync: Add pattern-based transfer ordering (Shurong Cao)+ - touch: Parse documented timestamp formats (Nicholas Velten)+- Bug Fixes+ - accounting: Fix totalBytes counting server-side copies and moves twice (ilya lavrenov)+ - bisync+ - Fix conflict winner missing from listings when Path2 wins (nielash)+ - Fix `--conflict-resolve` comparing a zero size or modtime (nielash)+ - Fix endless retry loop for skipped files (nielash)+ - Fix listings ignoring a modtime less than 1s older (nielash)+ - Fix rc ignoring global `--dry-run` and `--max-delete` settings (nielash)+ - Fix rc ignoring legacy backupdir1/2 and invalid values (nielash)+ - Fix internal error when a conflict rename reuses deleted names (TastyHeadphones)+ - fs: Fix durations like 1d12h being rejected (Rohit Behera)+ - lib/encoder: Fix decoding a trailing character after an escaped quote (kevin9327)+ - list: Fix ext sort error sometimes being ignored at the end of a listing (Nick Craig-Wood)+ - ncdu: Update to tcell v3 which fixes duplicated keys on Windows (Nick Craig-Wood)+ - operations+ - Fix making a directory with modtime asking twice with `--interactive` (Nick Craig-Wood)+ - Fix case- or Unicode-only renames deleting the file (nielash)+ - Fix moving a file onto its own name deleting it (nielash)+ - Fix cat `--offset` before the start of a file printing its end (b84330808)+ - Fix oversized getfile tail requests (唐僧洗发用飘柔)+ - rc: Fix LogLevel in `_config` being ignored (Nick Craig-Wood)+ - rest: Fix retries reading the request body after the request has finished (Nick Craig-Wood)+ - serve dlna: Fix crash on Browse with a negative StartingIndex or RequestedCount (phatlc)+ - serve restic: Fix temporary errors confusing restic into leaving locks behind (Azhar)+ - serve s3+ - Update to gofakes3 v0.0.8 which fixes 500 errors with go1.27 (Nick Craig-Wood)+ - Upgrade gofakes3 to v0.0.11 to fix IsTruncated and Docker Distribution Registry (DEVN)+ - Fix buckets being undeletable after an aborted multipart upload (Nick Craig-Wood)+ - Fix CopyObject metadata including the source's and an mtime (Nick Craig-Wood)+ - serve sftp: Fix rmdir deleting a file instead of returning an error (phatlc)+ - sync+ - Fix directory modtimes being set twice (Nick Craig-Wood)+ - Fix `--delete-empty-src-dirs` disabled by a prior error in a reused stats group (ghwany)+ - walk: Fix duplicate directories with `--fast-list` when directories are listed after their contents (Nick Craig-Wood)+- Mount+ - Fix VFS instance leaks on mount failures and unmounts (Hakan İSMAİL)+- VFS+ - Preserve shrinking truncate on uncached files (47th)+ - Fix crash renaming a directory while the dir cache expires (Shreyansh Jain)+ - Fix virtual entries leak in removed directory (Biagio Peccerillo)+ - Fix renamed items listed twice after a case or Unicode-only match (nielash)+ - Fix uploads never starting while a dirty file keeps being read (Neil Cawse)+- Local+ - Speed up default checksummed copies by writing in larger blocks (Dominik Sander)+ - Make out of space errors fatal during multi-thread transfers (ferrumclaudepilgrim)+ - Stop `--copy-links` following symlink loops (phatlc)+ - Limit the size of symlink targets read from .rclonelink files (Nick Craig-Wood)+ - Fix server-side copy on macOS when the destination exists (Ayush Sharma)+ - Set only the ownership a non-root user may set (Dirk Petersen)+- Crypt+ - Fix nil pointer panic in ObjectInfo.Hash (fixfon)+- Archive+ - Read symlinks stored in zip and squashfs archives (Lev Devaev)+ - Fix reading an archive in the root of a remote (Oshino Penguin)+- Azure Blob+ - Enable on Solaris (acoeur)+ - Fix directories on storage accounts with hierarchical namespace (Nick Craig-Wood)+ - Fix NewObject on a directory returning "is not a regular file" (Nick Craig-Wood)+ - Fix listing a file as a directory on hierarchical namespace accounts (Nick Craig-Wood)+ - Fix sync at the root failing on soft deleted containers (Ludovic Bondon)+- Box+ - Reuse multipart upload part buffers via the global pool (Nick Craig-Wood)+ - Fix ~ and environment variables not being expanded in `box_config_file` (Nick Craig-Wood)+- Chunker+ - Clean stale chunks after server-side overwrite (hahahahahayesyeseys)+- Compress+ - Buffer small uploads in the global memory pool (Nick Craig-Wood)+ - Stop buffering the compressibility check output in memory (Nick Craig-Wood)+- Drime+ - Fix endless listing of directories with more than 20,000 entries (Maksym Tymoshyk)+- Drive+ - Fetch metadata permissions concurrently (Loi Nguyen)+ - Reuse resumable-upload chunk buffers via the global memory pool (SillyZir)+ - Stop a fast-list error from hanging the sync (Sash)+- Dropbox+ - Reuse upload chunk buffers via the global memory pool (Nick Craig-Wood)+- Filelu+ - Reuse multipart upload buffers via the global pool and retry failed parts (Nick Craig-Wood)+ - Encode multipart init paths (wjc)+- FTP+ - Discard connections that fail with non-FTP errors (Daniel Bae)+ - Fix ranged reads failing when the server answers 425 to an early close (Robert Kubiś)+- Hasher+ - Keep cached hash when only the modtime changes (Aniket)+- Hidrive+ - Stop copying every upload chunk a second time (Nick Craig-Wood)+ - Buffer the start of each upload from the global memory pool (Nick Craig-Wood)+ - Reuse chunked upload buffers via the global memory pool (Nick Craig-Wood)+ - Buffer small file updates from the global memory pool (Nick Craig-Wood)+ - Fix truncated file left behind when an upload hits `--max-transfer` (Nick Craig-Wood)+- HTTP+ - Add metadata defaults for `--http-no-head` (Giridhar)+- Huaweidrive+ - Build small uploads in pooled memory instead of a growing buffer (Nick Craig-Wood)+ - Reuse resumable-upload chunk buffers via the global pool (Nick Craig-Wood)+ - Fix creating, copying and renaming files with non-ASCII names (Nick Craig-Wood)+- Iclouddrive+ - Mark the backend as case insensitive (berce)+- Imagekit+ - Stop multipart writers after failed uploads (Florian Josef Altendorfer)+- Jottacloud+ - Buffer MD5 pre-reads in the global memory pool (Nick Craig-Wood)+ - Fix server-side copy onto a trashed file failing as corrupted (Nick Craig-Wood)+ - Fix server-side copy onto a trashed file keeping the old name case (Nick Craig-Wood)+ - Fix authentication with whitelabel services elkjop and elko (albertony)+ - Fix traditional authentication failing with invalid_client (albertony)+- Linkbox+ - Buffer the hashed 10 MiB file prefix in the global memory pool (Nick Craig-Wood)+- Mailru+ - Buffer speedup hashing in the global memory pool and fix upload retries (Nick Craig-Wood)+- Onedrive+ - OneDrive for Business and SharePoint users must upgrade to v1.76+ before April 1, 2027 (Nick Craig-Wood)+ - Authenticate SharePoint downloads, uploads and copies after tempauth retirement (Nick Craig-Wood)+ - Return an error if a ranged download doesn't return the requested range (Nick Craig-Wood)+ - Reuse upload chunk buffers via the global memory pool (Nick Craig-Wood)+ - Allow `--onedrive-upload-cutoff` up to the documented limit of 250 MiB (Can Arslan)+ - Add configurable SharePoint tenant API version (Mattias Michaux)+ - Make `--onedrive-delta` quicker when not listing from the drive root (Nick Craig-Wood)+ - Keep the delta token when a ChangeNotify poll fails (xiaozou-wine)+ - Show the size of directories in listings (GingerAdonis)+ - Poll async copy status with the configured HTTP client (Soujas K)+- Opendrive+ - Reuse upload chunk buffers via the global memory pool (Nick Craig-Wood)+- Oracleobjectstorage+ - Add server-side object moves (tomaszni)+ - Encode carriage returns and line feeds (tomaszni)+ - Purge objects in batches (tomaszni)+ - Retry 409 ConcurrentObjectUpdate errors (berce)+- Pcloud+ - Stop multipart writers after failed uploads (Florian Josef Altendorfer)+- Pikpak+ - Share upload chunk buffers with the global memory pool (Nick Craig-Wood)+ - Use pooled memory for the gcid hash buffer and spool unknown-size uploads to disk (Nick Craig-Wood)+ - Fix downloads that return an HTTP error page as the file (Cestercian)+ - Stop multipart writers after failed uploads (Florian Josef Altendorfer)+ - Fix uploads failing when the server is slow to complete them (Nick Craig-Wood)+- Putio+ - Reuse upload chunk buffers via the global memory pool (Nick Craig-Wood)+- S3+ - Avoid buffering Object Lock uploads when the source MD5 is known (Nick Craig-Wood)+ - Add backend link command with signed response header overrides (eliotee)+ - Add Fastly EU Central 1 region (Leon Brocard)+ - Add the MD5 of multipart chunks to the debug log (interested.tortoise)+ - Report the real storage class when an object needs restoring (Dirk Petersen)+ - Speed up multi-thread uploads by checksumming chunks in 1 MiB reads (Roman Butor)+ - Add IT-MIL region to Scaleway (Alessio Margelli)+ - Add missing Huawei OBS regions (Enbiya Göral)+ - Fix rmdir of the root directory with `--s3-directory-markers` (Nick Craig-Wood)+ - Fix presigned URLs for GCS, Linode and Ceph (Leon Brocard)+- Seafile+ - Fix operations on files whose names are not in Unicode NFC form (Fred)+ - Stop multipart writers after failed uploads (Florian Josef Altendorfer)+ - Fix double slash in the URL when getting a 2FA token (Nick Craig-Wood)+- SFTP+ - Implement multi-thread uploads (Splainte)+ - Fix data race when using `--sftp-ssh` (Dirk Petersen)+ - Fix hash check for names starting with "-" with relative `--sftp-path-override` (Dirk Einecke)+ - Fix sync deleting files with control characters in their names (nielash)+ - Fix hashing files with a newline in their name on rsync.net (Nick Craig-Wood)+- Shade+ - Stop copying each upload chunk into a second heap buffer (Nick Craig-Wood)+- Smb+ - Add workstation field to smb backend (Kalin Stoyanov)+- Swift+ - Add Infomaniak Swiss Backup provider (Louis)+- WebDAV+ - Reject ranged responses which ignore the range (Morax)+ - Retry Nextcloud chunk uploads individually (Stochastic)+ - Fix Nextcloud public share access by sending X-Requested-With (drakeo338)+ - Fix memory leak uploading to Infinite Scale (jxj)+- Yandex+ - Add `app_folder` option to support `cloud_api:disk.app_folder` OAuth scope (Sanjay Kanth A)+- Zoho+ - Stop multipart writers after failed uploads (Florian Josef Altendorfer)++## v1.75.2 - 2026-10-09++[See commits](https://github.com/rclone/rclone/compare/v1.75.1...v1.75.2)++- Security+ - b2: Fix account token and SSE-C key leaking on redirects from `--b2-download-url` GHSA-gpw7-77cq-9f72 CVE-2026-116196 (Nick Craig-Wood)+ - build+ - Fix multiple CVEs by upgrading to go1.26.9 (Nick Craig-Wood)+ - CVE-2026-97032: net/http: HTTP/2 server crash due to HPACK encoder race+ - CVE-2026-78659: net/http: HTTP/2 server memory exhaustion due to Trailer headers+ - CVE-2026-97031: crypto/tls: reject malformed ECH outer extension references+ - CVE-2026-94444: cmd/go: checksum bypass for golang.org/fips140+ - CVE-2026-94447: cmd/go: checksum database bypass for golang.org/toolchain+ - CVE-2026-94448: html/template: reset context tracking on consecutive template expressions+ - CVE-2026-97030: html/template: recognize |yield| as regexp preceder keyword+ - CVE-2026-94440: net/textproto, mime/multipart: memory limit bypass when parsing MIME headers+ - CVE-2026-56866: net/http: HTTP/1 client connection desynchronization after CONNECT rejection+ - CVE-2026-94439: net/http: HTTP/1 server connection desynchronization after 2xx CONNECT response+ - CVE-2026-78669: net/http: excessive CPU consumption from repeated initial window changes+ - CVE-2026-78660: net/http: HTTP/2 transport accepts malformed framing-related headers+ - CVE-2026-56857: os: Root.Mkdir(All) can follow junctions out of the root on Windows+ - CVE-2026-78667: net/http: lack of limit on size of parsed Range headers+ - CVE-2026-78663: net/http: double flow control refund on HTTP/2 server streams+ - Update golang.org/x/net to v0.60.0 to fix multiple CVEs (Nick Craig-Wood)+ - CVE-2026-78659: net/http: HTTP/2 server memory exhaustion due to Trailer headers+ - CVE-2026-78669: net/http: excessive CPU consumption from repeated initial window changes+ - CVE-2026-78663: net/http: double flow control refund on HTTP/2 server streams+ - CVE-2026-97032: net/http: HTTP/2 server crash due to HPACK encoder race+ - Update google.golang.org/grpc to fix CVE-2026-84445 (dependabot[bot])+ - CVE-2026-84445: xds: server panic on a request missing both :authority and Host headers+ - fshttp: Drop `--header` values over HTTP after HTTPS downgrade GHSA-rrwf-2qr6-p832 CVE-PENDING (Nick Craig-Wood)+ - http: Stop a path starting with // moving the remote to another host GHSA-xqmq-v4wj-rw5m CVE-2026-116192 (Nick Craig-Wood)+ - lib/http: Require authentication for OPTIONS requests GHSA-gj73-fh6v-92fj CVE-2026-116193 (Nick Craig-Wood)+ - local: Fix .rclonelink names escaping the root with `--links` GHSA-55qp-jrwr-x827 CVE-2026-116195 (Nick Craig-Wood)+ - netstorage+ - Stop a path starting with // moving the remote to another host GHSA-xqmq-v4wj-rw5m CVE-2026-116192 (Nick Craig-Wood)+ - Stop a listing resume token moving signed requests to another host GHSA-xqmq-v4wj-rw5m CVE-2026-116192 (Nick Craig-Wood)+ - pcloud+ - Fix OAuth config accepting callbacks without state GHSA-hvpr-p4pv-4f46 CVE-2026-116194 (Nick Craig-Wood)+ - Only accept pCloud hosts from the hostname in the OAuth callback GHSA-hvpr-p4pv-4f46 (Nick Craig-Wood)+ - selfupdate+ - Fix hash check trusting unsigned data in the signed SHA256SUMS GHSA-w33h-3qgq-8mv8 CVE-2026-116189 (Nick Craig-Wood)+ - Fix release verification instructions trusting unsigned hashes GHSA-w33h-3qgq-8mv8 (Nick Craig-Wood)+ - serve nfs: Fix path traversal by refusing unsafe file names in the VFS GHSA-44pm-2q5p-fm63 CVE-2026-116191 (Nick Craig-Wood)+ - serve s3+ - Fix presigned URL being turned into a copy of any object GHSA-w3h7-jxxq-vfrp CVE-PENDING (Nick Craig-Wood)+ - Stop auth proxy users listing or using each other's multipart uploads and metadata GHSA-35g6-9fjv-rmx5 CVE-2026-116187 (Nick Craig-Wood)+ - Fix multipart buffer limits being bypassed by overflowing part sizes GHSA-qjw8-8hmp-c3m6 CVE-2026-116188 (Nick Craig-Wood)+ - Add `--multipart-max-uploads` to limit uploads in progress (Nick Craig-Wood)+ - Add `--multipart-streaming-buffer-total` to limit memory across uploads (Nick Craig-Wood)+ - Add `--metadata-max-objects` to limit the metadata kept in memory (Nick Craig-Wood)+ - serve webdav: Fix path traversal above the served directory GHSA-44pm-2q5p-fm63 CVE-2026-116191 (Nick Craig-Wood)+ - webdav: Fix headers option leaking to other hosts on redirect GHSA-3rqf-7h69-gg5v CVE-2026-116190 (Nick Craig-Wood)+- Bug Fixes+ - accounting+ - Stop averaging when the last check finishes (Kunpeng Xie)+ - Keep speed averaging active when transfers overlap (kaii9)+ - dedupe: Fix rename mode giving up after 100 names and make it faster (Nick Craig-Wood)+ - doc fixes (Dhevenddra K G, enkvadrat, GhostCoder6969, James Truitt, maximilize, n4n5, Nick Craig-Wood, PSR94, Rohit Behera, shaurya, subomi)+ - docker: Fix files uploaded with the wrong mime type (hsdfat)+ - fs+ - Fix about showing a negative total when a quota reaches the int64 maximum (youdie006)+ - Fix bandwidth limits below 1 KiB being multiplied by 1024 (Roland)+ - Fix `--bwlimit` timetables written out of order using the wrong limit (Rohit Behera)+ - Make BwTimetable.Set replace the timetable instead of appending to it (youdie006)+ - Retry "network is unreachable" and "network is down" errors (nielash)+ - fserrors: Fix out of space detection on Windows (ferrumclaudepilgrim)+ - lib/rest: Limit the size of HTTP response bodies read into memory (Nick Craig-Wood)+ - log: Fix race when adding a log output while logging (Nick Craig-Wood)+ - oauthutil: Fix Renew.Shutdown nil timer panic and race (TastyHeadphones)+ - operations+ - Stop `--copy-dest` replacing files with `--immutable` (phatlc)+ - Fix ignored error in rcat probe reads (Wang Chencheng)+ - Fix hang when deleting files and a fatal error occurs (Nick Craig-Wood)+ - Fix hang and lost context when moving a directory file by file (Nick Craig-Wood)+ - Don't sleep for a Retry-After error when the transfer is cancelled (Nick Craig-Wood)+ - rc+ - Make job/status and job/list require authentication (Nick Craig-Wood)+ - Fix rc API accepting an out of range number and overflowing 64 bits (KBS)+ - Reject out of range integer parameters instead of truncating them (Nick Craig-Wood)+ - Fix large numeric parameters being rejected on 32 bit builds (Jeremy Schoemaker)+ - selfupdate: Fix `--version X.Y` depending on ./ in the download site links (Nick Craig-Wood)+ - serve+ - Fix `--auth-proxy` stopping VFS cache uploads after 5 minutes (Nick Craig-Wood)+ - Fix `--auth-proxy` backends never being shut down after simultaneous logins (Nick Craig-Wood)+ - serve dlna: Log unescaped paths (phatlc)+ - serve docker: Fix volume path being lost when the plugin restarts (phatlc)+ - serve ftp+ - Fix transfers failing after 5 minutes with `--auth-proxy` (Nick Craig-Wood)+ - Fix VFS leak when the server fails to start (Nick Craig-Wood)+ - serve http+ - Fix downloads failing after 5 minutes with `--auth-proxy` (Nick Craig-Wood)+ - Escape filename in zip download Content-Disposition header (Harsh Raj Singhania)+ - serve restic+ - Fix repositories named data (kaii9)+ - Prevent concurrent append-only overwrites (Vasek Sraier)+ - serve s3+ - Apply the multipart upload limits and `--metadata-max-objects` to each user separately (Nick Craig-Wood)+ - Stream in-order multipart parts without buffering them (Nick Craig-Wood)+ - Start the upload to the backend on the first multipart part (Nick Craig-Wood)+ - Don't buffer re-uploads of multipart parts already streamed (Nick Craig-Wood)+ - Count buffered multipart parts in whole memory pages (Nick Craig-Wood)+ - Time out multipart parts waiting for buffer space (Nick Craig-Wood)+ - Limit the memory requests can use (Nick Craig-Wood)+ - Forget expired multipart uploads completely (Nick Craig-Wood)+ - Forget the metadata of deleted objects (Nick Craig-Wood)+ - List objects lazily so paging a deep hierarchy is fast (Nick Craig-Wood)+ - Fix multipart upload corruption if the backend write of an early part fails (Nick Craig-Wood)+ - Fix uploads failing when the auth proxy expires the user's backend (Nick Craig-Wood)+ - Fix stale metadata being returned for objects changed elsewhere (Nick Craig-Wood)+ - Fix `--etag-hash auto` crashing or using the wrong hash with `--auth-proxy` (Nick Craig-Wood)+ - Fix CopyObject of a missing object not returning NoSuchKey (Nick Craig-Wood)+ - Fix log messages with bad format strings (Nick Craig-Wood)+ - Document memory use and bring the docs up to date (Nick Craig-Wood)+ - serve webdav+ - Fix transfers failing after 5 minutes with `--auth-proxy` (Nick Craig-Wood)+ - Fix crash when the server fails to start (Nick Craig-Wood)+ - Escape filename in zip download Content-Disposition header (Harsh Raj Singhania)+ - vfscache: Fix hang when the cache cleaner is disabled (ferrumclaudepilgrim)+- Mount+ - Fix `--daemon` failing when `--rc` is enabled (somaz)+ - Ignore com.apple.* xattrs on the macOS FSKit backend (Eugene)+- VFS+ - Fix reads returning zeros after a file changes during handle caching (Neil Cawse)+ - Stop setting a file's modtime from discarding its cached data (Neil Cawse)+ - Fix crash reading the metadata of a file which is being written (Nick Craig-Wood)+ - Fix a VFS being reused while it is being shut down (Nick Craig-Wood)+ - Fix AddVirtual ignoring isDir (phatlc)+ - Stop `--links` objects called "...rclonelink" appearing as ".." in directory listings (Nick Craig-Wood)+- Local+ - Fix directories named *.rclonelink being treated as links with `--links` (Nick Craig-Wood)+- Archive+ - Fix corrupt listings when listing a zip directory more than once (Nick Craig-Wood)+ - Fix zip file entries named for a directory causing confusion (Nick Craig-Wood)+ - Fix listing entries with a leading slash as if they were in the root (Nick Craig-Wood)+ - Fix crash when creating archive to stdout (Vladimir Babin)+ - Fix a squashfs path to a single file listing its whole directory (Nick Craig-Wood)+- Azure Blob+ - Use the released Azure SDK for Apache Arrow listing (Nick Craig-Wood)+- Box+ - Fix log messages with bad format strings (Nick Craig-Wood)+- Chunker+ - Fix panic listing a directory after an interrupted upload with meta_format none (Nick Craig-Wood)+- Compress+ - Fix crash on ranged reads when gzip metadata is corrupted (Acts1631)+- Drime+ - Fix server-side copy over an existing file leaving a "name (1)" copy (Nick Craig-Wood)+ - Fix deleted files and directories still being listed with hard_delete (Nick Craig-Wood)+- Drive+ - Document the Branding step needed to publish your own client_id (Sanjay Kanth A)+- Dropbox+ - Fix shared folder mount for roots nested more than one level deep (phatlc)+ - Match shared-folder and received-file names case-insensitively (phatlc)+- Filescom+ - Fix corrupted uploads after a retried upload error (Nick Craig-Wood)+ - Document that names with / or \ in can't be stored (Nick Craig-Wood)+- HTTP+ - Fix crash when the server fails to start (Nick Craig-Wood)+- Iclouddrive+ - Fix potential crash looking up items (Nick Craig-Wood)+- Imagekit+ - Fix potential crash in rmdir and purge (Nick Craig-Wood)+- Internxt+ - Fix lookups of files starting with a dot and dropped uploads (jzunigax2)+ - Fix sync with `--backup-dir` or `--suffix` deleting the backed up file (Nick Craig-Wood)+ - Fix server-side moves failing with "Not Found" or "already exists" (Nick Craig-Wood)+ - Fix "directory not empty" and stale directories after moves and deletes (Nick Craig-Wood)+ - Fix server-side directory move failing after a gateway timeout (Nick Craig-Wood)+- Onedrive+ - Update docs for versions, links and time precision on personal accounts (Nick Craig-Wood)+ - Document that shared with me shortcuts may fail to list on personal (Nick Craig-Wood)+ - Update `--onedrive-hard-delete` docs to mention personal account support (foecmke)+- Oracleobjectstorage+ - Fix SSE-C server-side copies (tomaszni)+ - Upload empty streams without multipart (tomaszni)+- Pixeldrain+ - Fix corrupted uploads after a retried upload error (Nick Craig-Wood)+- Premiumizeme+ - Fix uploading files with ";" in their names (Nick Craig-Wood)+- S3+ - Disable signing Accept-Encoding for Ceph and Linode (Aditya)+ - Fix version-at listings with URL encoded keys (jxj)+ - Fix storage class missing from `--s3-versions` listings (Dirk Petersen)+- Seafile+ - Fix corrupted uploads after a retried upload error (Nick Craig-Wood)+- SFTP+ - Fix failed uploads leaving disk space in use on the server (Nick Craig-Wood)+- Shade+ - Fix potential crash in directory move (Nick Craig-Wood)+- Smb+ - Save the user name in the config even if it matches the current user (phatlc)+- WebDAV+ - Fix duplicated listing entries after retried PROPFIND (ZRHann)+ - Fix "XML syntax error" on paths which don't exist with ownCloud 10.16 (Nick Craig-Wood)++## v1.75.1 - 2026-09-04++[See commits](https://github.com/rclone/rclone/compare/v1.75.0...v1.75.1)++- Security+ - archive+ - Fix zip slip path traversal in untrusted zip files GHSA-66hp-wgxq-6f5q CVE-2026-88014 (Nick Craig-Wood)+ - Hide any archive entry which escapes the directory being listed GHSA-66hp-wgxq-6f5q (Nick Craig-Wood)+ - Reject unsafe entry names when mounting squashfs images GHSA-66hp-wgxq-6f5q (Nick Craig-Wood)+ - Fix zip subdirectory root matching sibling directories GHSA-66hp-wgxq-6f5q (Nick Craig-Wood)+ - Fix zip entry named "." hiding every other file GHSA-66hp-wgxq-6f5q (Nick Craig-Wood)+ - Fix "directory not found" for archive paths containing "./" or "//" GHSA-66hp-wgxq-6f5q (Nick Craig-Wood)+ - build+ - Fix multiple CVEs by upgrading to go1.26.6 (Nick Craig-Wood)+ - CVE-2026-56860: net/url: quadratic complexity in resolvePath+ - CVE-2026-56858: html/template: JavaScript regexp context tracking+ - CVE-2026-56862: crypto/tls: limit handshake messages accepted post-handshake+ - CVE-2026-56853: net/http: apply ReadHeaderTimeout to unencrypted HTTP/2 check+ - CVE-2026-56859: encoding/xml: recursion depth guard during decode+ - CVE-2026-33818: encoding/asn1: enforce maximum recursion depth+ - CVE-2026-46600: net: panic parsing an invalid SVCB or HTTPS RR in dnsmessage+ - CVE-2026-39821: net/http: reject ASCII-only Punycode-encoded labels in idna+ - Update golang.org/x/crypto to v0.56.0 to fix multiple CVEs (Nick Craig-Wood)+ - CVE-2026-56854: ssh: source-address critical option not enforced for non-public-key auth callbacks+ - CVE-2026-78662: ssh: a malicious peer could flood an undecided channel's incoming requests, deadlocking the connection+ - CVE-2026-56855: ssh: a malicious peer could send crafted messages on an established channel, deadlocking the connection+ - Update golang.org/x/image to v0.45.0 to fix CVE-2026-46603 (Nick Craig-Wood)+ - CVE-2026-46603: excessive memory allocation during VP8L decoding+ - fs: Confine directory listing entries that escape the root GHSA-3vxh-3pcx-9m8q CVE-2026-93986 GHSA-38xv-hf3p-h7mq CVE-2026-88046 (Nick Craig-Wood)+ - fshttp: Don't send `--header` values to other hosts on redirect GHSA-486v-q2wf-fp2r CVE-2026-88013 (Nick Craig-Wood)+ - http: Don't leak configured headers to other hosts or over plaintext on redirect GHSA-486v-q2wf-fp2r CVE-2026-88013 (Nick Craig-Wood)+ - lib/rest: Check HTTPS downgrades against the original request on redirect GHSA-486v-q2wf-fp2r CVE-2026-88013 (Nick Craig-Wood)+ - local+ - Fix dir metadata escaping the root through a planted symlink GHSA-f8g7-2xjc-7mfh CVE-2026-88016 (Nick Craig-Wood)+ - Fix btime escaping the root via a planted symlink GHSA-f8g7-2xjc-7mfh CVE-2026-88016 (Nick Craig-Wood)+ - Fix panic on Range request past the end of a symlink GHSA-p6m2-r3w9-mpxw CVE-2026-88015 (Nick Craig-Wood)+ - serve docker+ - Reject volume names that escape the base directory GHSA-p6vx-hf7p-98j6 CVE-2026-93987 (Nick Craig-Wood)+ - Reject volume names resolving to the base directory itself GHSA-p6vx-hf7p-98j6 CVE-2026-93987 (Nick Craig-Wood)+ - Re-derive volume mountpoint from name when restoring state GHSA-p6vx-hf7p-98j6 CVE-2026-93987 (Nick Craig-Wood)+ - serve ftp: Fix auth-proxy sessions sharing credentials by username GHSA-c476-6w5q-jw77 CVE-2026-88017 (Nick Craig-Wood)+ - serve s3+ - Fix memory exhaustion from client-declared multipart part size GHSA-2p48-j3qc-rx9f CVE-2026-88045 (Nick Craig-Wood)+ - Reject bogus multipart part sizes in the reorder buffer GHSA-2p48-j3qc-rx9f (Nick Craig-Wood)+ - Fix auth proxy accepting any request signed with an empty secret GHSA-xwwr-4h3p-r22c CVE-2026-88018 (Nick Craig-Wood)+ - **NB** the auth proxy protocol for `serve s3` has changed - the proxy program is now given the access key ID as `user` and must return the secret as `_secret_access_key`+ - Fix each server accepting the `--auth-key` credentials of all the others (Nick Craig-Wood)+ - Fix misleading anonymous access log when using an auth proxy via rc GHSA-p569-5gjg-9cmj CVE-2026-88044 (Nick Craig-Wood)+ - serve sftp: Fix auth proxy configured via rc being silently ignored GHSA-p569-5gjg-9cmj CVE-2026-88044 (Nick Craig-Wood)+- Bug Fixes+ - accounting+ - Fix memory leak on long-running rcd (nielash)+ - Fix memory leak from stats groups on long-running rcd (nielash)+ - Fix bwlimit burst overflow (Rayan Salhab)+ - bisync+ - Fix memory leak when running via the rc (nielash)+ - Fix failed transfers of empty files being recorded as synced (Nick Craig-Wood)+ - build: Make go1.26 the minimum required version as needed by golang.org/x/crypto v0.56.0 (Nick Craig-Wood)+ - config: Redact env var config values in logs (Pastalikek65)+ - doc fixes (Anton Karpov, CAOShurong, Dean Chen, Nick Craig-Wood, Recoordinate, Rodrigo Rodrigues, Shantanav Mukherjee, shaurya)+ - lib/batcher: Prevent commits racing shutdown (Loi Nguyen)+ - lib/transform: Fix panic in `truncate_keep_extension` (VXNCXNX)+ - multipart: Fix chunked uploads storing truncated objects when the source ends early (Nick Craig-Wood)+ - operations: Fix silent truncation of streaming uploads whose source ends early (Nick Craig-Wood)+ - serve+ - Fix VFS instance leaks on server startup failures and shutdown (Hakan İSMAİL)+ - Pass the client IP address to the auth proxy (am-at-enrollvb)+ - serve http: Prevent scrolling to the top on page reload (Sune Mølgaard)+ - serve nfs: Fix EIO when creating symlinks with `--vfs-links` (SillyZir)+ - serve s3+ - Fix failed uploads deleting or corrupting the object at the key (Nick Craig-Wood)+ - Fix crash when a multipart upload is aborted while a part is uploading (Nick Craig-Wood)+ - Fix modtime not being set when only mtime metadata is supplied on PUT (Nick Craig-Wood)+ - Upload all multipart uploads via the VFS so they obey `--bwlimit` and show in stats (Nick Craig-Wood)+ - Reserve the `.rclone_temp_` prefix for temporary objects (Nick Craig-Wood)+ - Clean up abandoned multipart uploads after `--multipart-expiry` (Nick Craig-Wood)+ - vfscache+ - Fix reader deadlock when the item size drops below the read offset (Dave)+ - Fix log message growing without bound on repeated write errors (Vijay Misal)+ - walk: Stop directory traversal when the context is cancelled (Rahman Yilmaz)+- VFS+ - Synchronize poll updates with shutdown (Loi Nguyen)+ - Make poll shutdown lifecycle deterministic (Loi Nguyen)+- Crypt+ - Fix hash mismatches with `no_data_encryption` on backends which check upload hashes (Nick Craig-Wood)+ - Fix directory names which look like versioned file names (TowyTowy)+ - Warn about directories with legacy version-like encrypted names (Nick Craig-Wood)+- Azure Blob+ - Fix Entra ID server-side copy source authentication (Edward Klesel)+ - Fix spurious vfs cache corruption errors during chunked reads (Nick Craig-Wood)+- Azurefiles+ - Fix zero padded files being created when the source ends early (Nick Craig-Wood)+- Box+ - Fix truncated files being uploaded successfully when the source ends early (Rohit Behera)+- Compress+ - Fix corrupted objects being created when the source ends early (Nick Craig-Wood)+- Drive+ - Don't list trashed files when removing a directory into the trash (alliasgher)+- Dropbox+ - Preserve Paper export paths on lookup (Loi Nguyen)+ - Fix context cancellation (e.g. `--max-duration` limit) not stopping in-flight requests (debaditya)+ - Fix chunked uploads of truncated files never finishing (Nick Craig-Wood)+ - Don't retry chunked upload requests when the upload has been cancelled (Nick Craig-Wood)+ - Decode received shared-file names (Sanjay Kanth A)+ - Fix ChangeNotify when the root's case differs from Dropbox's (Loi Nguyen)+- Filelu+ - Fix truncated files being uploaded successfully when the source ends early (Nick Craig-Wood)+ - Fix duplicate root path during multipart folder creation (kingston125)+- Huaweidrive+ - Fix truncated files being uploaded successfully when the source ends early (Rohit Behera)+- Iclouddrive+ - Fix uploads into an app container failing with 412 (Christian De Santis)+- Internetarchive+ - Fix corrupted files being created when the source ends early (Nick Craig-Wood)+- Internxt+ - Persist rotated token returned by the user info call (0rangeSeaW0lf)+- Onedrive+ - Fix 403 Forbidden for configuration personal onedrive (machsix)+ - Fall back to manual drive ID entry when drive listing fails (SillyZir)+ - Don't retry multipart upload chunk on 404 (upload session not found) (water)+- Overview+ - Fix "internal error: no overview data found" on 32 bit architectures (Nick Craig-Wood)+- Pikpak+ - Fix truncated files being created when the source ends early (Nick Craig-Wood)+ - Fix truncated single part uploads reported as ok when source ends early (Nick Craig-Wood)+- Protondrive+ - Fix files uploaded with v1.75.0 not being readable in the Proton apps (Nick Craig-Wood)+ - Fix corrupted uploads after a retried upload error (Nick Craig-Wood)+- Quatrix+ - Fix chunk upload retries and fix memory leak (Nick Craig-Wood)+- S3+ - Update Mega endpoints (Nick Craig-Wood)+ - Treat UploadPart success without ETag as retryable error (CAOShurong)+ - Fix server side copy failing with `--s3-no-head-object` (Anatoly Tarnavsky)+- Sia+ - Fix corrupted files being created when the source ends early (Nick Craig-Wood)+- Smb+ - Reuse the upload connection for SetModTime (alliasgher)+- WebDAV+ - Fix SetModTime failing and hashes missing on Nextcloud (Nick Craig-Wood)+- Yandex+ - Fix truncated files being uploaded successfully when the source ends early (Rohit Behera)+## v1.75.0 - 2026-07-31[See commits](https://github.com/rclone/rclone/compare/v1.74.0...v1.75.0)@@ -71490,26 +75342,26 @@ Options:- [Scality](https://rclone.org/s3/#scality) (RING / ARTESCA)- [Zero Services](https://rclone.org/s3/#zero-z3) (ZERO-Z3)- Security- - archive: Don't crash on malformed squashfs images GHSA-6jcg-q3wp-x2f4 CVE-PENDING (Nick Craig-Wood)- - ftp: Fix ftp command injection when encoding doesn't include CRLF GHSA-8c48-q9wj-3w37 CVE-PENDING (Nick Craig-Wood)+ - archive: Don't crash on malformed squashfs images GHSA-6jcg-q3wp-x2f4 CVE-2026-79775 (Nick Craig-Wood)+ - ftp: Fix ftp command injection when encoding doesn't include CRLF GHSA-8c48-q9wj-3w37 CVE-2026-71311 (Nick Craig-Wood)- lib/http: Use TLS on all `--addr` listeners when `--cert` and `--key` are set GHSA-mfvx-7rcj-9m5g (Nick Craig-Wood)- - lib/proxy: Fix unbounded HTTP CONNECT headers causing OOM GHSA-xhf4-832v-7xcr CVE-PENDING (Nick Craig-Wood)- - local: Stop source file names escaping the destination directory GHSA-7p4m-qxvv-g567 CVE-PENDING (Nick Craig-Wood)+ - lib/proxy: Fix unbounded HTTP CONNECT headers causing OOM GHSA-xhf4-832v-7xcr CVE-2026-71310 (Nick Craig-Wood)+ - local: Stop source file names escaping the destination directory GHSA-7p4m-qxvv-g567 CVE-2026-71313 (Nick Craig-Wood)- rc- - Don't expose pprof debug handlers on an unauthenticated server GHSA-mfvx-7rcj-9m5g CVE-PENDING (Nick Craig-Wood)+ - Don't expose pprof debug handlers on an unauthenticated server GHSA-mfvx-7rcj-9m5g CVE-2026-79776 (Nick Craig-Wood)- Require authentication to list the remotes with `--rc-serve` GHSA-mfvx-7rcj-9m5g (Nick Craig-Wood)- - Fix leaking stack traces on panics GHSA-gwfq-86j8-7qhv (Nick Craig-Wood)+ - Fix leaking stack traces on panics GHSA-gwfq-86j8-7qhv CVE-2026-79777 (Nick Craig-Wood)- s3- - Fix redirect credential leaks, reject HTTPS->HTTP and strip secrets GHSA-8mxv-9xhp-86h4 (Nick Craig-Wood)- - Strip S3 Express session token on cross-host redirects GHSA-8mxv-9xhp-86h4 (Nick Craig-Wood)+ - Fix redirect credential leaks, reject HTTPS->HTTP and strip secrets GHSA-8mxv-9xhp-86h4 CVE-2026-79780 (Nick Craig-Wood)+ - Strip S3 Express session token on cross-host redirects GHSA-8mxv-9xhp-86h4 CVE-2026-79780 (Nick Craig-Wood)- serve ftp: Use constant time comparison for password check GHSA-mfvx-7rcj-9m5g (Nick Craig-Wood)- - serve restic: Fix path traversal above the served directory GHSA-45pq-889g-fcgh CVE-PENDING (Nick Craig-Wood)- - serve sftp: Don't crash the whole server on a bad request GHSA-6jcg-q3wp-x2f4 (Nick Craig-Wood)- - sftp: Fix command injection via crafted filenames on PowerShell remotes GHSA-2m8m-jhrm-w6j2 CVE-PENDING (Nick Craig-Wood)- - vfs: Don't crash the process if a backend panics on a background goroutine GHSA-6jcg-q3wp-x2f4 (Nick Craig-Wood)+ - serve restic: Fix path traversal above the served directory GHSA-45pq-889g-fcgh CVE-2026-71309 (Nick Craig-Wood)+ - serve sftp: Don't crash the whole server on a bad request GHSA-6jcg-q3wp-x2f4 CVE-2026-79775 (Nick Craig-Wood)+ - sftp: Fix command injection via crafted filenames on PowerShell remotes GHSA-2m8m-jhrm-w6j2 CVE-2026-71312 (Nick Craig-Wood)+ - vfs: Don't crash the process if a backend panics on a background goroutine GHSA-6jcg-q3wp-x2f4 CVE-2026-79775 (Nick Craig-Wood)- webdav- - Fix HTTPS to HTTP redirects leaking credentials GHSA-h4mf-4v27-hggj (Nick Craig-Wood)- - Tus: fix potential nil pointer crash GHSA-3x6r-wxxg-53vv (Nick Craig-Wood)+ - Fix HTTPS to HTTP redirects leaking credentials GHSA-h4mf-4v27-hggj CVE-2026-79779 (Nick Craig-Wood)+ - Tus: fix potential nil pointer crash GHSA-3x6r-wxxg-53vv CVE-2026-79778 (Nick Craig-Wood)- Update `google.golang.org/grpc` to fix multiple security problems (Nick Craig-Wood)- New Features- build: Update all dependencies (Nick Craig-Wood)@@ -71738,14 +75590,14 @@ Options:- serve restic: Fix `--private-repos` isolation bypass CVE-2026-59733 (Nick Craig-Wood)- serve s3- Fix spurious 404 on HEAD/GET during VFS writeback (max)- - Fix path traversal letting clients see files in the root GHSA-8v25-v8p6-qf7v (Nick Craig-Wood)+ - Fix path traversal letting clients see files in the root GHSA-8v25-v8p6-qf7v CVE-2026-79781 (Nick Craig-Wood)- serve webdav: Fix MOVE overwrite failing without Overwrite header (Sanjay Santhanam)- serve/http: Fix `--disable-zip` so it works over rc (Nick Craig-Wood)- VFS- Fix hang reopening a file during the handle-caching grace period (Nick Craig-Wood)- Local- Stop `--links` symlinks escaping the destination directory CVE-2026-54572 (Nick Craig-Wood)- - Don't restore setuid/setgid/sticky bits from metadata by default GHSA-945v-v9p3-v5xw (Nick Craig-Wood)+ - Don't restore setuid/setgid/sticky bits from metadata by default GHSA-945v-v9p3-v5xw CVE-2026-79783 (Nick Craig-Wood)- Drive- Warn when non-exportable Google documents are skipped (Nick Craig-Wood)- Fix stray %!(EXTRA) in unexportable google document log message (Nick Craig-Wood)@@ -78511,7 +82363,7 @@ file.The FTP and SFTP backends have their own `http_proxy` settings tosupport an HTTP CONNECT proxy ([--ftp-http-proxy](https://rclone.org/ftp/#ftp-http-proxy) and-[--sftp-http-proxy](https://rclone.org/ftp/#sftp-http-proxy) )+[--sftp-http-proxy](https://rclone.org/sftp/#sftp-http-proxy) )### Rclone gives x509 SSL root certificates error@@ -79925,6 +83777,140 @@ put them back in again. -->- Socialpranker <[email protected]>- phatlc <[email protected]>- Punya Jain <[email protected]>+- acoeur <[email protected]>+- SillyZir <[email protected]>+- am-at-enrollvb <[email protected]>+- Loi Nguyen <[email protected]>+- debaditya <[email protected]>+- alliasgher <[email protected]>+- Anton Karpov <[email protected]>+- Edward Klesel <[email protected]>+- Christian De Santis <[email protected]>+- Rodrigo Rodrigues <[email protected]>+- Recoordinate <[email protected]>+- Jiacheng Xu <[email protected]>+- Teddy Tennant <[email protected]>+- Dean Chen <[email protected]>+- Pastalikek65 <[email protected]>+- Morax <[email protected]>+- Shurong Cao <[email protected]>+- Dave <[email protected]>+- max <[email protected]>+- Hakan İSMAİL <[email protected]>+- Shantanav Mukherjee <[email protected]>+- VXNCXNX <[email protected]>+- Rohit Behera <[email protected]>+- Rahman Yilmaz <[email protected]>+- Sanjay Kanth A <[email protected]>+- machsix <[email protected]>+- CAOShurong <[email protected]>+- shaurya <[email protected]>+- water <[email protected]>+- Rayan Salhab <[email protected]>+- Vijay Misal <[email protected]>+- Sune Mølgaard <[email protected]>+- Anatoly Tarnavsky <[email protected]>+- cyphercodes <[email protected]>+- shaurya <[email protected]>+- no-hup <[email protected]>+- Dhevenddra <[email protected]>+- Splainte <[email protected]>+- PSR94 <[email protected]>+- Shane McCarron <[email protected]>+- Dhevenddra K G <[email protected]>+- Vaibhav Mashal <[email protected]>+- Murat Topcu <[email protected]>+- tomaszni <[email protected]>+- eliotee <[email protected]>+- Aditya <[email protected]>+- Nicholas Velten <[email protected]>+- youdie006 <[email protected]>+- subomi <[email protected]>+- Can Arslan <[email protected]>+- Can Arslan <[email protected]>+- ZRHann <[email protected]>+- enkvadrat <[email protected]>+- Vladimir Babin <[email protected]>+- Eugene <[email protected]>+- Kunpeng Xie <[email protected]>+- foecmke <[email protected]>+- Vasek Sraier <[email protected]>+- Kalin Stoyanov <[email protected]>+- NytePlus <[email protected]>+- Jeremy Schoemaker <[email protected]>+- solunolab <[email protected]>+- Mattias Michaux <[email protected]>+- jxj <[email protected]>+- Harsh Raj Singhania <[email protected]>+- Roland <[email protected]>+- interested.tortoise <[email protected]>+- hsdfat <[email protected]>+- Dirk Petersen <[email protected]>+- TastyHeadphones <[email protected]>+- Neil Cawse <[email protected]>+- Lev Devaev <[email protected]>+- GhostCoder6969 <[email protected]>+- bounty-agent <[email protected]>+- mygrexit <[email protected]>+- Kaya <[email protected]>+- somaz <[email protected]>+- xiaozou-wine <[email protected]>+- Roman Butor <[email protected]>+- kaii9 <[email protected]>+- Stochastic <[email protected]>+- James Truitt <[email protected]>+- firatkaya <[email protected]>+- kevin9327 <[email protected]>+- drakeo338 <[email protected]>+- Alessio Margelli <[email protected]>+- Ludovic Bondon <[email protected]>+- fmterrors <[email protected]>+- GingerAdonis <[email protected]>+- Satvik Mishra <[email protected]>+- wjc <[email protected]>+- Maksym Tymoshyk <[email protected]>+- Aniket <[email protected]>+- Oshino Penguin <[email protected]>+- Yousef Issa <[email protected]>+- ghwany <[email protected]>+- DEVN <[email protected]>+- 47th <[email protected]>+- Sash <[email protected]>+- Shreyansh Jain <[email protected]>+- Biagio Peccerillo <[email protected]>+- gomitrah <[email protected]>+- Jim Counter <[email protected]>+- Dirk Einecke <[email protected]>+- ASRodrigo1 <[email protected]>+- chaogeshuodiannao <[email protected]>+- David Legrand <[email protected]>+- Azhar <[email protected]>+- Daniel Bae <[email protected]>+- frodleTian <[email protected]>+- Cestercian <[email protected]>+- Ayush Sharma <[email protected]>+- Dirk Petersen <[email protected]>+- Codex <[email protected]>+- bunnysayzz <[email protected]>+- hahahahahayesyeseys <[email protected]>+- qodelicious <[email protected]>+- fixfon <[email protected]>+- berce <[email protected]>+- Fred <[email protected]>+- Florian Josef Altendorfer <[email protected]>+- ashwinibailangadi <[email protected]>+- ilya lavrenov <[email protected]>+- Soujas K <[email protected]>+- Robert Kubiś <[email protected]>+- b84330808 <[email protected]>+- Guillaume MALLERET <[email protected]>+- Louis <[email protected]>+- 唐僧洗发用飘柔 <[email protected]>+- Seth Westphal <[email protected]>+- ZainnQureshii <[email protected]>+- Rishi Raj Jain <[email protected]>+- Enbiya Göral <[email protected]>+- Florian Josef Altendorfer <[email protected]># Contact the rclone projectdiff --git a/MANUAL.txt b/MANUAL.txtindex 784a34a88..d4298b105 100644--- a/MANUAL.txt+++ b/MANUAL.txt@@ -1,6 +1,6 @@rclone(1) User ManualNick Craig-Wood-Jul 31, 2026+Nov 06, 2026NAME@@ -38,6 +38,7 @@ SYNOPSISgui Open the web based GUI.hashsum Produces a hashsum file for all the objects in the path.help Show help for rclone commands, flags and backends.+ index Write static directory listings into a remote.link Generate public link to file/folder.listremotes List all the remotes in the config file and defined in environment variables.ls List the objects in the path with size and path.@@ -167,6 +168,7 @@ S3, that work out of the box.)- Akamai Netstorage- Alibaba Cloud (Aliyun) Object Storage System (OSS)- Amazon S3+- Autonomys Auto Drive- Bizfly Cloud Simple Storage- Backblaze B2- Box@@ -174,11 +176,13 @@ S3, that work out of the box.)- China Mobile Ecloud Elastic Object Storage (EOS)- Arvan Cloud Object Storage (AOS)- Citrix ShareFile+- Clever Cloud Cellar- Cloudflare R2- Cloudinary- Cubbit DS3- DigitalOcean Spaces- Digi Storage+- dosya.dev- Dreamhost- Drime- Dropbox@@ -212,8 +216,11 @@ S3, that work out of the box.)- IBM COS S3- IDrive e2- Impossible Cloud+- Infomaniak Swiss Backup (S3)+- Infomaniak Swiss Backup (Swift)- Intercolo Object Storage- IONOS Cloud+- kDrive- Koofr- Leviia Object Storage- Liara Object Storage@@ -229,6 +236,7 @@ S3, that work out of the box.)- Microsoft Azure Files Storage- Microsoft OneDrive- Minio+- Neon Object Storage- Nextcloud- Blomp Cloud Storage- OpenDrive@@ -253,6 +261,7 @@ S3, that work out of the box.)- RackCorp Object Storage- Rackspace Cloud Files- Rclone Serve S3+- RelAix.Storage- rsync.net- Scaleway- Scality (RING / ARTESCA)@@ -518,10 +527,6 @@ Docker installationThe rclone developers maintain a docker image for rclone.-Note: We also now offer a paid version of rclone with enterprise-grade-security and zero CVEs through our partner SecureBuild. If you are-interested, check out their website and the Rclone SecureBuild Image.-These images are built as part of the release process based on a minimalAlpine Linux.@@ -632,7 +637,7 @@ developers so it may be out of date. Its current version is as below.Source installation-Make sure you have git and Go installed. Go version 1.25 or newer is+Make sure you have git and Go installed. Go version 1.26 or newer isrequired, the latest release is recommended. You can get it from yourpackage manager, or download it from golang.org/dl. Then you can run thefollowing:@@ -958,6 +963,7 @@ See the following for detailed instructions for- Crypt - to encrypt other remotes- DigitalOcean Spaces- Digi Storage+- dosya.dev- Drime- Dropbox- Enterprise File Fabric@@ -979,6 +985,7 @@ See the following for detailed instructions for- Internet Archive- Internxt- Jottacloud+- kDrive- Koofr- Linkbox- Mail.ru Cloud@@ -2097,6 +2104,10 @@ recursion.The other list commands lsd,lsf,lsjson do not recurse by default - use-R to make them recurse.+On the bucket-based remotes (e.g. s3, gcs, b2) a directory is a key+prefix rather than an object, so a listing without -R shows each prefix+as one directory entry and not the objects stored under it.+List commands prefer a recursive method that uses more memory but fewertransactions by default. Use --disable ListR to suppress the behavior.See --fast-list for more details.@@ -2164,7 +2175,7 @@ Lists the directories in the source path to standard output. Does notrecurse by default. Use the -R flag to recurse.This command lists the total size of the directory (if known, -1 if-not), the modification time (if known, the current time if not), the+not), the modification time (if known, the --default-time if not), thenumber of objects in the directory (if known, -1 if not) and the name ofthe directory, E.g.@@ -2200,6 +2211,10 @@ recursion.The other list commands lsd,lsf,lsjson do not recurse by default - use-R to make them recurse.+On the bucket-based remotes (e.g. s3, gcs, b2) a directory is a key+prefix rather than an object, so a listing without -R shows each prefix+as one directory entry and not the objects stored under it.+List commands prefer a recursive method that uses more memory but fewertransactions by default. Use --disable ListR to suppress the behavior.See --fast-list for more details.@@ -2295,6 +2310,10 @@ recursion.The other list commands lsd,lsf,lsjson do not recurse by default - use-R to make them recurse.+On the bucket-based remotes (e.g. s3, gcs, b2) a directory is a key+prefix rather than an object, so a listing without -R shows each prefix+as one directory entry and not the objects stored under it.+List commands prefer a recursive method that uses more memory but fewertransactions by default. Use --disable ListR to suppress the behavior.See --fast-list for more details.@@ -3245,6 +3264,7 @@ Options--force Bypass --max-delete safety check and run the sync. Consider using with --verbose-h, --help help for bisync--ignore-listing-checksum Do not use checksums for listings (add --ignore-checksum to additionally skip post-copy checksum checks)+ --max-delete-renames-aware Exclude tracked renames from the --max-delete safety check (requires --track-renames)--max-lock Duration Consider lock files older than this to be expired (default: 0 (never expire)) (minimum: 2m) (default 0s)--no-cleanup Retain working files (useful for troubleshooting and testing).--no-slow-hash Ignore listing checksums only on backends where they are slow@@ -3562,7 +3582,8 @@ Output completion script for a given shell.SynopsisGenerates a shell completion script for rclone. Run with --help to list-the supported shells.+the supported shells. Enum-valued flags such as --log-level and+--vfs-cache-mode also complete their supported values.Options@@ -4683,10 +4704,10 @@ Examples:// Output: stories/The Quick Brown Fox!.txtrclone convmv "stories/The Quick Brown Fox!" --name-transform "date=-{YYYYMMDD}"- // Output: stories/The Quick Brown Fox!-20260731+ // Output: stories/The Quick Brown Fox!-20261106rclone convmv "stories/The Quick Brown Fox!" --name-transform "date=-{macfriendlytime}"- // Output: stories/The Quick Brown Fox!-2026-07-31 0340PM+ // Output: stories/The Quick Brown Fox!-2026-11-06 0220PMrclone convmv "stories/The Quick Brown Fox!.txt" --name-transform "all,regex=[\\.\\w]/ab"// Output: ababababababab/ababab ababababab ababababab ababab!abababab@@ -5321,7 +5342,10 @@ SynopsisRemove a single file from remote. Unlike delete it cannot be used toremove a directory and it doesn't obey include/exclude filters - if the-specified file exists, it will always be removed.+specified file exists, it will always be removed from its original+location.++The --backup-dir and --suffix flags are honored when set.rclone deletefile remote:path [flags]@@ -5332,6 +5356,24 @@ OptionsOptions shared with other commands are described next. See the globalflags page for global options not listed here.+Sync Options++Flags used for sync commands++ --backup-dir string Make backups into hierarchy based in DIR+ --delete-after When synchronizing, delete files on destination after transferring (default)+ --delete-before When synchronizing, delete files on destination before transferring+ --delete-during When synchronizing, delete files during transfer+ --fix-case Force rename of case insensitive dest to match source+ --ignore-errors Delete even if there are I/O errors+ --list-cutoff int To save memory, sort directory listings on disk above this threshold (default 1000000)+ --max-delete int When synchronizing, limit the number of deletes (default -1)+ --max-delete-size SizeSuffix When synchronizing, limit the total size of deletes (default off)+ --suffix string Suffix to add to changed files+ --suffix-keep-extension Preserve the extension when using --suffix+ --track-renames When synchronizing, track file renames and do a server-side move if possible+ --track-renames-strategy string Strategies to use when synchronizing using track-renames hash|modtime|leaf (default "hash")+Important OptionsImportant flags useful for most commands@@ -5535,11 +5577,6 @@ Flags to control the Remote Control API--rc-template string User-specified template--rc-user string User name for authentication--rc-user-from-header string User name from a defined HTTP header- --rc-web-fetch-url string URL to fetch the releases for webgui (default "https://api.github.com/repos/rclone/rclone-webui-react/releases/latest")- --rc-web-gui Launch WebGUI on localhost- --rc-web-gui-force-update Force update to latest version of web gui- --rc-web-gui-no-open-browser Don't open the browser automatically- --rc-web-gui-update Check and update to latest version of web guiSee Also@@ -5581,6 +5618,7 @@ Run without a hash to see the list of all supported hashes, e.g.- blake3- xxh3- xxh128+ - crc32cThen@@ -5642,256 +5680,575 @@ See Also- rclone - Show help for rclone commands, flags and backends.-rclone link--Generate public link to file/folder.--Synopsis--Create, retrieve or remove a public link to the given file or folder.-- rclone link remote:path/to/file- rclone link remote:path/to/folder/- rclone link --unlink remote:path/to/folder/- rclone link --expire 1d remote:path/to/file--If you supply the --expire flag, it will set the expiration time-otherwise it will use the default (100 years). Note not all backends-support the --expire flag - if the backend doesn't support it then the-link returned won't expire.--Use the --unlink flag to remove existing public links to the file or-folder. Note not all backends support "--unlink" flag - those that don't-will just ignore it.--If successful, the last line of the output will contain the link. Exact-capabilities depend on the remote, but the link will always by default-be created with the least constraints - e.g. no expiry, no password-protection, accessible without account.-- rclone link remote:path [flags]--Options-- --expire Duration The amount of time that the link will be valid (default off)- -h, --help help for link- --unlink Remove existing public link to file/folder--See the global flags page for global options not listed here.--See Also--- rclone - Show help for rclone commands, flags and backends.--rclone listremotes+rclone index-List all the remotes in the config file and defined in environment-variables.+Write static directory listings into a remote.Synopsis-Lists all the available remotes from the config file, or the remotes-matching an optional filter.--Prints the result in human-readable format by default, and as a simple-list of remote names, or if used with flag --long a tabular format-including the remote names, types and descriptions. Using flag --json-produces machine-readable output instead, which always includes all-attributes - including the source (file or environment).--Result can be filtered by a filter argument which applies to all-attributes, and/or filter flags specific for each attribute. The values-must be specified according to regular rclone filtering pattern syntax.+Write a directory listing, index.html by default, into every directory+of remote:path so that it can be browsed when it is served as a static+website, for example from an S3 website endpoint, a bucket behind a CDN,+or a plain web server.-By default filtering uses non-anchored matching, so --type box also-matches dropbox. Use --exact to match complete values only.-- rclone listremotes [<filter>] [flags]--Options-- --description string Filter remotes by description- --exact Match filter strings exactly instead of using non-anchored glob matching- -h, --help help for listremotes- --json Format output as JSON- --long Show type and description in addition to name- --name string Filter remotes by name- --order-by string Instructions on how to order the result, e.g. 'type,name=descending'- --source string Filter remotes by source, e.g. 'file' or 'environment'- --type string Filter remotes by type--See the global flags page for global options not listed here.--See Also--- rclone - Show help for rclone commands, flags and backends.--rclone lsf--List directories and objects in remote:path formatted for parsing.+This works like rclone sync: each run makes the minimum changes needed+to bring the listings into line with the files. Listings which haven't+changed aren't rewritten, so a run with no changes makes no uploads. The+listings never show the listing files themselves.-Synopsis+Note that rclone index treats every file with an output name (index.html+by default) in the directories it indexes as its own, and will overwrite+it or delete it without checking where it came from. To protect a+hand-written index, exclude its directory with --index-exclude.-List the contents of the source path (directories and objects) to-standard output in a form which is easy to parse by scripts. By default-this will just be the names of the objects and directories, one per-line. The directories will have a / suffix.+Outputs-E.g.+--output NAME=FORMAT says what to write in each directory and can be+repeated. NAME is the file name, and FORMAT is one of the built-in+formats below or the path of a Go template file.- $ rclone lsf swift:bucket- bevajer5jef- canole- diwogej7- ferejej3gux/- fubuwic+ -----------------------------------------------------------------------+ Format MIME type Contents+ ------------------- -------------------------- ------------------------+ html text/html; charset=utf-8 The HTML listing from+ the rclone serve http+ template++ json application/json JSON in the format of+ rclone lsjson for the+ directory++ caddy application/json JSON in the format of+ Caddy's+ file_server browse+ ------------------------------------------------------------------------Use the --format option to control what gets listed. By default this is-just the path, but you can use these parameters to control the output:+The default is --output index.html=html. The built-in formats are+templates, so they can be printed with --print-template FORMAT, copied+and changed. Outputs whose name ends in .html or .htm are rendered with+Go's html/template and everything else with text/template, which has a+json function for safe encoding. The template data is that of+rclone serve http with .Static set (see rclone serve http). Outputs are+uploaded with the MIME type of the built-in format, or that of the+output name's extension for template outputs, since static hosts serve+the stored type.- p - path- s - size- t - modification time- h - hash- i - ID of object- o - Original ID of underlying object- m - MimeType of object if known- e - encrypted name- T - tier of storage if known, e.g. "Hot" or "Cool"- M - Metadata of object in JSON blob format, eg {"key":"value"}+Filters-Some backends use composite IDs. In particular, Google Drive shortcuts-are normally dereferenced, and their "i" field contains the target-object's ID followed by the shortcut's ID, separated by a tab. This tab-is part of the ID field and is not changed by --separator. Use---drive-skip-shortcuts to omit shortcut entries.+The normal filters (--include, --exclude, --filter-from, --max-age, ...)+control what appears in the listings. An excluded file isn't listed, and+an excluded directory isn't listed or indexed.-So if you wanted the path, size and modification time, you would use---format "pst", or maybe --format "tsp" to put the path last.+The --index-include, --index-exclude, --index-filter flags and their+-from variants control which directories get listings. They take the+same patterns as the normal filters, applied to directory paths, and+--index-include implies excluding all other directories. A directory+excluded here still appears in its parent's listing, but nothing is+written or deleted inside it, so it can have its own hand-made index.+--index-max-depth limits how deep listings are written.-E.g.+Modification times- $ rclone lsf --format "tsp" swift:bucket- 2016-06-25 18:55:41;60295;bevajer5jef- 2016-06-25 18:55:43;90613;canole- 2016-06-25 18:55:43;94467;diwogej7- 2018-04-26 08:50:45;0;ferejej3gux/- 2016-06-25 18:55:40;37600;fubuwic+Listings show the modification time of each file. When using the s3,+oracleobjectstorage and swift backends rclone stores the modification+time as metadata which isn't returned by the bucket listing, so every+object needs a HEAD request, which can mean thousands of transactions+per run. Avoid this with one of:-If you specify "h" in the format you will get the MD5 hash by default,-use the --hash flag to change which hash you want. Note that this can be-returned as an empty string if it isn't available on the object (and for-directories), "ERROR" if there was an error reading it from the object-and "UNSUPPORTED" if that object does not support that hash type.+- --use-server-modtime uses the time the object was uploaded instead,+ which is what the web server sends as Last-Modified. Recommended for+ s3 unless the original modification times matter.+- --no-modtime shows no times at all, which is the cheapest option.-For example, to emulate the md5sum command you can use+--dir-time sets the time shown for directories:- rclone lsf -R --hash MD5 --format hp --separator " " --files-only .+- newest (default): the newest modification time of anything below the+ directory. Note that this means one new file changes the listing of+ every directory above it.+- dir: the directory's own modification time. On most file systems+ writing a listing into a directory changes its time, so this can+ take more than one run to settle.+- none: no time.-E.g.+The listing files are given the newest modification time below their+directory, so the Last-Modified header of a listing is meaningful.- $ rclone lsf -R --hash MD5 --format hp --separator " " --files-only swift:bucket- 7908e352297f0f530b84a756f188baa3 bevajer5jef- cd65ac234e6fea5925974a51cdd865cc canole- 03b5341b4f234b9d984d03ad076bae91 diwogej7- 8fd37c3810dd660778137ac3a66cc06d fubuwic- 99713e14a4c4ff553acaf1930fad985b gixacuh7ku--(Though "rclone md5sum ." is an easier way of typing this.)+Links-By default the separator is ";" this can be changed with the --separator-flag. Note that separators aren't escaped in the path so putting it last-is a good strategy.+Directory links are dir/ by default, which works on hosts that serve+index documents: S3 website endpoints, GCS and Azure static websites,+nginx and Apache, and R2 custom domains with a URL rewrite rule. For+hosts that don't (plain S3 REST URLs, B2 friendly URLs) use --link-index+to make directory links dir/index.html instead.++Partial runs++After a sync which changed a known set of files there is no need to walk+the whole remote. Tell rclone index what changed and it re-indexes only+the directories that could be affected:++ rclone index r2:bucket --changed v1.76.0/ --changed version.txt+ rclone index r2:bucket --changed-from changes.txt+ rclone sync ./public r2:bucket --exclude index.html --combined - | rclone index r2:bucket --changed-combined -++- --changed PATH names a changed file or directory relative to+ remote:path and can be repeated. A directory, given with a trailing+ slash or found to be one, is re-indexed completely.+- --changed-from FILE reads one path per line exactly as written.+- --changed-combined FILE reads the --combined report of rclone sync:+ lines starting with +, -, * and ! are changes and = lines are+ ignored.+- Both files accept - for standard input and all three flags can be+ combined.++A changed file re-indexes its directory and every directory above it,+since their times and their listings can change too, and on bucket based+storage a directory can cease to exist. Each of those is one directory+listing. With --dir-time newest the directories above also need the+times of their unchanged subdirectories, which are read from the+modification times of those subdirectories' listings at the cost of one+small request each. With --use-server-modtime that is the time the+listing was written rather than the time of the newest file, so partial+runs can drift by the indexing delay until the next full run.+--dir-time dir and none need no such lookups.++Directories which are neither named nor above a named path are left+alone, as are directories deeper than --max-depth. An empty change list+does nothing, and naming the root does a full run. There is no automatic+cutover to a full run since that depends on the size of the whole+remote: a partial run costs one listing per affected directory, a full+run one request per thousand objects on S3-like storage or one per+directory elsewhere. The number of directories a partial run will list+is logged with -v, and --changed-max-dirs makes it fall back to a full+run above that number. A good pattern is a partial run after each upload+and a scheduled full run to catch anything missed.++Site icon++The listings carry no icon or branding of their own. To give a site an+icon, upload a favicon.ico to the root of the remote, which browsers+fetch from there for every page. Exclude it from the listings with+--exclude /favicon.ico so it stays on the site but isn't shown.++Deleting listings++On backends which can't have empty directories (S3, B2 and other+bucket-based storage) a directory only exists while it has files in it.+Once the last file in a directory has been removed its listing is+deleted, since the directory only existed because of it. Backends which+can have empty directories keep a listing for each empty directory.++Using with sync++A sync from a local build deletes files on the destination which aren't+in the source, which includes the listings. Either index the local build+before syncing:++ rclone index ./public+ rclone sync ./public r2:bucket++or exclude the listings from the sync so it neither uploads nor deletes+them, then index the remote:++ rclone sync ./public r2:bucket --exclude index.html+ rclone index r2:bucket-E.g.+Examples- $ rclone lsf --separator "," --format "tshp" swift:bucket- 2016-06-25 18:55:41,60295,7908e352297f0f530b84a756f188baa3,bevajer5jef- 2016-06-25 18:55:43,90613,cd65ac234e6fea5925974a51cdd865cc,canole- 2016-06-25 18:55:43,94467,03b5341b4f234b9d984d03ad076bae91,diwogej7- 2018-04-26 08:52:53,0,,ferejej3gux/- 2016-06-25 18:55:40,37600,8fd37c3810dd660778137ac3a66cc06d,fubuwic+An S3 website bucket, using the upload times so no object needs a HEAD+request:-You can output in CSV standard format. This will escape things in " if-they contain,+ rclone index s3:my-site --use-server-modtime-E.g.+HTML and rclone JSON listings, setting Cache-Control on them:- $ rclone lsf --csv --files-only --format ps remote:path- test.log,22355- test.sh,449- "this file contains a comma, in the file name.txt",6+ rclone index s3:bucket --output index.html=html --output index.json=json --header-upload "Cache-Control: max-age=300"-Note that the --absolute parameter is useful for making lists of files-to pass to an rclone copy with the --files-from-raw flag.+Only list release files, and don't write listings under /private:-For example, to find all the files modified within one day and copy-those only (without traversing the whole directory structure):+ rclone index r2:bucket --include "*.{zip,deb,rpm}" --index-exclude "/private/**"- rclone lsf --absolute --files-only --max-age 1d /path/to/local > new_files- rclone copy --files-from-raw new_files /path/to/local remote:path+A B2 bucket served from its friendly URLs:-The default time format is '2006-01-02 15:04:05'. Other formats can be-specified with the --time-format flag. Examples:+ rclone index b2:bucket --link-index- rclone lsf remote:path --format pt --time-format 'Jan 2, 2006 at 3:04pm (MST)'- rclone lsf remote:path --format pt --time-format '2006-01-02 15:04:05.000000000'- rclone lsf remote:path --format pt --time-format '2006-01-02T15:04:05.999999999Z07:00'- rclone lsf remote:path --format pt --time-format RFC3339- rclone lsf remote:path --format pt --time-format DateOnly- rclone lsf remote:path --format pt --time-format max- rclone lsf remote:path --format pt --time-format unix- rclone lsf remote:path --format pt --time-format unixnano+See what would change:---time-format max will automatically truncate-2006-01-02 15:04:05.000000000 to the maximum precision supported by the-remote.+ rclone index r2:bucket --dry-run -v-Any of the filtering options can be applied to this command.+Listings which haven't changed aren't normally rewritten, so changing+the headers or metadata set on them needs --index-rewrite, which writes+every listing a run would render whether it changed or not:-There are several related list commands+ rclone index r2:bucket --index-rewrite --header-upload "Cache-Control: max-age=3600"-- ls to list size and path of objects only-- lsl to list modification time, size and path of objects only-- lsd to list directories only-- lsf to list objects and directories in easy to parse format-- lsjson to list objects and directories in JSON format+ rclone index remote:path [flags]-ls,lsl,lsd are designed to be human-readable. lsf is designed to be-human and machine-readable. lsjson is designed to be machine-readable.+Options-Note that ls and lsl recurse by default - use --max-depth 1 to stop the-recursion.+ --changed stringArray Only re-index directories affected by this changed path (directories end in /)+ --changed-combined stringArray Read changed paths from a sync --combined report (use - to read from stdin)+ --changed-from stringArray Read changed paths from file, one per line (use - to read from stdin)+ --changed-max-dirs int Do a full index if a partial one would list more than this many directories (0 for no limit)+ --dir-time newest|dir|none How to work out the time shown for a directory (default newest)+ -h, --help help for index+ --index-exclude stringArray Don't write listings in directories matching pattern+ --index-exclude-from stringArray Read directory exclude patterns from file (use - to read from stdin)+ --index-filter stringArray Add a rule for which directories get listings+ --index-filter-from stringArray Read directory rules from a file (use - to read from stdin)+ --index-include stringArray Only write listings in directories matching pattern+ --index-include-from stringArray Read directory include patterns from file (use - to read from stdin)+ --index-max-depth int Only write listings this many directories deep (-1 for no limit) (default -1)+ --index-rewrite Write every listing even if it is unchanged+ --link-index Link to dir/NAME rather than dir/ for hosts without index documents+ --no-modtime Don't show modification times in listings+ --output stringArray Listing to write in each directory as NAME=FORMAT (default [index.html=html])+ --print-template string Print the built-in template for FORMAT and exit-The other list commands lsd,lsf,lsjson do not recurse by default - use--R to make them recurse.+Options shared with other commands are described next. See the global+flags page for global options not listed here.-List commands prefer a recursive method that uses more memory but fewer-transactions by default. Use --disable ListR to suppress the behavior.-See --fast-list for more details.+Copy Options-Listing a nonexistent directory will produce an error except for remotes-which can't have empty directories (e.g. s3, swift, or gcs - the-bucket-based remotes).+Flags for anything which can copy a file- rclone lsf remote:path [flags]+ --check-first Do all the checks before starting transfers+ -c, --checksum Check for changes with size & checksum (if available, or fallback to size only)+ --compare-dest stringArray Include additional server-side paths during comparison+ --copy-dest stringArray Implies --compare-dest but also copies files from paths into destination+ --cutoff-mode HARD|SOFT|CAUTIOUS Mode to stop transfers when reaching the max transfer limit HARD|SOFT|CAUTIOUS (default HARD)+ --ignore-case-sync Ignore case when synchronizing+ --ignore-checksum Skip post copy check of checksums+ --ignore-existing Skip all files that exist on destination+ --ignore-size Ignore size when skipping use modtime or checksum+ -I, --ignore-times Don't skip items that match size and time - transfer all unconditionally+ --immutable Do not modify files, fail if existing files have been modified+ --inplace Download directly to destination file instead of atomic download to temp/rename+ -l, --links Translate symlinks to/from regular files with a '.rclonelink' extension+ --max-backlog int Maximum number of objects in sync or check backlog (default 10000)+ --max-duration Duration Maximum duration rclone will transfer data for (default 0s)+ --max-transfer SizeSuffix Maximum size of data to transfer (default off)+ -M, --metadata If set, preserve metadata when copying objects+ --modify-window Duration Max time diff to be considered the same (default 1ns)+ --multi-thread-chunk-size SizeSuffix Chunk size for multi-thread downloads / uploads, if not set by filesystem (default 64Mi)+ --multi-thread-cutoff SizeSuffix Use multi-thread downloads for files above this size (default 256Mi)+ --multi-thread-streams int Number of streams to use for multi-thread downloads (default 4)+ --multi-thread-write-buffer-size SizeSuffix In memory buffer size for writing when in multi-thread mode (default 128Ki)+ --name-transform stringArray Transform paths during the copy process+ --no-check-dest Don't check the destination, copy regardless+ --no-traverse Don't traverse destination file system on copy+ --no-update-dir-modtime Don't update directory modification times+ --no-update-modtime Don't update destination modtime if files identical+ --order-by string Instructions on how to order the transfers, e.g. 'size,descending'+ --partial-suffix string Add partial-suffix to temporary file name when --inplace is not used (default ".partial")+ --refresh-times Refresh the modtime of remote files+ --server-side-across-configs Allow server-side operations (e.g. copy) to work across different configs+ --size-only Skip based on size only, not modtime or checksum+ --streaming-upload-cutoff SizeSuffix Cutoff for switching to chunked upload if file size is unknown, upload starts after reaching cutoff or when file ends (default 100Ki)+ -u, --update Skip files that are newer on the destination-Options+Important Options- --absolute Put a leading / in front of path names- --csv Output in CSV format- -d, --dir-slash Append a slash to directory names (default true)- --dirs-only Only list directories- --files-only Only list files- -F, --format string Output format - see help for details (default "p")- --hash h Use this hash when h is used in the format MD5|SHA-1|DropboxHash (default "md5")- -h, --help help for lsf- -R, --recursive Recurse into the listing- -s, --separator string Separator for the items in the format (default ";")- -t, --time-format string Specify a custom time format - see docs for details (default: 2006-01-02 15:04:05)+Important flags useful for most commands-Options shared with other commands are described next. See the global-flags page for global options not listed here.+ -n, --dry-run Do a trial run with no permanent changes+ -i, --interactive Enable interactive mode+ -v, --verbose count Print lots more stuff (repeat for more)++Filter Options++Flags for filtering directory listings++ --delete-excluded Delete files on dest excluded from sync+ --exclude stringArray Exclude files matching pattern+ --exclude-from stringArray Read file exclude patterns from file (use - to read from stdin)+ --exclude-if-present stringArray Exclude directories if filename is present+ --files-from stringArray Read list of source-file names from file (use - to read from stdin)+ --files-from-raw stringArray Read list of source-file names from file without any processing of lines (use - to read from stdin)+ --files-from0 stringArray Read list of source-file names from file using NUL as separator (use - to read from stdin)+ -f, --filter stringArray Add a file filtering rule+ --filter-from stringArray Read file filtering patterns from a file (use - to read from stdin)+ --hash-filter string Partition filenames by hash k/n or randomly @/n+ --ignore-case Ignore case in filters (case insensitive)+ --include stringArray Include files matching pattern+ --include-from stringArray Read file include patterns from file (use - to read from stdin)+ --max-age Duration Only transfer files younger than this in s or suffix ms|s|m|h|d|w|M|y (default off)+ --max-depth int If set limits the recursion depth to this (default -1)+ --max-size SizeSuffix Only transfer files smaller than this in KiB or suffix B|K|M|G|T|P (default off)+ --metadata-exclude stringArray Exclude metadatas matching pattern+ --metadata-exclude-from stringArray Read metadata exclude patterns from file (use - to read from stdin)+ --metadata-filter stringArray Add a metadata filtering rule+ --metadata-filter-from stringArray Read metadata filtering patterns from a file (use - to read from stdin)+ --metadata-include stringArray Include metadatas matching pattern+ --metadata-include-from stringArray Read metadata include patterns from file (use - to read from stdin)+ --min-age Duration Only transfer files older than this in s or suffix ms|s|m|h|d|w|M|y (default off)+ --min-size SizeSuffix Only transfer files bigger than this in KiB or suffix B|K|M|G|T|P (default off)++Listing Options++Flags for listing directories++ --default-time Time Time to show if modtime is unknown for files and directories (default 2000-01-01T00:00:00Z)+ --fast-list Use recursive list if available; uses more memory but fewer transactions++See Also++- rclone - Show help for rclone commands, flags and backends.++rclone link++Generate public link to file/folder.++Synopsis++Create, retrieve or remove a public link to the given file or folder.++ rclone link remote:path/to/file+ rclone link remote:path/to/folder/+ rclone link --unlink remote:path/to/folder/+ rclone link --expire 1d remote:path/to/file++If you supply the --expire flag, it will set the expiration time+otherwise it will use the default (100 years). Note not all backends+support the --expire flag - if the backend doesn't support it then the+link returned won't expire.++Use the --unlink flag to remove existing public links to the file or+folder. Note not all backends support "--unlink" flag - those that don't+will just ignore it.++If successful, the last line of the output will contain the link. Exact+capabilities depend on the remote, but the link will always by default+be created with the least constraints - e.g. no expiry, no password+protection, accessible without account.++ rclone link remote:path [flags]++Options++ --expire Duration The amount of time that the link will be valid (default off)+ -h, --help help for link+ --unlink Remove existing public link to file/folder++See the global flags page for global options not listed here.++See Also++- rclone - Show help for rclone commands, flags and backends.++rclone listremotes++List all the remotes in the config file and defined in environment+variables.++Synopsis++Lists all the available remotes from the config file, or the remotes+matching an optional filter.++Prints the result in human-readable format by default, and as a simple+list of remote names, or if used with flag --long a tabular format+including the remote names, types and descriptions. Using flag --json+produces machine-readable output instead, which always includes all+attributes - including the source (file or environment).++Result can be filtered by a filter argument which applies to all+attributes, and/or filter flags specific for each attribute. The values+must be specified according to regular rclone filtering pattern syntax.++By default filtering uses non-anchored matching, so --type box also+matches dropbox. Use --exact to match complete values only.++ rclone listremotes [<filter>] [flags]++Options++ --description string Filter remotes by description+ --exact Match filter strings exactly instead of using non-anchored glob matching+ -h, --help help for listremotes+ --json Format output as JSON+ --long Show type and description in addition to name+ --name string Filter remotes by name+ --order-by string Instructions on how to order the result, e.g. 'type,name=descending'+ --source string Filter remotes by source, e.g. 'file' or 'environment'+ --type string Filter remotes by type++See the global flags page for global options not listed here.++See Also++- rclone - Show help for rclone commands, flags and backends.++rclone lsf++List directories and objects in remote:path formatted for parsing.++Synopsis++List the contents of the source path (directories and objects) to+standard output in a form which is easy to parse by scripts. By default+this will just be the names of the objects and directories, one per+line. The directories will have a / suffix.++E.g.++ $ rclone lsf swift:bucket+ bevajer5jef+ canole+ diwogej7+ ferejej3gux/+ fubuwic++Use the --format option to control what gets listed. By default this is+just the path, but you can use these parameters to control the output:++ p - path+ s - size+ t - modification time+ h - hash+ i - ID of object+ o - Original ID of underlying object+ m - MimeType of object if known+ e - encrypted name+ T - tier of storage if known, e.g. "Hot" or "Cool"+ M - Metadata of object in JSON blob format, eg {"key":"value"}++Some backends use composite IDs. In particular, Google Drive shortcuts+are normally dereferenced, and their "i" field contains the target+object's ID followed by the shortcut's ID, separated by a tab. This tab+is part of the ID field and is not changed by --separator. Use+--drive-skip-shortcuts to omit shortcut entries.++So if you wanted the path, size and modification time, you would use+--format "pst", or maybe --format "tsp" to put the path last.++E.g.++ $ rclone lsf --format "tsp" swift:bucket+ 2016-06-25 18:55:41;60295;bevajer5jef+ 2016-06-25 18:55:43;90613;canole+ 2016-06-25 18:55:43;94467;diwogej7+ 2018-04-26 08:50:45;0;ferejej3gux/+ 2016-06-25 18:55:40;37600;fubuwic++If you specify "h" in the format you will get the MD5 hash by default,+use the --hash flag to change which hash you want. Note that this can be+returned as an empty string if it isn't available on the object (and for+directories), "ERROR" if there was an error reading it from the object+and "UNSUPPORTED" if that object does not support that hash type.++For example, to emulate the md5sum command you can use++ rclone lsf -R --hash MD5 --format hp --separator " " --files-only .++E.g.++ $ rclone lsf -R --hash MD5 --format hp --separator " " --files-only swift:bucket+ 7908e352297f0f530b84a756f188baa3 bevajer5jef+ cd65ac234e6fea5925974a51cdd865cc canole+ 03b5341b4f234b9d984d03ad076bae91 diwogej7+ 8fd37c3810dd660778137ac3a66cc06d fubuwic+ 99713e14a4c4ff553acaf1930fad985b gixacuh7ku++(Though "rclone md5sum ." is an easier way of typing this.)++By default the separator is ";" this can be changed with the --separator+flag. Note that separators aren't escaped in the path so putting it last+is a good strategy.++E.g.++ $ rclone lsf --separator "," --format "tshp" swift:bucket+ 2016-06-25 18:55:41,60295,7908e352297f0f530b84a756f188baa3,bevajer5jef+ 2016-06-25 18:55:43,90613,cd65ac234e6fea5925974a51cdd865cc,canole+ 2016-06-25 18:55:43,94467,03b5341b4f234b9d984d03ad076bae91,diwogej7+ 2018-04-26 08:52:53,0,,ferejej3gux/+ 2016-06-25 18:55:40,37600,8fd37c3810dd660778137ac3a66cc06d,fubuwic++You can output in CSV standard format. This will escape things in " if+they contain,++E.g.++ $ rclone lsf --csv --files-only --format ps remote:path+ test.log,22355+ test.sh,449+ "this file contains a comma, in the file name.txt",6++Note that the --absolute parameter is useful for making lists of files+to pass to an rclone copy with the --files-from-raw flag.++For example, to find all the files modified within one day and copy+those only (without traversing the whole directory structure):++ rclone lsf --absolute --files-only --max-age 1d /path/to/local > new_files+ rclone copy --files-from-raw new_files /path/to/local remote:path++The default time format is '2006-01-02 15:04:05'. Other formats can be+specified with the --time-format flag. Examples:++ rclone lsf remote:path --format pt --time-format 'Jan 2, 2006 at 3:04pm (MST)'+ rclone lsf remote:path --format pt --time-format '2006-01-02 15:04:05.000000000'+ rclone lsf remote:path --format pt --time-format '2006-01-02T15:04:05.999999999Z07:00'+ rclone lsf remote:path --format pt --time-format RFC3339+ rclone lsf remote:path --format pt --time-format DateOnly+ rclone lsf remote:path --format pt --time-format max+ rclone lsf remote:path --format pt --time-format unix+ rclone lsf remote:path --format pt --time-format unixnano++--time-format max will automatically truncate+2006-01-02 15:04:05.000000000 to the maximum precision supported by the+remote.++Any of the filtering options can be applied to this command.++There are several related list commands++- ls to list size and path of objects only+- lsl to list modification time, size and path of objects only+- lsd to list directories only+- lsf to list objects and directories in easy to parse format+- lsjson to list objects and directories in JSON format++ls,lsl,lsd are designed to be human-readable. lsf is designed to be+human and machine-readable. lsjson is designed to be machine-readable.++Note that ls and lsl recurse by default - use --max-depth 1 to stop the+recursion.++The other list commands lsd,lsf,lsjson do not recurse by default - use+-R to make them recurse.++On the bucket-based remotes (e.g. s3, gcs, b2) a directory is a key+prefix rather than an object, so a listing without -R shows each prefix+as one directory entry and not the objects stored under it.++List commands prefer a recursive method that uses more memory but fewer+transactions by default. Use --disable ListR to suppress the behavior.+See --fast-list for more details.++Listing a nonexistent directory will produce an error except for remotes+which can't have empty directories (e.g. s3, swift, or gcs - the+bucket-based remotes).++ rclone lsf remote:path [flags]++Options++ --absolute Put a leading / in front of path names+ --csv Output in CSV format+ -d, --dir-slash Append a slash to directory names (default true)+ --dirs-only Only list directories+ --files-only Only list files+ -F, --format string Output format - see help for details (default "p")+ --hash h Use this hash when h is used in the format MD5|SHA-1|DropboxHash (default "md5")+ -h, --help help for lsf+ -R, --recursive Recurse into the listing+ -s, --separator string Separator for the items in the format (default ";")+ -t, --time-format string Specify a custom time format - see docs for details (default: 2006-01-02 15:04:05)++Options shared with other commands are described next. See the global+flags page for global options not listed here.Filter Options@@ -6041,6 +6398,10 @@ recursion.The other list commands lsd,lsf,lsjson do not recurse by default - use-R to make them recurse.+On the bucket-based remotes (e.g. s3, gcs, b2) a directory is a key+prefix rather than an object, so a listing without -R shows each prefix+as one directory entry and not the objects stored under it.+List commands prefer a recursive method that uses more memory but fewertransactions by default. Use --disable ListR to suppress the behavior.See --fast-list for more details.@@ -6388,6 +6749,11 @@ macOS. For details, see vfs-case-sensitivity.NFS mount+For macOS (and other platforms where this path is supported), prefer the+dedicated rclone nfsmount command. It starts the NFS server and performs+the mount for you. rclone mount itself still uses FUSE (macFUSE/FUSE-T)+and does not switch to NFS via a flag.+This method spins up an NFS server using serve nfs command and mounts itto the specified mountpoint. If you run this in background mode using|--daemon|, you will need to send SIGTERM signal to the rclone process@@ -6704,8 +7070,8 @@ at all times. The buffered data is bound to one open file and won't beshared.This flag is a upper limit for the used memory per open file. The buffer-will only use memory for data that is downloaded but not not yet read.-If the buffer is empty, only a small amount of memory will be used.+will only use memory for data that is downloaded but not yet read. If+the buffer is empty, only a small amount of memory will be used.The maximum memory used by rclone for buffering can be up to--buffer-size * open files.@@ -6754,11 +7120,12 @@ that haven't been accessed for the longest. This cache flushing strategyis efficient and more relevant files are likely to remain cached.The --vfs-cache-max-age will evict files from the cache after the set-time since last access has passed. The default value of 1 hour will-start evicting files from cache that haven't been accessed for 1 hour.-When a cached file is accessed the 1 hour timer is reset to 0 and will-wait for 1 more hour before evicting. Specify the time with standard-notation, s, m, h, d, w .+time since last access has passed; it is based on access time, not on+when the file was first added to the cache. The default value of 1 hour+will start evicting files from cache that haven't been accessed for 1+hour. When a cached file is accessed the 1 hour timer is reset to 0 and+will wait for 1 more hour before evicting. Specify the time with+standard notation, s, m, h, d, w .You should not run two copies of rclone using the same VFS cache withthe same or overlapping remotes if using --vfs-cache-mode > off. This@@ -6905,8 +7272,8 @@ Rclone reads --vfs-read-chunk-streams chunks of size--vfs-read-chunk-size concurrently. The size for each read will stayconstant.-This improves performance performance massively on high latency links or-very high bandwidth links to high performance object stores.+This improves performance massively on high latency links or very high+bandwidth links to high performance object stores.Some experimentation will be needed to find the optimum values of--vfs-read-chunk-size and --vfs-read-chunk-streams as these will depend@@ -7796,6 +8163,11 @@ macOS. For details, see vfs-case-sensitivity.NFS mount+For macOS (and other platforms where this path is supported), prefer the+dedicated rclone nfsmount command. It starts the NFS server and performs+the mount for you. rclone mount itself still uses FUSE (macFUSE/FUSE-T)+and does not switch to NFS via a flag.+This method spins up an NFS server using serve nfs command and mounts itto the specified mountpoint. If you run this in background mode using|--daemon|, you will need to send SIGTERM signal to the rclone process@@ -8112,8 +8484,8 @@ at all times. The buffered data is bound to one open file and won't beshared.This flag is a upper limit for the used memory per open file. The buffer-will only use memory for data that is downloaded but not not yet read.-If the buffer is empty, only a small amount of memory will be used.+will only use memory for data that is downloaded but not yet read. If+the buffer is empty, only a small amount of memory will be used.The maximum memory used by rclone for buffering can be up to--buffer-size * open files.@@ -8162,11 +8534,12 @@ that haven't been accessed for the longest. This cache flushing strategyis efficient and more relevant files are likely to remain cached.The --vfs-cache-max-age will evict files from the cache after the set-time since last access has passed. The default value of 1 hour will-start evicting files from cache that haven't been accessed for 1 hour.-When a cached file is accessed the 1 hour timer is reset to 0 and will-wait for 1 more hour before evicting. Specify the time with standard-notation, s, m, h, d, w .+time since last access has passed; it is based on access time, not on+when the file was first added to the cache. The default value of 1 hour+will start evicting files from cache that haven't been accessed for 1+hour. When a cached file is accessed the 1 hour timer is reset to 0 and+will wait for 1 more hour before evicting. Specify the time with+standard notation, s, m, h, d, w .You should not run two copies of rclone using the same VFS cache withthe same or overlapping remotes if using --vfs-cache-mode > off. This@@ -8313,8 +8686,8 @@ Rclone reads --vfs-read-chunk-streams chunks of size--vfs-read-chunk-size concurrently. The size for each read will stayconstant.-This improves performance performance massively on high latency links or-very high bandwidth links to high performance object stores.+This improves performance massively on high latency links or very high+bandwidth links to high performance object stores.Some experimentation will be needed to find the optimum values of--vfs-read-chunk-size and --vfs-read-chunk-streams as these will depend@@ -8847,6 +9220,14 @@ will accept in the HTTP header.responses, will overriding existing values. The flag may be repeated toadd multiple headers. Use the format Header-Name: value.+--rc-allow-origin enables cross-origin (CORS) requests from browsers by+setting the Access-Control-Allow-Origin header on every response to the+origin given, e.g. --rc-allow-origin https://app.example.com. Browser+CORS preflight OPTIONS requests are answered by rclone without needing+authentication as browsers never send credentials with them. Setting+this to * allows any origin which is a security risk on an authenticated+server.+--rc-baseurl controls the URL prefix that rclone serves from. By defaultrclone will serve from the root. If you used --rc-baseurl "/rclone" thenrclone would serve from a URL starting with "/rclone/". This is useful@@ -8935,6 +9316,34 @@ be used within the template to server pages:.Query Currently unused.+ .Path The path of the+ directory from the+ root with leading and+ trailing slashes, e.g.+ '/dir/'.++ .IsRoot Boolean for if this is+ the root directory.++ .UpLink The link to the parent+ directory.++ .Static Boolean for if the+ listing is being+ written as a static+ page by rclone index.++ .NumDirs The number of+ directories in the+ listing.++ .NumFiles The number of files in+ the listing.++ .TotalSize The total size in+ bytes of the files in+ the listing.+.Breadcrumb Allows for creating arelative navigation.@@ -8963,6 +9372,11 @@ be used within the template to server pages:.ModTime The UTC timestamp ofan entry.++ .MimeType The MIME type of an+ entry guessed from its+ name, or+ 'inode/directory'.-----------------------------------------------------------------------The server also makes the following functions available so that they can@@ -9014,8 +9428,20 @@ To create an htpasswd file:htpasswd -B htpasswd userhtpasswd -B htpasswd anotherUser+Use the same hash format and cost for every entry in the file. If the+entries differ, the time taken to reject a wrong password can reveal+which usernames exist.+The password file can be updated while rclone is running.+When authentication is enabled every request must be authenticated,+including HTTP OPTIONS requests, as the answer to an OPTIONS request can+reveal whether a path exists. Health checks or monitoring which use+unauthenticated OPTIONS requests will need to send credentials. The+exception is browser CORS preflight requests when --rc-allow-origin is+set, which are answered without authentication as browsers never send+credentials with them.+Use --rc-realm to set the authentication realm.Use --rc-salt to change the password hashing salt from the default.@@ -9059,11 +9485,6 @@ Flags to control the Remote Control API--rc-template string User-specified template--rc-user string User name for authentication--rc-user-from-header string User name from a defined HTTP header- --rc-web-fetch-url string URL to fetch the releases for webgui (default "https://api.github.com/repos/rclone/rclone-webui-react/releases/latest")- --rc-web-gui Launch WebGUI on localhost- --rc-web-gui-force-update Force update to latest version of web gui- --rc-web-gui-no-open-browser Don't open the browser automatically- --rc-web-gui-update Check and update to latest version of web guiSee Also@@ -9314,8 +9735,8 @@ at all times. The buffered data is bound to one open file and won't beshared.This flag is a upper limit for the used memory per open file. The buffer-will only use memory for data that is downloaded but not not yet read.-If the buffer is empty, only a small amount of memory will be used.+will only use memory for data that is downloaded but not yet read. If+the buffer is empty, only a small amount of memory will be used.The maximum memory used by rclone for buffering can be up to--buffer-size * open files.@@ -9364,11 +9785,12 @@ that haven't been accessed for the longest. This cache flushing strategyis efficient and more relevant files are likely to remain cached.The --vfs-cache-max-age will evict files from the cache after the set-time since last access has passed. The default value of 1 hour will-start evicting files from cache that haven't been accessed for 1 hour.-When a cached file is accessed the 1 hour timer is reset to 0 and will-wait for 1 more hour before evicting. Specify the time with standard-notation, s, m, h, d, w .+time since last access has passed; it is based on access time, not on+when the file was first added to the cache. The default value of 1 hour+will start evicting files from cache that haven't been accessed for 1+hour. When a cached file is accessed the 1 hour timer is reset to 0 and+will wait for 1 more hour before evicting. Specify the time with+standard notation, s, m, h, d, w .You should not run two copies of rclone using the same VFS cache withthe same or overlapping remotes if using --vfs-cache-mode > off. This@@ -9515,8 +9937,8 @@ Rclone reads --vfs-read-chunk-streams chunks of size--vfs-read-chunk-size concurrently. The size for each read will stayconstant.-This improves performance performance massively on high latency links or-very high bandwidth links to high performance object stores.+This improves performance massively on high latency links or very high+bandwidth links to high performance object stores.Some experimentation will be needed to find the optimum values of--vfs-read-chunk-size and --vfs-read-chunk-streams as these will depend@@ -9953,8 +10375,8 @@ at all times. The buffered data is bound to one open file and won't beshared.This flag is a upper limit for the used memory per open file. The buffer-will only use memory for data that is downloaded but not not yet read.-If the buffer is empty, only a small amount of memory will be used.+will only use memory for data that is downloaded but not yet read. If+the buffer is empty, only a small amount of memory will be used.The maximum memory used by rclone for buffering can be up to--buffer-size * open files.@@ -10003,11 +10425,12 @@ that haven't been accessed for the longest. This cache flushing strategyis efficient and more relevant files are likely to remain cached.The --vfs-cache-max-age will evict files from the cache after the set-time since last access has passed. The default value of 1 hour will-start evicting files from cache that haven't been accessed for 1 hour.-When a cached file is accessed the 1 hour timer is reset to 0 and will-wait for 1 more hour before evicting. Specify the time with standard-notation, s, m, h, d, w .+time since last access has passed; it is based on access time, not on+when the file was first added to the cache. The default value of 1 hour+will start evicting files from cache that haven't been accessed for 1+hour. When a cached file is accessed the 1 hour timer is reset to 0 and+will wait for 1 more hour before evicting. Specify the time with+standard notation, s, m, h, d, w .You should not run two copies of rclone using the same VFS cache withthe same or overlapping remotes if using --vfs-cache-mode > off. This@@ -10154,8 +10577,8 @@ Rclone reads --vfs-read-chunk-streams chunks of size--vfs-read-chunk-size concurrently. The size for each read will stayconstant.-This improves performance performance massively on high latency links or-very high bandwidth links to high performance object stores.+This improves performance massively on high latency links or very high+bandwidth links to high performance object stores.Some experimentation will be needed to find the optimum values of--vfs-read-chunk-size and --vfs-read-chunk-streams as these will depend@@ -10544,8 +10967,8 @@ at all times. The buffered data is bound to one open file and won't beshared.This flag is a upper limit for the used memory per open file. The buffer-will only use memory for data that is downloaded but not not yet read.-If the buffer is empty, only a small amount of memory will be used.+will only use memory for data that is downloaded but not yet read. If+the buffer is empty, only a small amount of memory will be used.The maximum memory used by rclone for buffering can be up to--buffer-size * open files.@@ -10594,11 +11017,12 @@ that haven't been accessed for the longest. This cache flushing strategyis efficient and more relevant files are likely to remain cached.The --vfs-cache-max-age will evict files from the cache after the set-time since last access has passed. The default value of 1 hour will-start evicting files from cache that haven't been accessed for 1 hour.-When a cached file is accessed the 1 hour timer is reset to 0 and will-wait for 1 more hour before evicting. Specify the time with standard-notation, s, m, h, d, w .+time since last access has passed; it is based on access time, not on+when the file was first added to the cache. The default value of 1 hour+will start evicting files from cache that haven't been accessed for 1+hour. When a cached file is accessed the 1 hour timer is reset to 0 and+will wait for 1 more hour before evicting. Specify the time with+standard notation, s, m, h, d, w .You should not run two copies of rclone using the same VFS cache withthe same or overlapping remotes if using --vfs-cache-mode > off. This@@ -10745,8 +11169,8 @@ Rclone reads --vfs-read-chunk-streams chunks of size--vfs-read-chunk-size concurrently. The size for each read will stayconstant.-This improves performance performance massively on high latency links or-very high bandwidth links to high performance object stores.+This improves performance massively on high latency links or very high+bandwidth links to high performance object stores.Some experimentation will be needed to find the optimum values of--vfs-read-chunk-size and --vfs-read-chunk-streams as these will depend@@ -10973,16 +11397,18 @@ This config generated must have this extra parameter- _root - root to use for the backend-And it may have this parameter+And it may have these parameters- _obscure - comma separated strings for parameters to obscure+- _secret_access_key - the secret for S3 access key auth (see below)If password authentication was used by the client, input to the proxyprocess (on STDIN) would look similar to this:{"user": "me",- "pass": "mypassword"+ "pass": "mypassword",+ "client_ip": "192.168.1.1"}If public-key authentication was used by the client, input to the proxy@@ -10990,9 +11416,41 @@ process (on STDIN) would look similar to this:{"user": "me",- "public_key": "AAAAB3NzaC1yc2EAAAADAQABAAABAQDuwESFdAe14hVS6omeyX7edc...JQdf"+ "public_key": "AAAAB3NzaC1yc2EAAAADAQABAAABAQDuwESFdAe14hVS6omeyX7edc...JQdf",+ "client_ip": "192.168.1.1"}+If the client authenticated with an S3 access key (rclone serve s3), the+client never sends its secret, only a signature made with it, so the+input contains just the access key ID as the user with no pass or+public_key:++ {+ "user": "AKIAIOSFODNN7EXAMPLE",+ "client_ip": "192.168.1.1"+ }++In this case the program must look up the secret access key for that+access key ID and return it in the _secret_access_key field of the+output. Rclone then uses that secret to verify the signature on the+request, refusing the request if it does not match. This means the proxy+program is the source of truth for both the credentials and the backend+they map to. If the program does not return _secret_access_key or+returns it empty the request is refused.++The program's answer for an access key ID is cached (see below) but is+checked with the program again after 5 minutes even if the access key ID+is in constant use, so revoking an access key ID in the program takes+effect within 5 minutes. A rotated secret takes effect on the first+request signed with it.++The client_ip key holds the IP address the client connected from,+without a port number. It can be used to restrict logins to certain+networks, or to log authentication attempts centrally. It is omitted if+the client has no IP address, for example when connecting over a unix+socket. Note that if rclone is behind a reverse proxy this will be the+address of the reverse proxy and not the original client.+And as an example return this on STDOUT{@@ -11016,11 +11474,19 @@ [email protected] and then set the host to example.com in the output andthe user to user. For security you'd probably want to restrict the hostto a limited list.-An internal cache of backends is keyed on the user and a hash of the-pass or public_key. This means that if a user's password or public-key-changes, or the proxy returns different config parameters (eg a rotated-api_key), a fresh backend will be created on the next request rather-than the cached one being reused.+An internal cache of backends is keyed on the user, the kind of+credential presented, a hash of the pass or public_key, and the+client_ip. This means that if a user's password or public-key changes,+the client connects from a new IP address, or the proxy returns+different config parameters (eg a rotated api_key), a fresh backend will+be created on the next request rather than the cached one being reused.+A backend created for a public key login is never reused for a password+login, so a public key cannot be sent as a password to skip the program.++A cached answer is checked with the program again after 5 minutes even+if the user is in constant use, so revoking a user in the program takes+effect within 5 minutes. For serve sftp this applies to new logins - an+existing connection is not closed.This can be used to build general purpose proxies to any kind of backendthat rclone supports.@@ -11153,6 +11619,14 @@ accept in the HTTP header.will overriding existing values. The flag may be repeated to addmultiple headers. Use the format Header-Name: value.+--allow-origin enables cross-origin (CORS) requests from browsers by+setting the Access-Control-Allow-Origin header on every response to the+origin given, e.g. --allow-origin https://app.example.com. Browser CORS+preflight OPTIONS requests are answered by rclone without needing+authentication as browsers never send credentials with them. Setting+this to * allows any origin which is a security risk on an authenticated+server.+--baseurl controls the URL prefix that rclone serves from. By defaultrclone will serve from the root. If you used --baseurl "/rclone" thenrclone would serve from a URL starting with "/rclone/". This is useful@@ -11240,6 +11714,34 @@ be used within the template to server pages:.Query Currently unused.+ .Path The path of the+ directory from the+ root with leading and+ trailing slashes, e.g.+ '/dir/'.++ .IsRoot Boolean for if this is+ the root directory.++ .UpLink The link to the parent+ directory.++ .Static Boolean for if the+ listing is being+ written as a static+ page by rclone index.++ .NumDirs The number of+ directories in the+ listing.++ .NumFiles The number of files in+ the listing.++ .TotalSize The total size in+ bytes of the files in+ the listing.+.Breadcrumb Allows for creating arelative navigation.@@ -11268,6 +11770,11 @@ be used within the template to server pages:.ModTime The UTC timestamp ofan entry.++ .MimeType The MIME type of an+ entry guessed from its+ name, or+ 'inode/directory'.-----------------------------------------------------------------------The server also makes the following functions available so that they can@@ -11319,8 +11826,20 @@ To create an htpasswd file:htpasswd -B htpasswd userhtpasswd -B htpasswd anotherUser+Use the same hash format and cost for every entry in the file. If the+entries differ, the time taken to reject a wrong password can reveal+which usernames exist.+The password file can be updated while rclone is running.+When authentication is enabled every request must be authenticated,+including HTTP OPTIONS requests, as the answer to an OPTIONS request can+reveal whether a path exists. Health checks or monitoring which use+unauthenticated OPTIONS requests will need to send credentials. The+exception is browser CORS preflight requests when --allow-origin is set,+which are answered without authentication as browsers never send+credentials with them.+Use --realm to set the authentication realm.Use --salt to change the password hashing salt from the default.@@ -11380,8 +11899,8 @@ at all times. The buffered data is bound to one open file and won't beshared.This flag is a upper limit for the used memory per open file. The buffer-will only use memory for data that is downloaded but not not yet read.-If the buffer is empty, only a small amount of memory will be used.+will only use memory for data that is downloaded but not yet read. If+the buffer is empty, only a small amount of memory will be used.The maximum memory used by rclone for buffering can be up to--buffer-size * open files.@@ -11430,11 +11949,12 @@ that haven't been accessed for the longest. This cache flushing strategyis efficient and more relevant files are likely to remain cached.The --vfs-cache-max-age will evict files from the cache after the set-time since last access has passed. The default value of 1 hour will-start evicting files from cache that haven't been accessed for 1 hour.-When a cached file is accessed the 1 hour timer is reset to 0 and will-wait for 1 more hour before evicting. Specify the time with standard-notation, s, m, h, d, w .+time since last access has passed; it is based on access time, not on+when the file was first added to the cache. The default value of 1 hour+will start evicting files from cache that haven't been accessed for 1+hour. When a cached file is accessed the 1 hour timer is reset to 0 and+will wait for 1 more hour before evicting. Specify the time with+standard notation, s, m, h, d, w .You should not run two copies of rclone using the same VFS cache withthe same or overlapping remotes if using --vfs-cache-mode > off. This@@ -11581,8 +12101,8 @@ Rclone reads --vfs-read-chunk-streams chunks of size--vfs-read-chunk-size concurrently. The size for each read will stayconstant.-This improves performance performance massively on high latency links or-very high bandwidth links to high performance object stores.+This improves performance massively on high latency links or very high+bandwidth links to high performance object stores.Some experimentation will be needed to find the optimum values of--vfs-read-chunk-size and --vfs-read-chunk-streams as these will depend@@ -11809,16 +12329,18 @@ This config generated must have this extra parameter- _root - root to use for the backend-And it may have this parameter+And it may have these parameters- _obscure - comma separated strings for parameters to obscure+- _secret_access_key - the secret for S3 access key auth (see below)If password authentication was used by the client, input to the proxyprocess (on STDIN) would look similar to this:{"user": "me",- "pass": "mypassword"+ "pass": "mypassword",+ "client_ip": "192.168.1.1"}If public-key authentication was used by the client, input to the proxy@@ -11826,9 +12348,41 @@ process (on STDIN) would look similar to this:{"user": "me",- "public_key": "AAAAB3NzaC1yc2EAAAADAQABAAABAQDuwESFdAe14hVS6omeyX7edc...JQdf"+ "public_key": "AAAAB3NzaC1yc2EAAAADAQABAAABAQDuwESFdAe14hVS6omeyX7edc...JQdf",+ "client_ip": "192.168.1.1"}+If the client authenticated with an S3 access key (rclone serve s3), the+client never sends its secret, only a signature made with it, so the+input contains just the access key ID as the user with no pass or+public_key:++ {+ "user": "AKIAIOSFODNN7EXAMPLE",+ "client_ip": "192.168.1.1"+ }++In this case the program must look up the secret access key for that+access key ID and return it in the _secret_access_key field of the+output. Rclone then uses that secret to verify the signature on the+request, refusing the request if it does not match. This means the proxy+program is the source of truth for both the credentials and the backend+they map to. If the program does not return _secret_access_key or+returns it empty the request is refused.++The program's answer for an access key ID is cached (see below) but is+checked with the program again after 5 minutes even if the access key ID+is in constant use, so revoking an access key ID in the program takes+effect within 5 minutes. A rotated secret takes effect on the first+request signed with it.++The client_ip key holds the IP address the client connected from,+without a port number. It can be used to restrict logins to certain+networks, or to log authentication attempts centrally. It is omitted if+the client has no IP address, for example when connecting over a unix+socket. Note that if rclone is behind a reverse proxy this will be the+address of the reverse proxy and not the original client.+And as an example return this on STDOUT{@@ -11852,11 +12406,19 @@ [email protected] and then set the host to example.com in the output andthe user to user. For security you'd probably want to restrict the hostto a limited list.-An internal cache of backends is keyed on the user and a hash of the-pass or public_key. This means that if a user's password or public-key-changes, or the proxy returns different config parameters (eg a rotated-api_key), a fresh backend will be created on the next request rather-than the cached one being reused.+An internal cache of backends is keyed on the user, the kind of+credential presented, a hash of the pass or public_key, and the+client_ip. This means that if a user's password or public-key changes,+the client connects from a new IP address, or the proxy returns+different config parameters (eg a rotated api_key), a fresh backend will+be created on the next request rather than the cached one being reused.+A backend created for a public key login is never reused for a password+login, so a public key cannot be sent as a password to skip the program.++A cached answer is checked with the program again after 5 minutes even+if the user is in constant use, so revoking a user in the program takes+effect within 5 minutes. For serve sftp this applies to new logins - an+existing connection is not closed.This can be used to build general purpose proxies to any kind of backendthat rclone supports.@@ -12102,8 +12664,8 @@ at all times. The buffered data is bound to one open file and won't beshared.This flag is a upper limit for the used memory per open file. The buffer-will only use memory for data that is downloaded but not not yet read.-If the buffer is empty, only a small amount of memory will be used.+will only use memory for data that is downloaded but not yet read. If+the buffer is empty, only a small amount of memory will be used.The maximum memory used by rclone for buffering can be up to--buffer-size * open files.@@ -12152,11 +12714,12 @@ that haven't been accessed for the longest. This cache flushing strategyis efficient and more relevant files are likely to remain cached.The --vfs-cache-max-age will evict files from the cache after the set-time since last access has passed. The default value of 1 hour will-start evicting files from cache that haven't been accessed for 1 hour.-When a cached file is accessed the 1 hour timer is reset to 0 and will-wait for 1 more hour before evicting. Specify the time with standard-notation, s, m, h, d, w .+time since last access has passed; it is based on access time, not on+when the file was first added to the cache. The default value of 1 hour+will start evicting files from cache that haven't been accessed for 1+hour. When a cached file is accessed the 1 hour timer is reset to 0 and+will wait for 1 more hour before evicting. Specify the time with+standard notation, s, m, h, d, w .You should not run two copies of rclone using the same VFS cache withthe same or overlapping remotes if using --vfs-cache-mode > off. This@@ -12303,8 +12866,8 @@ Rclone reads --vfs-read-chunk-streams chunks of size--vfs-read-chunk-size concurrently. The size for each read will stayconstant.-This improves performance performance massively on high latency links or-very high bandwidth links to high performance object stores.+This improves performance massively on high latency links or very high+bandwidth links to high performance object stores.Some experimentation will be needed to find the optimum values of--vfs-read-chunk-size and --vfs-read-chunk-streams as these will depend@@ -12701,6 +13264,14 @@ accept in the HTTP header.will overriding existing values. The flag may be repeated to addmultiple headers. Use the format Header-Name: value.+--allow-origin enables cross-origin (CORS) requests from browsers by+setting the Access-Control-Allow-Origin header on every response to the+origin given, e.g. --allow-origin https://app.example.com. Browser CORS+preflight OPTIONS requests are answered by rclone without needing+authentication as browsers never send credentials with them. Setting+this to * allows any origin which is a security risk on an authenticated+server.+--baseurl controls the URL prefix that rclone serves from. By defaultrclone will serve from the root. If you used --baseurl "/rclone" thenrclone would serve from a URL starting with "/rclone/". This is useful@@ -12784,8 +13355,20 @@ To create an htpasswd file:htpasswd -B htpasswd userhtpasswd -B htpasswd anotherUser+Use the same hash format and cost for every entry in the file. If the+entries differ, the time taken to reject a wrong password can reveal+which usernames exist.+The password file can be updated while rclone is running.+When authentication is enabled every request must be authenticated,+including HTTP OPTIONS requests, as the answer to an OPTIONS request can+reveal whether a path exists. Health checks or monitoring which use+unauthenticated OPTIONS requests will need to send credentials. The+exception is browser CORS preflight requests when --allow-origin is set,+which are answered without authentication as browsers never send+credentials with them.+Use --realm to set the authentication realm.Use --salt to change the password hashing salt from the default.@@ -12842,6 +13425,21 @@ correctly in the request. (See the AWS docs).--auth-key can be repeated for multiple auth pairs. If --auth-key is notprovided then serve s3 will allow anonymous access.+Alternatively --auth-proxy can be used to look up the secret for each+access key ID and choose the backend it maps to (see Auth Proxy below).+When an auth proxy is in use --auth-key is ignored and every request+must be signed with the secret the proxy returns for its access key ID.+Each access key ID is a separate user, whose multipart uploads and+object metadata are private to it, even if the proxy maps several access+key IDs to the same backend.++As on AWS, every x-amz-* header in an authenticated request must be in+its signed headers list; a request with an unsigned x-amz-* header is+refused with AccessDenied. S3 clients sign these headers as a matter of+course, but a header added after signing (for example by a proxy in+front of the server, or x-amz-meta-* headers sent with a presigned URL+which only signs the host) is refused.+Like all rclone flags --auth-key can be set via environment variables,in this case RCLONE_AUTH_KEY. Since this flag can be repeated, the inputto RCLONE_AUTH_KEY is CSV encoded. Because the accessKey,secretKey has a@@ -12869,7 +13467,15 @@ part of the hostname (such as mybucket.local)Use --etag-hash if you want to change the hash uses for the ETag. Notethat using anything other than MD5 (the default) is likely to cause-problems for S3 clients which rely on the Etag being the MD5.+problems for S3 clients which rely on the Etag being the MD5. Use+--etag-hash auto to use the best hash the backend supports - with+--auth-proxy that of each user's backend.++The ETag of an object made by a multipart upload is its hash, like any+other object, rather than the -N suffixed ETag AWS S3 gives these. If+the remote doesn't have the MD5 of objects, HeadObject and listings only+return the MD5 of objects uploaded through serve s3 while the object's+metadata is kept (see below).Quickstart@@ -12904,29 +13510,76 @@ which is defined like this:access_key_id = ACCESS_KEY_IDsecret_access_key = SECRET_ACCESS_KEY-Multipart uploads+Object uploads (PUT)++A PutObject upload only ever changes the object at its key atomically,+on success, a failed or interrupted PUT neither removes nor overwrites+the object already stored at the key, and never leaves a partial object+visible at it.-By default serve s3 streams each multipart upload, in part-number order,-into a single PutStream upload to the underlying remote, so the whole-file is never buffered in memory - memory use stays bounded by the parts-in flight. The remote then performs its own internal upload (for example-its own multipart upload, still with bounded memory). This works for any-remote that supports PutStream, which is nearly all of them, including-through crypt.+Remotes that upload atomically (e.g. object stores such as s3) are+streamed straight to the destination. On remotes where a partial upload+would otherwise be visible (e.g. local), and whenever --vfs-cache-mode+is writes or above, the upload is written to a temporary object that is+renamed into place on success; these remotes need to support a+server-side move or copy for this (nearly all do - without move or copy+the upload is written directly and a failed PUT may leave a partial+object at the key). If serve s3 is killed part-way through an upload the+temporary object (named with a leading .rclone_temp_put_) may be left+behind; it is hidden from S3 listings but must be removed manually.++Multipart uploads-The upload is atomic so the destination object only ever changes on a+Multipart uploads are written, in part-number order, to a temporary+object which is renamed into place, server-side, on completion, so the+upload is atomic. The object at the key only ever changes on asuccessful completion. A failed or aborted upload never affects any-object already stored under that name. Remotes that upload atomically-already (object stores such as s3) are streamed straight to the-destination. On remotes where a partial upload would otherwise be-visible (such as local), the parts are streamed to a temporary object-that is moved into place, server-side, on completion; these remotes-therefore also need to support a server-side move or copy.+object already stored under that name and a partly-uploaded object never+becomes visible under it.++With the default --vfs-cache-mode off serve s3 streams each multipart+upload, in part-number order, into a single streaming upload to the+underlying remote, so the whole file is never buffered in memory: only+parts which arrive out of order are buffered, within the limits+described below. The remote then performs its own internal upload (for+example its own multipart upload, still with bounded memory). Remotes+that don't support streaming uploads (those that must know the file size+before the upload starts, such as onedrive, pcloud, jottacloud, mailru,+opendrive, putio, protondrive and zoho) have the parts spooled to a+temporary file on local disk instead, and uploaded with the size then+known on completion, so they need local disk space for the largest+objects in flight rather than memory.++With --vfs-cache-mode writes (or full) the parts are written to a+temporary file in the VFS cache and uploaded by the VFS write-back - see+Multipart uploads and the VFS cache below.++The rename into place needs the remote to support a server-side move or+copy, which nearly all do. It is a cheap rename on most remotes, but on+object stores without a real rename (such as s3 itself) the move is+performed as a server-side copy and delete of the whole object, which+can take time and API calls for large objects. Concurrent multipart+uploads of the same key (which S3 permits) are safe. Each writes its own+temporary object and the last to complete wins.++On the few remotes that support neither server side move nor copy, the+parts are written straight to the destination object instead and never+buffered in memory. This is at some cost in atomicity - the incomplete+object is visible under its final name while the upload is in flight, as+it also is for a plain object PUT on such remotes, and concurrent+multipart uploads of the same key write to the same object and can+interleave. A failed or aborted upload still leaves any pre-existing+object untouched provided the remote uploads atomically and the VFS+cache is off; on a remote where partial uploads are visible it may leave+partial data at the key (like a plain PUT there), and with+--vfs-cache-mode writes (or full) a write to the cache cannot be+abandoned, so an aborted upload's partial data is written back to the+remote as if it had completed.Features-- The whole object is never buffered in memory; memory use is bounded- by the parts in flight, not the upload size.+- The whole object is never buffered in memory; only parts arriving+ out of order are, within fixed limits, whatever the upload size.- Parts can be any size. Clients that don't produce uniform-sizedparts work fine - for example PostgreSQL backup tools such aspgBarman and pgBackRest, which flush an upload buffer once it grows@@ -12937,68 +13590,265 @@ Features- The destination object only ever changes atomically, on completion:an aborted or failed upload leaves any pre-existing object of thesame name untouched, and a partly-uploaded object never becomes- visible.-- Backend-agnostic - it only needs the remote to support PutStream- (plus a server-side move or copy on remotes that don't upload- atomically).+ visible (except on the few remotes with no server-side move or copy,+ as above).+- Multipart uploads go through the VFS like any other upload, so they+ show in rclone's transfer stats and obey --bwlimit.+- Backend-agnostic - it only needs the remote to support a server-side+ move or copy for the rename into place, which nearly all do; a+ remote without streaming upload support spools to local disk as+ above.Limitations- Parts must arrive in ascending, contiguous part-number order (1, 2,- 3, ...). Parts the client uploads concurrently or out of order are- buffered until their turn. The memory used for this buffering is- capped, per upload, by --multipart-streaming-buffer-limit (default- 256M, 0 for no limit): a part that would take the buffer over the- limit is stalled until the stream drains, so a client that uploads- faster than the remote can accept sees backpressure rather than- unbounded server memory use. Since a stalled part holds its HTTP- request open, clients whose upload concurrency times chunk size- exceeds the limit may need a longer read timeout when the remote is- slow. Non-contiguous part numbers are rejected on completion.- Configure the client to upload in part order, ideally with low- concurrency, for the lowest memory use.+ 3, ...). The part the stream needs next is written straight through+ as it arrives, so it is never buffered, whatever its size. Parts the+ client uploads concurrently or out of order are buffered until their+ turn. The memory used for this buffering is capped, per upload, by+ --multipart-streaming-buffer-limit (default 256M, 0 for no limit),+ and across all the uploads of each user (each access key) by+ --multipart-streaming-buffer-total (default 1G, 0 for no limit),+ counting each buffered part in the whole 1 MiB memory pages it+ occupies: a part that would take the buffer over either limit is+ stalled until the stream drains (or, if it is bigger than the+ limits, until it is the next part), so a client that uploads faster+ than the remote can accept sees backpressure rather than unbounded+ server memory use. A stalled part holds its HTTP request open, and+ one stalled for more than a minute is failed with a SlowDown error,+ which S3 clients retry. Clients whose upload concurrency times part+ size exceeds the limits will see these stalls and retries when the+ remote is slower than the client. Non-contiguous part numbers are+ rejected on completion. Configure the client to upload in part+ order, ideally with low concurrency, for the lowest memory use.- A part uploaded again before completion - typically a clientretrying after a timeout - is accepted: if the earlier copy is stillbuffered it is replaced, and if it has already been streamed an- identical re-upload is a no-op. What isn't possible is replacing a- part that has already been streamed with different content - that is- rejected. A failure in the stream to the remote itself still aborts- the whole upload and the client must start it again. (The remote's- own upload still retries its internal chunks.)+ identical re-upload is a no-op (a re-upload which arrives while the+ part is being streamed waits for it to finish first). If a part+ fails part way through being streamed (for example the client's+ connection drops) the client's retry of it carries on where it+ stopped, provided it has the same content. What isn't possible is+ replacing a part that has already been streamed, in whole or in+ part, with different content - that is rejected. A part which+ arrives in full but is then rejected, for example because it doesn't+ match the Content-MD5 the client declared, has already been+ streamed, so the whole upload is failed and the client must start it+ again. So is a failure in the stream to the remote itself. (The+ remote's own upload still retries its internal chunks.)+- A CompleteMultipartUpload which fails because the parts it lists+ don't match those uploaded, or there is a gap in the part numbers,+ leaves the upload open, as on S3, so the client can upload any+ missing parts and complete it again. If writing the completed object+ to the remote fails, the upload is failed and must be started again.- Parts are serialised into one stream, so ingest from the client is- effectively single-threaded, although the remote's own upload still- runs concurrently.-- On remotes that don't upload atomically (such as local), the- completed object is moved into place with a server-side operation.- This is a cheap rename on most such remotes. On these remotes, if- serve s3 is killed part-way through an upload the temporary object- (named with a leading .rclone_multipart_upload_) may be left behind;- it is hidden from S3 listings but must be removed manually.+ effectively single-threaded. When streaming, the remote's own upload+ runs concurrently with the parts arriving; with the local disk spool+ or the VFS cache the upload to the remote only starts on completion.+- If serve s3 is killed part-way through an upload the temporary+ object (named with a leading .rclone_temp_multipart_) may be left+ behind; it is hidden from S3 listings but must be removed manually.++Multipart uploads and the VFS cache++With --vfs-cache-mode writes (or full) multipart uploads do not stream+to the remote at all. The parts are written, in part-number order, to a+temporary file in the VFS cache. On completion the file is renamed into+place and uploaded by the VFS write-back, exactly like a plain object+PUT. This needs no streaming upload support from the remote. The rename+normally happens in the cache before the upload has started, but the VFS+requires the remote to support a server-side move or copy to rename+files at all (and uses one if the temporary file has already been+written back, e.g. with --vfs-write-back 0). On remotes without either,+the parts are written to the cache directly under the final key instead:+the upload still never touches memory, but it loses its atomicity - the+in-flight upload is visible at the key, and an aborted upload cannot be+abandoned once in the cache, so its partial data is written back to the+remote as if it were a completed object.++Remotes that benefit from --vfs-cache-mode writes:++- Remotes over slow or unreliable links. A failure in a streamed+ upload aborts the whole multipart upload and the client must start+ again from the first part; a failed write-back upload is retried by+ the VFS (see --vfs-cache-max-age and friends) without the client+ being involved. Ingest from the client also runs at local disk speed+ rather than being throttled to the remote's pace.+- Workloads that read back or overwrite what they just wrote. The+ completed object stays in the cache, so subsequent GET/HEAD requests+ are served locally, and plain PUTs and multipart uploads to the same+ key go through the same cache entry so the last write wins+ regardless of upload style.++The trade-offs of the VFS cache:++- The whole object lands on local disk, so the cache (--cache-dir)+ needs space for the largest objects in flight; --vfs-cache-max-size+ cannot evict files which are still being uploaded.+- The 200 OK for CompleteMultipartUpload means the data is safely in+ the local cache, not yet on the remote - the same durability the+ cache gives plain PUTs. If an acknowledgement must mean the data has+ reached the remote (for example WAL archiving), use the default+ --vfs-cache-mode off.+- The upload to the remote only starts on completion, rather than+ overlapping with the parts arriving, so the data reaches the remote+ later than with streaming.+- If serve s3 is killed part-way through an upload, the temporary file+ survives in the cache and the VFS cache recovery uploads it to the+ remote on restart as a temporary object (named with a leading+ .rclone_temp_multipart_); as with the streaming path, it is hidden+ from S3 listings but must be removed manually.++Cleaning up temporary objects++If serve s3 is killed part-way through an upload it can leave a+temporary object behind, named with a leading .rclone_temp_. This whole+prefix is reserved: any object whose name (the last /-separated segment+of its key) starts with .rclone_temp_ is hidden from S3 listings, so+don't give real objects such names - an existing object with such a name+disappears from listings (though it stays accessible directly by its+key: only listings hide reserved names, GET, HEAD and DELETE of the+exact key still work). A temporary object never holds acknowledged+data - uploads whose temporary object survived were never confirmed to+the client - so old ones are safe to delete:++ rclone delete --min-age 24h --include ".rclone_temp_*" remote:path++The --min-age protects uploads which are still in progress: make sure it+is longer than your longest upload, especially if several serve s3+instances share the same remote.++rclone v1.75 named its temporary multipart objects+.rclone_multipart_upload_*; leftovers from an older server are also+hidden from listings and can be cleaned up the same way.++Abandoned uploads++A client which starts a multipart upload and vanishes without either+completing or aborting it would otherwise hold on to its resources+forever.++An incomplete multipart upload which has had no activity for+--multipart-expiry (default 24h) is therefore aborted and cleaned up,+exactly as if the client had called AbortMultipartUpload, and a NOTICE+is logged.++An upload with a part still being received is never expired, however+slowly the part is arriving, and each completed part restarts the clock,+so the expiry only needs to outlast the client's pauses between parts,+not the whole upload.++Late operations on an expired upload fail with NoSuchUpload, as they do+on real S3 when a lifecycle rule has aborted the upload. Set+--multipart-expiry 0 to keep incomplete uploads forever - but note that+abandoned uploads are then never cleaned up and go on counting towards+--multipart-max-uploads below, so enough of them will stop the user+starting any new multipart upload until the server is restarted.++Each multipart upload in progress holds server resources - for example+the buffers the remote's own upload uses once the first part has+arrived - so each user (each access key) can have no more than+--multipart-max-uploads (default 1000, 0 for no limit) in progress at+once. Starting another fails with a SlowDown error, which S3 clients+retry, and a one-off NOTICE is logged.Disabling streaming-If you pass --disable-multipart-streaming, or the remote doesn't support-PutStream (or doesn't upload atomically and can't move or copy-server-side), multipart uploads are instead buffered in memory by the-underlying S3 library: every part is held in memory and the whole object-is written out in one go when the upload completes (the previous-behaviour). This removes the in-order/contiguous-part restriction above,+If you pass --disable-multipart-streaming, multipart uploads are instead+buffered in memory by the underlying S3 library: every part is held in+memory and the whole object is written out in one go when the upload+completes. This removes the in-order/contiguous-part restriction above,so parts can be uploaded in any order, but memory use grows with thesize of the upload, so it is only suitable for small objects. A one-off-NOTICE is logged the first time this happens.--Alternatively, if the client is an rclone s3 remote (like the [serves3]-example above), you can set use_multipart_uploads = false on it so it-uploads each object as a single stream and skips multipart uploads-altogether.+NOTICE is logged the first time this happens. This flag is the only+thing that makes multipart uploads buffer in memory - it is never done+because of missing remote capabilities. The limits on multipart uploads+above (--multipart-streaming-buffer-limit,+--multipart-streaming-buffer-total and --multipart-max-uploads) do not+apply to uploads buffered in memory, so don't use this flag if the+server is open to clients you don't trust. Consider+--vfs-cache-mode writes instead, which buffers the upload in the VFS+cache on disk and takes precedence over --disable-multipart-streaming.++Memory use++serve s3 never buffers whole objects in memory (except with+--disable-multipart-streaming, above), and limits the memory each kind+of request can use, so that a client, even a hostile one, can't make it+use unbounded memory.++- Object data. PutObject bodies and multipart parts arriving in order+ are streamed straight through to the remote. Parts arriving out of+ order are buffered, up to --multipart-streaming-buffer-limit per+ upload and --multipart-streaming-buffer-total per user.+- Uploads in progress. Each user can have at most+ --multipart-max-uploads multipart uploads in progress. An upload+ holds little memory until its first part arrives. After that, with+ --vfs-cache-mode off, it holds a --streaming-upload-cutoff sized+ buffer plus whatever the remote's own upload buffers, which for many+ remotes is up to their chunk size times their upload concurrency+ (for example --s3-chunk-size times --s3-upload-concurrency). Plain+ PutObject uploads use the same while they are in progress.+- Metadata. The metadata of at most --metadata-max-objects objects of+ each user is kept, each limited to 2 KB.+- Request bodies. The XML bodies of requests such as+ CompleteMultipartUpload and DeleteObjects are limited to 10 MiB, and+ DeleteObjects to 1000 keys. A browser form (POST) upload holds at+ most 1 MiB in memory, spooling the rest to a temporary file, and is+ limited to 5 GB.+- Listings. Each page of a ListObjects listing reads only the+ directories needed to fill it, and returns at most 1000 keys.+ Directory listings are cached by the VFS directory cache.++The largest of these is usually the remote's upload buffers, which can+be up to --multipart-max-uploads times the buffers of one upload for+each user if clients start many uploads and send part of each. The+limits apply to each user separately, so that one user can't stop others+uploading, which means the memory use grows with the number of users. To+limit it, lower --multipart-max-uploads, the remote's chunk size or+upload concurrency, or set --max-buffer-memory to limit the memory used+by the buffers of all uploads to remotes which use rclone's buffer pool+(including s3, azureblob and b2). Buffered multipart parts come from the+same pool, so if you set --max-buffer-memory make it comfortably bigger+than --multipart-streaming-buffer-total times the number of users+uploading at once, otherwise buffered parts can use up all the memory+the remote needs to upload the parts which would free them.++The number of connections is not limited, so if the server is open to+clients you don't trust, use --auth-key or --auth-proxy and consider+putting it behind a reverse proxy which limits connections.++Design goals++These goals explain many of the limitations below.++- Objects are plain files on the remote. A bucket is a directory in+ the root and a key is a path in it, so the remote stays usable by+ other tools, such as rclone mount or rclone sync. So keys must be+ paths (see Object names).+- No database. serve s3 stores nothing but the objects. What the+ remote can't hold, such as user metadata, tags, the MD5 ETag of an+ object on a remote without MD5 and the directory markers put, is+ kept in memory as a best effort, and lost on restart.+- Works with any rclone remote. What serve s3 can do depends on what+ the remote supports, such as the hashes used for ETags and+ modification times, rather than needing a particular backend.+- Bounded memory. Uploads, including multipart uploads, are streamed+ to the remote and the memory each request can use is limited (see+ Memory use). Nothing is written to local disk unless the VFS cache+ is used or the remote can't stream uploads (see Multipart uploads).+- Refuse rather than ignore. Unsupported features are refused, or+ answered as S3 answers when the feature isn't configured, so a+ client is never told something was done which wasn't (see+ Compatibility).+- A storage gateway, not a full S3. serve s3 implements the parts of+ the S3 API clients use to store and fetch objects. Versioning, ACLs,+ bucket policies, lifecycle rules and the like are out of scope.+- Compatible with rclone's s3 backend. This includes its directory+ markers, which the tests check.Bugs-Multipart server side copies do not work (see #7454). These take a very-long time and eventually fail. The default threshold for multipart-server side copies is 5G which is the maximum it can be, so files above-this side will fail to be server side copied.-For a current list of serve s3 bugs see the serve s3 bug category onGitHub.@@ -13006,56 +13856,221 @@ Limitationsserve s3 will treat all directories in the root as buckets and ignoreall files in the root. You can use CreateBucket to create folders under-the root, but you can't create empty folders under other folders not in-the root.+the root, and put a directory marker to create an empty folder below+them (see Object names).When using PutObject or DeleteObject, rclone will automatically createor clean up empty folders. If you don't want to clean up empty folders-automatically, use --no-cleanup.--When using ListObjects, rclone will use / when the delimiter is empty.-This reduces backend requests with no effect on most operations, but if-the delimiter is something other than / and empty, rclone will do a full-recursive search of the backend, which can take some time.--Versioning is not currently supported.--Metadata will only be saved in memory other than the rclone mtime-metadata which will be set as the modification time of the file.+automatically, use --directory-markers (see Object names). The+deprecated --no-cleanup does the same.++When using ListObjects with the delimiter /, rclone lists the directory+the prefix is in, showing its subdirectories as common prefixes,+including empty ones. Without a delimiter the listing is recursive:+rclone walks the directory tree below the prefix, reading only as many+directories as it needs to fill each page, and lists empty directories+as their directory markers (see below). Any other delimiter is supported+too, but needs the whole directory tree below the prefix walking, unless+the delimiter is in the directory names.++Versioning is not supported (see Design goals).++Conditional writes, PutObject and CompleteMultipartUpload with+If-None-Match or If-Match, are supported. The conditions are checked+before the object is written, and again as it is moved into place when+it is written to a temporary object first. This happens for multipart+uploads and, for PutObject, with remotes which can upload partial files+(such as local disk) or with --vfs-cache-mode minimal or above. The+second check is atomic against other writes through serve s3, but+neither is atomic against changes made to the remote some other way. If+the remote doesn't support the --etag-hash, objects have no ETag so+If-Match with an ETag always fails.++Metadata will only be saved in memory (see Design goals) other than the+rclone mtime metadata which will be set as the modification time of the+file. The metadata of at most --metadata-max-objects objects (default+100000, 0 for no limit) of each user (each access key) is kept; after+that the metadata of the user's least recently used object is forgotten.+The metadata of an object is ignored once the object has been changed+some other way, for example by another --auth-proxy user or directly on+the backend. Object tags are kept with the metadata in the same way.Object namesserve s3 stores objects as files in the backend, so object keys aremapped to file paths rather than treated as the opaque strings AWS S3allows. Keys must be in canonical path form: keys that contain .. or .-path segments, repeated slashes (//), or a leading or trailing slash are-rejected with a 400 Bad Request (InvalidArgument) instead of being-normalised, since normalising them could alias two distinct keys to the-same file or resolve a key outside its bucket. This matches the-behaviour of other S3 servers such as MinIO.--Supported operations--serve s3 currently supports the following operations.--- Bucket- - ListBuckets- - CreateBucket- - DeleteBucket-- Object- - HeadObject- - ListObjects- - GetObject- - PutObject- - DeleteObject- - DeleteObjects- - CreateMultipartUpload- - CompleteMultipartUpload- - AbortMultipartUpload- - CopyObject- - UploadPart--Other operations will return error Unimplemented.+path segments, repeated slashes (//), or a leading slash are rejected+with a 400 Bad Request (InvalidArgument) instead of being normalised,+since normalising them could alias two distinct keys to the same file or+resolve a key outside its bucket. This matches the behaviour of other S3+servers such as MinIO.++A key ending in /, such as dir/, is a directory marker, standing for the+directory dir. Putting an empty directory marker creates the directory+(one with data is rejected with 400 Bad Request), and deleting it+removes the directory if it is empty. Directory markers are found by+HeadObject, GetObject and CopyObject for any directory, but only the+markers of empty directories, and of the directories a marker was put+for, are listed as objects, as other directories may not have been made+by putting a marker. The markers put are remembered in memory with the+metadata (see Limitations), so after a restart, or once forgotten, the+marker of a directory holding other keys is no longer listed. While a+marker is remembered its directory isn't removed when the last key below+it is deleted.++With --directory-markers every directory is listed as a directory+marker, so the markers are listed after a restart too, as needed by+clients which rely on them, such as rclone's s3 backend with+--s3-directory-markers. As every directory has a marker, a directory+isn't removed when the last key below it is deleted. Deleting the marker+of a directory which holds other keys stops it being listed, and the+directory is then removed with the last key below it, but this is only+remembered in memory, like the markers put without --directory-markers.++As keys are file paths, a key can't be both an object and a directory+holding other objects, for example a/b and a/b/c. Writing an object+which would need this fails with 400 Bad Request (InvalidRequest).++Compatibility++serve s3 implements the parts of the S3 API that S3 clients commonly use+to store and fetch objects. Features it doesn't support are refused with+501 NotImplemented, or answered as AWS S3 answers for a bucket without+that feature configured, rather than being silently ignored.++ -------------------------------------------------------------------------------------------+ Feature Supported Notes+ ----------------------- ------------------------------ ------------------------------------+ ListBuckets, Yes Buckets are the directories in the+ CreateBucket, root of the remote.+ DeleteBucket,+ HeadBucket,+ GetBucketLocation++ ListObjects, Yes Prefixes, delimiters, markers,+ ListObjectsV2 start-after, continuation tokens,+ fetch-owner and encoding-type=url+ all work. Delimiters other than /+ need the whole directory tree+ walking.++ HeadObject, GetObject Yes Including Range, partNumber, the+ response-* header overrides and the+ If-Match, If-None-Match,+ If-Modified-Since and+ If-Unmodified-Since conditions.++ PutObject Yes Including Content-MD5, aws-chunked+ uploads and uploads with no+ Content-Length.++ CopyObject Yes Including the metadata and tagging+ directives and the+ x-amz-copy-source-if-* conditions.++ DeleteObject, Yes+ DeleteObjects++ Multipart uploads Yes CreateMultipartUpload, UploadPart,+ UploadPartCopy,+ CompleteMultipartUpload,+ AbortMultipartUpload,+ ListMultipartUploads and ListParts.+ The ETag of the object is its hash,+ not the -N suffixed one AWS gives.+ The 5 MiB minimum part size isn't+ enforced.++ Conditional writes Partial If-Match and If-None-Match on+ PutObject and+ CompleteMultipartUpload are checked,+ but not atomically against changes+ made to the remote some other way+ (see Limitations).++ Authentication Partial AWS Signature Version 4, in headers,+ presigned URLs and browser form+ (POST) uploads. Signature Version 2+ isn't supported.++ Anonymous access Partial All requests are allowed without+ --auth-key or --auth-proxy, and none+ are with them, as buckets and+ objects can't be made public.++ Metadata Partial x-amz-meta-*, the storage class and+ the standard headers such as+ Content-Type are kept in memory only+ (see Limitations), except the+ modification time. The storage class+ is only recorded, so doesn't change+ how the object is stored.++ Directory markers Partial Keys ending in / are stored as+ directories, so can't hold data. A+ marker put for a directory holding+ other keys is only listed while it+ is remembered in memory, unless+ --directory-markers is set (see+ Object names).++ Checksums Partial The x-amz-checksum-* of a PutObject+ (CRC32, CRC32C, CRC64NVME, SHA1 or+ SHA256) is checked and returned with+ the metadata. The Content-MD5 and+ x-amz-checksum-* of the bodies of+ other requests, such as+ DeleteObjects, are checked. The+ checksums of multipart uploads+ aren't supported.++ GetObjectAttributes Partial Everything but the checksums of+ multipart uploads.++ Object tagging Partial GetObjectTagging, PutObjectTagging+ and DeleteObjectTagging work, but+ tags are kept in memory with the+ metadata.++ Bucket owner No serve s3 has no AWS account IDs, so+ x-amz-expected-bucket-owner and+ x-amz-source-expected-bucket-owner+ are ignored.++ ACLs Partial GetBucketAcl and GetObjectAcl give+ the owner full control. Only the+ private ACL can be set.++ CORS Partial Preflight (OPTIONS) requests are+ answered, allowing any origin. CORS+ configuration (PutBucketCors) isn't+ supported.++ Server side encryption No SSE-S3 (AES256) is accepted but the+ data isn't encrypted. SSE-KMS and+ SSE-C are refused.++ Versioning No Buckets are reported as unversioned+ and ListObjectVersions lists each+ object as version null. Enabling+ versioning is refused.++ Object lock, retention No+ and legal hold++ Bucket configuration No Bucket policies, tagging, lifecycle,+ logging, encryption, website,+ replication, notifications and the+ like. Getting them fails as for a+ bucket without them configured.++ SelectObjectContent, No+ RestoreObject, torrents+ -------------------------------------------------------------------------------------------++The ceph s3-tests can be run against serve s3 to check its+compatibility. See cmd/serve/s3/s3tests for how.Authentication@@ -13085,8 +14100,20 @@ To create an htpasswd file:htpasswd -B htpasswd userhtpasswd -B htpasswd anotherUser+Use the same hash format and cost for every entry in the file. If the+entries differ, the time taken to reject a wrong password can reveal+which usernames exist.+The password file can be updated while rclone is running.+When authentication is enabled every request must be authenticated,+including HTTP OPTIONS requests, as the answer to an OPTIONS request can+reveal whether a path exists. Health checks or monitoring which use+unauthenticated OPTIONS requests will need to send credentials. The+exception is browser CORS preflight requests when --allow-origin is set,+which are answered without authentication as browsers never send+credentials with them.+Use --realm to set the authentication realm.Use --salt to change the password hashing salt from the default.@@ -13119,6 +14146,14 @@ accept in the HTTP header.will overriding existing values. The flag may be repeated to addmultiple headers. Use the format Header-Name: value.+--allow-origin enables cross-origin (CORS) requests from browsers by+setting the Access-Control-Allow-Origin header on every response to the+origin given, e.g. --allow-origin https://app.example.com. Browser CORS+preflight OPTIONS requests are answered by rclone without needing+authentication as browsers never send credentials with them. Setting+this to * allows any origin which is a security risk on an authenticated+server.+--baseurl controls the URL prefix that rclone serves from. By defaultrclone will serve from the root. If you used --baseurl "/rclone" thenrclone would serve from a URL starting with "/rclone/". This is useful@@ -13229,8 +14264,8 @@ at all times. The buffered data is bound to one open file and won't beshared.This flag is a upper limit for the used memory per open file. The buffer-will only use memory for data that is downloaded but not not yet read.-If the buffer is empty, only a small amount of memory will be used.+will only use memory for data that is downloaded but not yet read. If+the buffer is empty, only a small amount of memory will be used.The maximum memory used by rclone for buffering can be up to--buffer-size * open files.@@ -13279,11 +14314,12 @@ that haven't been accessed for the longest. This cache flushing strategyis efficient and more relevant files are likely to remain cached.The --vfs-cache-max-age will evict files from the cache after the set-time since last access has passed. The default value of 1 hour will-start evicting files from cache that haven't been accessed for 1 hour.-When a cached file is accessed the 1 hour timer is reset to 0 and will-wait for 1 more hour before evicting. Specify the time with standard-notation, s, m, h, d, w .+time since last access has passed; it is based on access time, not on+when the file was first added to the cache. The default value of 1 hour+will start evicting files from cache that haven't been accessed for 1+hour. When a cached file is accessed the 1 hour timer is reset to 0 and+will wait for 1 more hour before evicting. Specify the time with+standard notation, s, m, h, d, w .You should not run two copies of rclone using the same VFS cache withthe same or overlapping remotes if using --vfs-cache-mode > off. This@@ -13430,8 +14466,8 @@ Rclone reads --vfs-read-chunk-streams chunks of size--vfs-read-chunk-size concurrently. The size for each read will stayconstant.-This improves performance performance massively on high latency links or-very high bandwidth links to high performance object stores.+This improves performance massively on high latency links or very high+bandwidth links to high performance object stores.Some experimentation will be needed to find the optimum values of--vfs-read-chunk-size and --vfs-read-chunk-streams as these will depend@@ -13636,6 +14672,122 @@ If the file has no metadata it will be returned as {} and if there is anerror reading the metadata the error will be returned as{"error":"error string"}.+Auth Proxy++If you supply the parameter --auth-proxy /path/to/program then rclone+will use that program to generate backends on the fly which then are+used to authenticate incoming requests. This uses a simple JSON based+protocol with input on STDIN and output on STDOUT.++PLEASE NOTE: --auth-proxy and --authorized-keys cannot be used together,+if --auth-proxy is set the authorized keys option will be ignored.++There is an example program bin/test_proxy.py in the rclone source code.++The program's job is to take a user and pass on the input and turn those+into the config for a backend on STDOUT in JSON format. This config will+have any default parameters for the backend added, but it won't use+configuration from environment variables or command line options - it is+the job of the proxy program to make a complete config.++This config generated must have this extra parameter++- _root - root to use for the backend++And it may have these parameters++- _obscure - comma separated strings for parameters to obscure+- _secret_access_key - the secret for S3 access key auth (see below)++If password authentication was used by the client, input to the proxy+process (on STDIN) would look similar to this:++ {+ "user": "me",+ "pass": "mypassword",+ "client_ip": "192.168.1.1"+ }++If public-key authentication was used by the client, input to the proxy+process (on STDIN) would look similar to this:++ {+ "user": "me",+ "public_key": "AAAAB3NzaC1yc2EAAAADAQABAAABAQDuwESFdAe14hVS6omeyX7edc...JQdf",+ "client_ip": "192.168.1.1"+ }++If the client authenticated with an S3 access key (rclone serve s3), the+client never sends its secret, only a signature made with it, so the+input contains just the access key ID as the user with no pass or+public_key:++ {+ "user": "AKIAIOSFODNN7EXAMPLE",+ "client_ip": "192.168.1.1"+ }++In this case the program must look up the secret access key for that+access key ID and return it in the _secret_access_key field of the+output. Rclone then uses that secret to verify the signature on the+request, refusing the request if it does not match. This means the proxy+program is the source of truth for both the credentials and the backend+they map to. If the program does not return _secret_access_key or+returns it empty the request is refused.++The program's answer for an access key ID is cached (see below) but is+checked with the program again after 5 minutes even if the access key ID+is in constant use, so revoking an access key ID in the program takes+effect within 5 minutes. A rotated secret takes effect on the first+request signed with it.++The client_ip key holds the IP address the client connected from,+without a port number. It can be used to restrict logins to certain+networks, or to log authentication attempts centrally. It is omitted if+the client has no IP address, for example when connecting over a unix+socket. Note that if rclone is behind a reverse proxy this will be the+address of the reverse proxy and not the original client.++And as an example return this on STDOUT++ {+ "type": "sftp",+ "_root": "",+ "_obscure": "pass",+ "user": "me",+ "pass": "mypassword",+ "host": "sftp.example.com"+ }++This would mean that an SFTP backend would be created on the fly for the+user and pass/public_key returned in the output to the host given. Note+that since _obscure is set to pass, rclone will obscure the pass+parameter before creating the backend (which is required for sftp+backends).++The program can manipulate the supplied user in any way, for example to+make proxy to many different sftp backends, you could make the user be[email protected] and then set the host to example.com in the output and+the user to user. For security you'd probably want to restrict the host+to a limited list.++An internal cache of backends is keyed on the user, the kind of+credential presented, a hash of the pass or public_key, and the+client_ip. This means that if a user's password or public-key changes,+the client connects from a new IP address, or the proxy returns+different config parameters (eg a rotated api_key), a fresh backend will+be created on the next request rather than the cached one being reused.+A backend created for a public key login is never reused for a password+login, so a public key cannot be sent as a password to skip the program.++A cached answer is checked with the program again after 5 minutes even+if the user is in constant use, so revoking a user in the program takes+effect within 5 minutes. For serve sftp this applies to new logins - an+existing connection is not closed.++This can be used to build general purpose proxies to any kind of backend+that rclone supports.+rclone serve s3 remote:path [flags]Options@@ -13649,7 +14801,8 @@ Options--client-ca string Client certificate authority to verify clients with--dir-cache-time Duration Time to cache directory entries for (default 5m0s)--dir-perms FileMode Directory permissions (default 777)- --disable-multipart-streaming Buffer multipart uploads in memory instead of streaming them to the backend (see the Multipart uploads docs section)+ --directory-markers List every directory as a directory marker object+ --disable-multipart-streaming Buffer multipart uploads in memory instead of streaming them to the backend--etag-hash string Which hash to use for the ETag, or auto or blank for off (default "MD5")--file-perms FileMode File permissions (default 666)--force-path-style If true use path style access if false use virtual hosted style (default true)@@ -13659,10 +14812,14 @@ Options--key string TLS PEM Private key--link-perms FileMode Link permissions (default 666)--max-header-bytes int Maximum size of request header (default 4096)+ --metadata-max-objects int Maximum number of objects of a user whose metadata is kept in memory, 0 for unlimited (default 100000)--min-tls-version string Minimum TLS version that is acceptable (default "tls1.0")- --multipart-streaming-buffer-limit SizeSuffix Maximum memory buffered per streamed multipart upload for parts arriving out of order, 0 for unlimited (see the Multipart uploads docs section) (default 256Mi)+ --multipart-expiry Duration Abort incomplete multipart uploads idle for longer than this, 0 to keep forever (default 1d)+ --multipart-max-uploads int Maximum number of streamed multipart uploads a user can have in progress at once, 0 for unlimited (default 1000)+ --multipart-streaming-buffer-limit SizeSuffix Maximum memory buffered per streamed multipart upload for parts arriving out of order, 0 for unlimited (default 256Mi)+ --multipart-streaming-buffer-total SizeSuffix Maximum memory buffered by all the streamed multipart uploads of a user for parts arriving out of order, 0 for unlimited (default 1Gi)--no-checksum Don't compare checksums on up/download- --no-cleanup Not to cleanup empty folder after object is deleted+ --no-cleanup Deprecated: use --directory-markers instead--no-modtime Don't read/write the modification time (can speed things up)--no-seek Don't allow seeking in files--pass string Password for authentication@@ -13861,8 +15018,8 @@ at all times. The buffered data is bound to one open file and won't beshared.This flag is a upper limit for the used memory per open file. The buffer-will only use memory for data that is downloaded but not not yet read.-If the buffer is empty, only a small amount of memory will be used.+will only use memory for data that is downloaded but not yet read. If+the buffer is empty, only a small amount of memory will be used.The maximum memory used by rclone for buffering can be up to--buffer-size * open files.@@ -13911,11 +15068,12 @@ that haven't been accessed for the longest. This cache flushing strategyis efficient and more relevant files are likely to remain cached.The --vfs-cache-max-age will evict files from the cache after the set-time since last access has passed. The default value of 1 hour will-start evicting files from cache that haven't been accessed for 1 hour.-When a cached file is accessed the 1 hour timer is reset to 0 and will-wait for 1 more hour before evicting. Specify the time with standard-notation, s, m, h, d, w .+time since last access has passed; it is based on access time, not on+when the file was first added to the cache. The default value of 1 hour+will start evicting files from cache that haven't been accessed for 1+hour. When a cached file is accessed the 1 hour timer is reset to 0 and+will wait for 1 more hour before evicting. Specify the time with+standard notation, s, m, h, d, w .You should not run two copies of rclone using the same VFS cache withthe same or overlapping remotes if using --vfs-cache-mode > off. This@@ -14062,8 +15220,8 @@ Rclone reads --vfs-read-chunk-streams chunks of size--vfs-read-chunk-size concurrently. The size for each read will stayconstant.-This improves performance performance massively on high latency links or-very high bandwidth links to high performance object stores.+This improves performance massively on high latency links or very high+bandwidth links to high performance object stores.Some experimentation will be needed to find the optimum values of--vfs-read-chunk-size and --vfs-read-chunk-streams as these will depend@@ -14290,16 +15448,18 @@ This config generated must have this extra parameter- _root - root to use for the backend-And it may have this parameter+And it may have these parameters- _obscure - comma separated strings for parameters to obscure+- _secret_access_key - the secret for S3 access key auth (see below)If password authentication was used by the client, input to the proxyprocess (on STDIN) would look similar to this:{"user": "me",- "pass": "mypassword"+ "pass": "mypassword",+ "client_ip": "192.168.1.1"}If public-key authentication was used by the client, input to the proxy@@ -14307,9 +15467,41 @@ process (on STDIN) would look similar to this:{"user": "me",- "public_key": "AAAAB3NzaC1yc2EAAAADAQABAAABAQDuwESFdAe14hVS6omeyX7edc...JQdf"+ "public_key": "AAAAB3NzaC1yc2EAAAADAQABAAABAQDuwESFdAe14hVS6omeyX7edc...JQdf",+ "client_ip": "192.168.1.1"}+If the client authenticated with an S3 access key (rclone serve s3), the+client never sends its secret, only a signature made with it, so the+input contains just the access key ID as the user with no pass or+public_key:++ {+ "user": "AKIAIOSFODNN7EXAMPLE",+ "client_ip": "192.168.1.1"+ }++In this case the program must look up the secret access key for that+access key ID and return it in the _secret_access_key field of the+output. Rclone then uses that secret to verify the signature on the+request, refusing the request if it does not match. This means the proxy+program is the source of truth for both the credentials and the backend+they map to. If the program does not return _secret_access_key or+returns it empty the request is refused.++The program's answer for an access key ID is cached (see below) but is+checked with the program again after 5 minutes even if the access key ID+is in constant use, so revoking an access key ID in the program takes+effect within 5 minutes. A rotated secret takes effect on the first+request signed with it.++The client_ip key holds the IP address the client connected from,+without a port number. It can be used to restrict logins to certain+networks, or to log authentication attempts centrally. It is omitted if+the client has no IP address, for example when connecting over a unix+socket. Note that if rclone is behind a reverse proxy this will be the+address of the reverse proxy and not the original client.+And as an example return this on STDOUT{@@ -14333,11 +15525,19 @@ [email protected] and then set the host to example.com in the output andthe user to user. For security you'd probably want to restrict the hostto a limited list.-An internal cache of backends is keyed on the user and a hash of the-pass or public_key. This means that if a user's password or public-key-changes, or the proxy returns different config parameters (eg a rotated-api_key), a fresh backend will be created on the next request rather-than the cached one being reused.+An internal cache of backends is keyed on the user, the kind of+credential presented, a hash of the pass or public_key, and the+client_ip. This means that if a user's password or public-key changes,+the client connects from a new IP address, or the proxy returns+different config parameters (eg a rotated api_key), a fresh backend will+be created on the next request rather than the cached one being reused.+A backend created for a public key login is never reused for a password+login, so a public key cannot be sent as a password to skip the program.++A cached answer is checked with the program again after 5 minutes even+if the user is in constant use, so revoking a user in the program takes+effect within 5 minutes. For serve sftp this applies to new logins - an+existing connection is not closed.This can be used to build general purpose proxies to any kind of backendthat rclone supports.@@ -14538,6 +15738,14 @@ accept in the HTTP header.will overriding existing values. The flag may be repeated to addmultiple headers. Use the format Header-Name: value.+--allow-origin enables cross-origin (CORS) requests from browsers by+setting the Access-Control-Allow-Origin header on every response to the+origin given, e.g. --allow-origin https://app.example.com. Browser CORS+preflight OPTIONS requests are answered by rclone without needing+authentication as browsers never send credentials with them. Setting+this to * allows any origin which is a security risk on an authenticated+server.+--baseurl controls the URL prefix that rclone serves from. By defaultrclone will serve from the root. If you used --baseurl "/rclone" thenrclone would serve from a URL starting with "/rclone/". This is useful@@ -14625,6 +15833,34 @@ be used within the template to server pages:.Query Currently unused.+ .Path The path of the+ directory from the+ root with leading and+ trailing slashes, e.g.+ '/dir/'.++ .IsRoot Boolean for if this is+ the root directory.++ .UpLink The link to the parent+ directory.++ .Static Boolean for if the+ listing is being+ written as a static+ page by rclone index.++ .NumDirs The number of+ directories in the+ listing.++ .NumFiles The number of files in+ the listing.++ .TotalSize The total size in+ bytes of the files in+ the listing.+.Breadcrumb Allows for creating arelative navigation.@@ -14653,6 +15889,11 @@ be used within the template to server pages:.ModTime The UTC timestamp ofan entry.++ .MimeType The MIME type of an+ entry guessed from its+ name, or+ 'inode/directory'.-----------------------------------------------------------------------The server also makes the following functions available so that they can@@ -14704,8 +15945,20 @@ To create an htpasswd file:htpasswd -B htpasswd userhtpasswd -B htpasswd anotherUser+Use the same hash format and cost for every entry in the file. If the+entries differ, the time taken to reject a wrong password can reveal+which usernames exist.+The password file can be updated while rclone is running.+When authentication is enabled every request must be authenticated,+including HTTP OPTIONS requests, as the answer to an OPTIONS request can+reveal whether a path exists. Health checks or monitoring which use+unauthenticated OPTIONS requests will need to send credentials. The+exception is browser CORS preflight requests when --allow-origin is set,+which are answered without authentication as browsers never send+credentials with them.+Use --realm to set the authentication realm.Use --salt to change the password hashing salt from the default.@@ -14765,8 +16018,8 @@ at all times. The buffered data is bound to one open file and won't beshared.This flag is a upper limit for the used memory per open file. The buffer-will only use memory for data that is downloaded but not not yet read.-If the buffer is empty, only a small amount of memory will be used.+will only use memory for data that is downloaded but not yet read. If+the buffer is empty, only a small amount of memory will be used.The maximum memory used by rclone for buffering can be up to--buffer-size * open files.@@ -14815,11 +16068,12 @@ that haven't been accessed for the longest. This cache flushing strategyis efficient and more relevant files are likely to remain cached.The --vfs-cache-max-age will evict files from the cache after the set-time since last access has passed. The default value of 1 hour will-start evicting files from cache that haven't been accessed for 1 hour.-When a cached file is accessed the 1 hour timer is reset to 0 and will-wait for 1 more hour before evicting. Specify the time with standard-notation, s, m, h, d, w .+time since last access has passed; it is based on access time, not on+when the file was first added to the cache. The default value of 1 hour+will start evicting files from cache that haven't been accessed for 1+hour. When a cached file is accessed the 1 hour timer is reset to 0 and+will wait for 1 more hour before evicting. Specify the time with+standard notation, s, m, h, d, w .You should not run two copies of rclone using the same VFS cache withthe same or overlapping remotes if using --vfs-cache-mode > off. This@@ -14966,8 +16220,8 @@ Rclone reads --vfs-read-chunk-streams chunks of size--vfs-read-chunk-size concurrently. The size for each read will stayconstant.-This improves performance performance massively on high latency links or-very high bandwidth links to high performance object stores.+This improves performance massively on high latency links or very high+bandwidth links to high performance object stores.Some experimentation will be needed to find the optimum values of--vfs-read-chunk-size and --vfs-read-chunk-streams as these will depend@@ -15194,16 +16448,18 @@ This config generated must have this extra parameter- _root - root to use for the backend-And it may have this parameter+And it may have these parameters- _obscure - comma separated strings for parameters to obscure+- _secret_access_key - the secret for S3 access key auth (see below)If password authentication was used by the client, input to the proxyprocess (on STDIN) would look similar to this:{"user": "me",- "pass": "mypassword"+ "pass": "mypassword",+ "client_ip": "192.168.1.1"}If public-key authentication was used by the client, input to the proxy@@ -15211,9 +16467,41 @@ process (on STDIN) would look similar to this:{"user": "me",- "public_key": "AAAAB3NzaC1yc2EAAAADAQABAAABAQDuwESFdAe14hVS6omeyX7edc...JQdf"+ "public_key": "AAAAB3NzaC1yc2EAAAADAQABAAABAQDuwESFdAe14hVS6omeyX7edc...JQdf",+ "client_ip": "192.168.1.1"+ }++If the client authenticated with an S3 access key (rclone serve s3), the+client never sends its secret, only a signature made with it, so the+input contains just the access key ID as the user with no pass or+public_key:++ {+ "user": "AKIAIOSFODNN7EXAMPLE",+ "client_ip": "192.168.1.1"}+In this case the program must look up the secret access key for that+access key ID and return it in the _secret_access_key field of the+output. Rclone then uses that secret to verify the signature on the+request, refusing the request if it does not match. This means the proxy+program is the source of truth for both the credentials and the backend+they map to. If the program does not return _secret_access_key or+returns it empty the request is refused.++The program's answer for an access key ID is cached (see below) but is+checked with the program again after 5 minutes even if the access key ID+is in constant use, so revoking an access key ID in the program takes+effect within 5 minutes. A rotated secret takes effect on the first+request signed with it.++The client_ip key holds the IP address the client connected from,+without a port number. It can be used to restrict logins to certain+networks, or to log authentication attempts centrally. It is omitted if+the client has no IP address, for example when connecting over a unix+socket. Note that if rclone is behind a reverse proxy this will be the+address of the reverse proxy and not the original client.+And as an example return this on STDOUT{@@ -15237,11 +16525,19 @@ [email protected] and then set the host to example.com in the output andthe user to user. For security you'd probably want to restrict the hostto a limited list.-An internal cache of backends is keyed on the user and a hash of the-pass or public_key. This means that if a user's password or public-key-changes, or the proxy returns different config parameters (eg a rotated-api_key), a fresh backend will be created on the next request rather-than the cached one being reused.+An internal cache of backends is keyed on the user, the kind of+credential presented, a hash of the pass or public_key, and the+client_ip. This means that if a user's password or public-key changes,+the client connects from a new IP address, or the proxy returns+different config parameters (eg a rotated api_key), a fresh backend will+be created on the next request rather than the cached one being reused.+A backend created for a public key login is never reused for a password+login, so a public key cannot be sent as a password to skip the program.++A cached answer is checked with the program again after 5 minutes even+if the user is in constant use, so revoking a user in the program takes+effect within 5 minutes. For serve sftp this applies to new logins - an+existing connection is not closed.This can be used to build general purpose proxies to any kind of backendthat rclone supports.@@ -15577,7 +16873,7 @@ This command runs a series of uploads and downloads to the remote,measuring and printing the speed of each test using varying file sizesand numbers of files.-Test time can be innaccurate with small file caps and large files. As it+Test time can be inaccurate with small file caps and large files. As ituses the results of an initial test to determine how many files to usein each subsequent test.@@ -15634,9 +16930,11 @@ will touch --transfers files concurrently.If --timestamp is used then sets the modification time to that timeinstead of the current time. Times may be specified as one of:-- 'YYMMDD' - e.g. 17.10.30+- 'YYMMDD' - e.g. 171030+- 'YYYY-MM-DD' - e.g. 2006-01-02- 'YYYY-MM-DDTHH:MM:SS' - e.g. 2006-01-02T15:04:05- 'YYYY-MM-DDTHH:MM:SS.SSS' - e.g. 2006-01-02T15:04:05.123456789+- 'YYYY-MM-DDTHH:MM:SSZ' - e.g. 2006-01-02T15:04:05ZNote that value of --timestamp is in UTC. If you want local time thenadd the --localtime flag.@@ -16015,6 +17313,11 @@ them. This is mostly a problem on Windows, where the consoletraditionally uses a non-Unicode character set - defined by theso-called "code page".+If you want to configure a remote using environment variables then its+name becomes part of the environment variable name, which in practice+restricts the characters you can use in it. See the Config file section+under Environment variables for details.+Do not use single character names on Windows as it creates ambiguitywith Windows drives' names, e.g.: remote called C is indistinguishablefrom C drive. Rclone will always assume that single letter name refers@@ -16061,6 +17364,7 @@ Not all global variables make sense to be overridden like this as theconfig is only applied during the remote creation. Here is a nonexhaustive list of ones which might be useful:+- allow_https_downgrade- bind_addr- ca_cert- client_cert@@ -16356,6 +17660,125 @@ time of the source object.Hashes are not included in system metadata as there is a well definedway of reading those already.+Performance++Rclone's defaults are fairly conservative. They are chosen to work on+small machines, slow links and with cloud providers which rate limit+heavily. If you have plenty of bandwidth, memory and a fast storage+system then rclone can usually be made to run a great deal faster by+adjusting a few flags.++The right values depend on the size of your files, your network and the+storage systems at each end, so treat the numbers below as starting+points. Change one thing at a time and measure the result.++Measuring++Use rclone copy with -P/--progress to watch the transfer rate while you+are experimenting. For unattended runs log the stats instead, e.g.+--stats 1m --stats-one-line --stats-log-level NOTICE.++A transfer is always limited by something: the network bandwidth, the+rate at which the storage system will accept requests, disk IO, CPU or+memory. Tuning is working out which limit you are hitting and raising it+until you hit the next one. Watch the network, disk, CPU and memory use+of the machine rclone is running on as well as rclone's stats.++Lots of files++The time taken to transfer small files is dominated by the round trip+time of the requests rather than by the bandwidth, so the way to make+them go faster is to do more of them at once.++--transfers sets the number of files transferred in parallel. The+default is 4. Double this until the transfer rate stops increasing or+the storage system starts returning errors. Values of 32 to 64 are+common and values in the hundreds work well with the large object+storage systems (e.g. s3, azureblob, googlecloudstorage, b2) when+transferring lots of small files.++--checkers sets the number of directory listings and file comparisons+done in parallel. The default is 8. Raising this speeds up syncs where+most of the files are already present on the destination and stops the+checking being the bottleneck when --transfers is large. 16 to 64 is a+reasonable range.++Not all storage systems like lots of parallel requests. If you see rate+limiting errors (e.g. HTTP 429 or 503 errors) or lots of low level+retries in the log then you have gone too far, so reduce --transfers and+--checkers or use --tpslimit to cap the request rate. Local disks,+especially spinning ones, and network file systems may get slower rather+than faster with too much parallelism.++Large files++A single stream often won't fill a fast link, so rclone can transfer a+large file as several chunks in parallel. These are called multi-thread+transfers and are used for files bigger than --multi-thread-cutoff on+backends which support them.++The number of chunks transferred in parallel is set by+--multi-thread-streams (default 4) and this works for all backends so+increase this first to make large files transfer faster.++The size of the chunks is set by the backend, usually with a+--backend-chunk-size flag, e.g. --s3-chunk-size. Backends without a+natural chunk size (e.g. local, smb) use --multi-thread-chunk-size.+Larger chunks mean fewer requests which increases throughput on big+files. See the docs for the backend you are using for details.++Chunked uploads are usually buffered in memory, so in the worst case+rclone will use++ --transfers * upload concurrency * chunk size++of memory for buffers. For example --transfers 256 with an upload+concurrency of 8 and a chunk size of 16 MiB could use up to 32 GiB. When+you raise --transfers to speed up small files you may therefore need to+lower the upload concurrency, or set --max-buffer-memory which limits+the total memory used for buffers and makes transfers wait when the+limit is reached.++--use-mmap makes rclone return buffer memory to the OS as soon as it has+finished with it which is helpful when using lots of transfers or large+chunks.++A mixture of large and small files++By default rclone transfers files in roughly the order it finds them.+With a mixture of file sizes this means that at times all the transfers+are large files, which fill the bandwidth but make few requests, and at+other times all the transfers are small files, which make lots of+requests but use little of the bandwidth.++Using --order-by size,mixed,50 makes half of the transfers take the+smallest files in the backlog and half of them take the largest. This+keeps both large and small files moving at once so the bandwidth and the+request rate are both kept full for the whole of the run.++This works best if rclone has lots of files to choose from so use it+with a big --max-backlog, e.g. --max-backlog 2000000. Each item in the+backlog uses in the order of 1 KiB of memory.++Doing less work++Fewer requests to the storage system means a faster transfer.++- If the source and destination are on the same remote then rclone+ will use server-side copy where it is supported so the data doesn't+ need to be downloaded and uploaded again.+- --fast-list lists bucket-based remotes with fewer transactions at+ the cost of using more memory. This may be faster or slower than the+ default so it is worth testing.+- --no-traverse stops rclone listing the destination which helps when+ copying a small number of files into a large destination.+- --size-only or --checksum avoid reading the modification time. On+ some object storage systems (e.g. s3, swift) this needs an extra+ request for each file. Alternatively use --use-server-modtime with+ --update.+- --no-check-dest skips checking the destination entirely. Only use+ this if you know that none of the files are there already.+OptionsRclone has a number of options to control its behaviour. These are@@ -16421,7 +17844,7 @@ treated as a number of seconds. The following suffixes are valid:- M - Months- y - Years-Examples: "10", "300ms", "-1.5h" or "2h45m".+Examples: "10", "300ms", "-1.5h", "2h45m" or "1d12h".Size options@@ -16436,6 +17859,21 @@ See also --human-readable.Main options+--allow-https-downgrade++By default rclone refuses to follow a redirect from an https:// URL to a+plain http:// URL, as any credentials or other headers in the request+would then be sent unencrypted, and the response could be altered by+anyone on the network path. The request fails with an error which+mentions this flag.++Set --allow-https-downgrade to follow such redirects, for example when+downloading from a mirror which redirects to plain HTTP. Some backends+still refuse to follow them, as doing so would send their credentials in+cleartext.++This option defaults to false.+--backup-dir stringWhen using sync, copy or move, any files which would have been@@ -16556,7 +17994,7 @@ example:Is equivalent to this:- --bwlimit "Mon-00:00,512Mon-12:00,1M Tue-12:00,1M Wed-12:00,1M Thu-12:00,1M Fri-12:00,1M Sat-12:00,1M Sun-12:00,1M Sun-20:00,off"+ --bwlimit "Mon-00:00,512 Mon-12:00,1M Tue-12:00,1M Wed-12:00,1M Thu-12:00,1M Fri-12:00,1M Sat-12:00,1M Sun-12:00,1M Sun-20:00,off"Bandwidth limit apply to the data transfer for all backends. For mostbackends the directory listing bandwidth is also included (exceptions@@ -16856,7 +18294,7 @@ the dedupe command for more information as to what these options mean.--default-time Time-If a file or directory does have a modification time rclone can read+If a file or directory does not have a modification time rclone can readthen rclone will display this fixed time instead.The default is 2000-01-01 00:00:00 UTC. This can be configured in any of@@ -17010,6 +18448,12 @@ This flag is supported for all HTTP based backends even those notsupported by --header-upload and --header-download so may be used as aworkaround for those with care.+The headers are only sent to the host of the original request. If the+server redirects to another host (including a subdomain or a different+port), or redirects an https request to http, the headers are not sent+to it, or to any further hop in that redirect chain. Rclone logs this at+debug level (-vv).+rclone ls remote:test --header "X-Rclone: Foo" --header "X-LetMeIn: Yes"--header-download stringArray@@ -17186,6 +18630,11 @@ Note that on the local file system if you don't use --inplace hard links(Unix only) will be broken. And if you do use --inplace you won't beable to update in use executables.+On macOS, if rclone clones a file (see --local-no-clone) and the+destination already exists, the existing file is replaced with a new+one, even with --inplace, so hard links to it will be broken. Use+--local-no-clone to keep them.+Note also that versions of rclone prior to v1.63.0 behave as if the--inplace flag is always supplied.@@ -17235,8 +18684,8 @@ like symlinks under Windows). Ignored files won't be copied, moved ordeleted in a sync.If you supply this flag then rclone will copy symbolic links from any-supported backend backend, and store them as text files, with a-.rclonelink suffix in the destination.+supported backend, and store them as text files, with a .rclonelink+suffix in the destination.The text file will contain the target of the symbolic link.@@ -17749,8 +19198,8 @@ will use multiple threads to transfer the file (default 256M).Capable backends are marked in the overview as MultithreadUpload. (Theyneed to implement either the OpenWriterAt or OpenChunkWriter internal-interfaces). These include include, local, s3, azureblob, b2,-oracleobjectstorage and smb at the time of writing.+interfaces). These include local, s3, azureblob, b2, oracleobjectstorage+and smb at the time of writing.On the local disk, rclone preallocates the file (usingfallocate(FALLOC_FL_KEEP_SIZE) on unix or NTSetInformationFile on@@ -17772,8 +19221,8 @@ Most multi-thread transfers do not take additional memory, but some do(for example uploading to s3). In the worst case memory usage can be atmaximum --transfers * --multi-thread-chunk-size * --multi-thread-streamsor specifically for the s3 backend --transfers * --s3-chunk-size *---s3-concurrency. However you can use the the --max-buffer-memory flag-to control the maximum memory used here.+--s3-concurrency. However you can use the --max-buffer-memory flag to+control the maximum memory used here.NB that this only works with supported backends as the destination butwill work with any backend as the source.@@ -17889,8 +19338,10 @@ what aspect is being measured:- size - order by the size of the files- name - order by the full path of the files- modtime - order by the modification date of the files+- pattern - order by a list of path glob patterns-This can have a modifier appended with a comma:+The size, name and modtime comparisons can have a modifier appended with+a comma:- ascending or asc - order so that the smallest (or oldest) isprocessed first@@ -17910,11 +19361,19 @@ bandwidth and the small files will be processed continuously.If no modifier is supplied then the order is ascending.+For pattern, append one or more comma-separated patterns. Patterns use+the filter pattern syntax. Files matching the first pattern are+processed first, followed by files matching the second pattern, and so+on. Files which do not match a pattern are processed last. Within each+group, files are ordered by full path. Patterns cannot contain a comma.+For example- --order-by size,desc - send the largest files first- --order-by modtime,ascending - send the oldest files first- --order-by name - send the files with alphabetically by path first+- --order-by 'pattern,*.tar.gz,*.md5' - send archives before checksum+ filesIf the --order-by flag is not supplied or it is supplied with an emptystring then the default ordering will be used which is as scanned. With@@ -17957,13 +19416,16 @@ the config password for the first time.The argument to this should be a command with a space separated list ofarguments. If one of the arguments has a space in then enclose it in ",if you want a literal " in an argument then enclose the argument in "-and double the ". See CSV encoding for more info.+and double the ". This includes the command itself: if the path to the+executable contains a space, it must be quoted too. See CSV encoding for+more info.Eg--password-command "echo hello"--password-command 'echo "hello with space"'--password-command 'echo "hello with ""quotes"" and space"'+ --password-command '"/path with a space/get-password.sh"'Note that when changing the configuration password the environmentvariable RCLONE_PASSWORD_CHANGE=1 will be set. This can be used to@@ -18974,9 +20436,12 @@ for each backend.To find the name of the environment variable, you need to set, takeRCLONE_CONFIG_ + name of remote + _ + name of config file option and-make it all uppercase. Note one implication here is the remote's name-must be convertible into a valid environment variable name, so it can-only contain letters, digits, or the _ (underscore) character.+make it all uppercase. Note that the name of the remote is only+converted to uppercase - characters such as - and . are not changed to+_ - so a remote called my-remote needs RCLONE_CONFIG_MY-REMOTE_TYPE.+Many shells only allow letters, digits and _ (underscore) in environment+variable names, so it is best to use only those characters in the name+of a remote you want to configure this way.For example, to configure an S3 remote named mys3: without a config file(using unix ways of setting environment variables):@@ -19210,7 +20675,7 @@ Rclone matching rules follow a glob style:* matches any sequence of non-separator (/) characters** matches any sequence of characters including / separators? matches any single non-separator (/) character- [ [ ! ] { character-range } ]+ [ [ ! | ^ ] { character-range } ]character class (must be non-empty){ pattern-list }pattern alternatives@@ -20171,6 +21636,10 @@ When you run rclone gui this is what happens- The URL already contains the username & password, in which case theGUI will use those values and log you in automatically.+The rc address is passed to the GUI via the url query parameter (the URL+printed in the terminal); append ?url=http://... to point it at a+different rc server.+SecurityIt's important to think first about what rclone has access to and what@@ -20379,10 +21848,20 @@ Default Off.Set the allowed Access-Control-Allow-Origin for rc requests.-Can be used with --rc-web-gui if the rclone is running on different IP-than the web-gui.+This enables cross-origin (CORS) requests from browsers on the origin+given, for example --rc-allow-origin https://app.example.com. Browser+CORS preflight OPTIONS requests are answered by rclone without+authentication as browsers never send credentials with them. All other+requests, including other OPTIONS requests, must be authenticated when+authentication is enabled.++The rclone gui command sets this automatically to the origin the GUI is+served from if it is not set explicitly.++Setting this to * allows any origin which is a security risk on an+authenticated server.-Default is IP address on which rc is running.+Default is off (no CORS).--rc-web-fetch-url@@ -20431,6 +21910,9 @@ If this is set then no authorisation will be required on the server touse these methods. The alternative is to use --rc-user and --rc-pass anduse these credentials in the request.+Without authorisation or this flag, POST request bodies are limited to+64 KiB.+Default Off.--rc-baseurl@@ -20499,6 +21981,27 @@ Rather thanrclone rc operations/list --json '{"fs": "/tmp", "remote": "test", "opt": {"showHash": true}}'+Paths on Windows++Prefer forward slashes (/) in Windows paths passed to the rc API, for+example C:/path1/path2. If you use backslashes (\) in a JSON string,+escape each one as \\. These JSON objects contain equivalent Windows+paths:++ {"remote": "C:/path1/path2"}++ {"remote": "C:\\path1\\path2"}++This applies to JSON request bodies and to JSON-encoded object+parameters, such as parameters in config/create, even when that JSON is+sent in a query string or form data. An unescaped backslash can cause a+JSON parsing error or be interpreted as an escape sequence, such as \t+for a tab.++String parameters sent directly in a query string or form data, without+JSON encoding, do not need JSON escaping. Use your HTTP client's URL+encoding for query strings and form data.+Special parametersThe rc interface supports some special parameters which apply to all@@ -20651,6 +22154,14 @@ Or flat:If you wish to check the _config assignment has worked properly thencalling options/local will show what the value got set to.+Setting LogLevel sets the log level for the logs made by the rc call+only, for example to debug a single call.++ "_config":{"LogLevel": "DEBUG"}++Note that not all of rclone's logs support this yet. In particular the+backends log at the global log level.+Setting filter flags with _filterIf you wish to set filters for the duration of an rc call only then pass@@ -21363,6 +22874,7 @@ Returns the following values:"totalTransfers": total number of transfers in the group,"transferTime" : total time spent on running jobs,"transfers": number of transferred files,+ "updatedDirs": number of directories updated (modtime or metadata set),"transferring": an array of currently active file transfers:[{@@ -21636,8 +23148,6 @@ Results:- runningIds - array of integer job ids that are running- finishedIds - array of integer job ids that are finished-Authentication is not required for this call.-job/status: Reads the status of the job IDParameters:@@ -21662,8 +23172,6 @@ Results:synchronously- progress - output of the progress related to the underlying job-Authentication is not required for this call.-job/stop: Stop the running jobParameters:@@ -22010,6 +23518,30 @@ instead:rclone rc --loopback operations/fsinfo fs=remote:+operations/getfile: Get a single file into a JSON response.++This takes the following parameters:++- fs - a remote name string e.g. "drive:path/to/file" or+ "drive:path/to/dir"+- remote - (optional) a path within that remote e.g. "file.txt"+- offset - (optional) start reading at offset N (or from end if+ negative) (default 0)+- count - (optional) only read N bytes (default -1)+- head - (optional) only read the first N bytes (default 0)+- tail - (optional) only read the last N bytes (default 0)+- maxSize - (optional) maximum size limit in bytes (can only lower the+ built-in maximum) (default 2097152 / 2MiB)+- base64 - (optional) if true, returns the file contents+ base64-encoded (default false)++Returns:++- result - contents of the file as string or base64-encoded string++For streaming file content over HTTP with Range support, use the+--rc-serve flag instead.+operations/hashsum: Produces a hashsum file for all the objects in the path.Produces a hash file for all the objects in the path using the hash@@ -22086,6 +23618,33 @@ Example:See the hashsum command for more information on the above.+operations/index: Write static directory listings into a remote++This takes the following parameters:++- fs - a remote name string e.g. "drive:"+- opt - a dictionary of options to control the indexing (optional)+ - output - list of listings to write in each directory as+ "NAME=FORMAT" (default ["index.html=html"])+ - rules - which directories get listings, e.g. {"ExcludeRule":+ ["/private/**"]}+ - maxDepth - only make listings this many directories deep+ (default -1 for no limit)+ - linkIndex - link to "dir/NAME" rather than "dir/"+ - dirTime - how to work out directory times: "newest", "dir" or+ "none"+ - noModTime - don't show modification times+ - rewrite - write every listing even if it is unchanged+ - changed - list of changed paths for a partial run, directories+ ending in "/"+ - changedFrom - list of files of changed paths, one per line+ - changedCombined - list of files in the sync --combined report+ format+ - changedMaxDirs - do a full run if a partial one would list more+ directories than this++See the index command for more information on the above.+operations/list: List the given remote and path in JSON formatThis takes the following parameters:@@ -22310,91 +23869,6 @@ And this sets NOTICE level logs (normal without -v)rclone rc options/set --json '{"main": {"LogLevel": "NOTICE"}}'-pluginsctl/addPlugin: Add a plugin using url--Used for adding a plugin to the webgui.--This takes the following parameters:--- url - http url of the github repo where the plugin is hosted- (http://github.com/rclone/rclone-webui-react).--Example:--rclone rc pluginsctl/addPlugin--pluginsctl/getPluginsForType: Get plugins with type criteria--This shows all possible plugins by a mime type.--This takes the following parameters:--- type - supported mime type by a loaded plugin e.g. (video/mp4,- audio/mp3).-- pluginType - filter plugins based on their type e.g. (DASHBOARD,- FILE_HANDLER, TERMINAL).--Returns:--- loadedPlugins - list of current production plugins.-- testPlugins - list of temporarily loaded development plugins,- usually running on a different server.--Example:--rclone rc pluginsctl/getPluginsForType type=video/mp4--pluginsctl/listPlugins: Get the list of currently loaded plugins--This allows you to get the currently enabled plugins and their details.--This takes no parameters and returns:--- loadedPlugins - list of current production plugins.-- testPlugins - list of temporarily loaded development plugins,- usually running on a different server.--E.g.--rclone rc pluginsctl/listPlugins--pluginsctl/listTestPlugins: Show currently loaded test plugins--Allows listing of test plugins with the rclone.test set to true in-package.json of the plugin.--This takes no parameters and returns:--- loadedTestPlugins - list of currently available test plugins.--E.g.-- rclone rc pluginsctl/listTestPlugins--pluginsctl/removePlugin: Remove a loaded plugin--This allows you to remove a plugin using it's name.--This takes parameters:--- name - name of the plugin in the format author/plugin_name.--E.g.--rclone rc pluginsctl/removePlugin name=rclone/video-plugin--pluginsctl/removeTestPlugin: Remove a test plugin--This allows you to remove a plugin using it's name.--This takes the following parameters:--- name - name of the plugin in the format author/plugin_name.--Example:-- rclone rc pluginsctl/removeTestPlugin name=rclone/rclone-webui-react-rc/error: This returns an errorThis returns an error with the input as part of its error string. Useful@@ -22571,6 +24045,9 @@ This takes the following parameters:- path2 (required) - (string) a remote directory string e.g.drive:path2- dryRun - (bool) dry-run mode+- maxDelete - (int) Safety check on maximum percentage of deleted+ files allowed. If exceeded, the bisync run will abort. (default:+ 50%)- backupDir1 - (string) --backup-dir for Path1. Must be anon-overlapping path on the same remote.- backupDir2 - (string) --backup-dir for Path2. Must be a@@ -22603,6 +24080,8 @@ This takes the following parameters:- ignoreListingChecksum - (bool) Do not use checksums for listings(add --ignore- checksum to additionally skip post-copy checksumchecks)+- maxDeleteRenamesAware - (bool) Exclude tracked renames from the+ --max-delete safety check (requires --track-renames)- maxLock - (Duration) Consider lock files older than this to beexpired (default: 0 (never expire)) (minimum: 2m)- noCleanup - (bool) Retain working files (useful for troubleshooting@@ -22636,6 +24115,34 @@ This takes the following parameters:- dstFs - a remote name string e.g. "drive:dst" for the destination- createEmptySrcDirs - create empty src directories on destination ifset+- combined - make a combined report of changes (default false)+- missingOnSrc - report all files missing from the source (default+ false)+- missingOnDst - report all files missing from the destination+ (default false)+- match - report all matching files (default false)+- differ - report all non-matching files (default false)+- error - report all files with errors (hashing or reading) (default+ false)+- destAfter - report all files that exist on the destination post-copy+ (default false)++Returns:++- combined - array of strings of combined report of changes+- missingOnSrc - array of strings of all files missing from the source+- missingOnDst - array of strings of all files missing from the+ destination+- match - array of strings of all matching files+- differ - array of strings of all non-matching files+- error - array of strings of all files with errors (hashing or+ reading)+- destAfter - array of strings of all files that exist on the+ destination post-copy++Each report is only returned if its parameter is set to true. If the+operation fails the reports are only available if it was run with+_async, as part of the job output.See the copy command for more information on the above.@@ -22648,6 +24155,34 @@ This takes the following parameters:- createEmptySrcDirs - create empty src directories on destination ifset- deleteEmptySrcDirs - delete empty src directories if set+- combined - make a combined report of changes (default false)+- missingOnSrc - report all files missing from the source (default+ false)+- missingOnDst - report all files missing from the destination+ (default false)+- match - report all matching files (default false)+- differ - report all non-matching files (default false)+- error - report all files with errors (hashing or reading) (default+ false)+- destAfter - report all files that exist on the destination post-move+ (default false)++Returns:++- combined - array of strings of combined report of changes+- missingOnSrc - array of strings of all files missing from the source+- missingOnDst - array of strings of all files missing from the+ destination+- match - array of strings of all matching files+- differ - array of strings of all non-matching files+- error - array of strings of all files with errors (hashing or+ reading)+- destAfter - array of strings of all files that exist on the+ destination post-move++Each report is only returned if its parameter is set to true. If the+operation fails the reports are only available if it was run with+_async, as part of the job output.See the move command for more information on the above.@@ -22659,6 +24194,34 @@ This takes the following parameters:- dstFs - a remote name string e.g. "drive:dst" for the destination- createEmptySrcDirs - create empty src directories on destination ifset+- combined - make a combined report of changes (default false)+- missingOnSrc - report all files missing from the source (default+ false)+- missingOnDst - report all files missing from the destination+ (default false)+- match - report all matching files (default false)+- differ - report all non-matching files (default false)+- error - report all files with errors (hashing or reading) (default+ false)+- destAfter - report all files that exist on the destination post-sync+ (default false)++Returns:++- combined - array of strings of combined report of changes+- missingOnSrc - array of strings of all files missing from the source+- missingOnDst - array of strings of all files missing from the+ destination+- match - array of strings of all matching files+- differ - array of strings of all non-matching files+- error - array of strings of all files with errors (hashing or+ reading)+- destAfter - array of strings of all files that exist on the+ destination post-sync++Each report is only returned if its parameter is set to true. If the+operation fails the reports are only available if it was run with+_async, as part of the job output.See the sync command for more information on the above.@@ -22883,9 +24446,11 @@ The keys in the error response are:CORS-The sever implements basic CORS support and allows all origins for that.-The response to a preflight OPTIONS request will echo the requested-"Access-Control-Request-Headers" back.+The server implements basic CORS support when --rc-allow-origin is set.+Browser CORS preflight OPTIONS requests are then answered by rclone+without authentication with the Access-Control-* headers. Without+--rc-allow-origin no CORS headers are sent and a preflight request is+authenticated like any other request.Using POST with URL parameters only@@ -23727,6 +25292,7 @@ NetworkingFlags for general networking and HTTP stuff.+ --allow-https-downgrade Follow redirects from HTTPS to HTTP (insecure)--bind string Local address to bind to for outgoing connections, IPv4, IPv6 or name--bwlimit BwTimetable Bandwidth limit in KiB/s, or use suffix B|K|M|G|T|P or a full timetable--bwlimit-file BwTimetable Bandwidth limit per file in KiB/s, or use suffix B|K|M|G|T|P or a full timetable@@ -23750,7 +25316,7 @@ Flags for general networking and HTTP stuff.--tpslimit float Limit HTTP transactions per second to this--tpslimit-burst int Max burst of transactions for --tpslimit (default 1)--use-cookies Enable session cookiejar- --user-agent string Set the user-agent to a specified string (default "rclone/v1.75.0")+ --user-agent string Set the user-agent to a specified string (default "rclone/v1.76.0")Performance@@ -23907,11 +25473,6 @@ Flags to control the Remote Control API.--rc-template string User-specified template--rc-user string User name for authentication--rc-user-from-header string User name from a defined HTTP header- --rc-web-fetch-url string URL to fetch the releases for webgui (default "https://api.github.com/repos/rclone/rclone-webui-react/releases/latest")- --rc-web-gui Launch WebGUI on localhost- --rc-web-gui-force-update Force update to latest version of web gui- --rc-web-gui-no-open-browser Don't open the browser automatically- --rc-web-gui-update Check and update to latest version of web guiMetrics@@ -23967,8 +25528,10 @@ Backend-only flags (these can be set in the config file also).--azureblob-encoding Encoding The encoding for the backend (default Slash,BackSlash,Del,Ctl,RightPeriod,InvalidUtf8)--azureblob-endpoint string Endpoint for the service--azureblob-env-auth Read credentials from runtime (environment variables, CLI or MSI)+ --azureblob-hns Tristate Set if the storage account has a hierarchical namespace (default unset)--azureblob-key string Storage Account Shared Key--azureblob-list-chunk int Size of blob list (default 5000)+ --azureblob-list-parallelism int Number of parallel shards to list a directory with--azureblob-msi-client-id string Object ID of the user-assigned MSI to use, if any--azureblob-msi-mi-res-id string Azure resource ID of the user-assigned MSI to use, if any--azureblob-msi-object-id string Object ID of the user-assigned MSI to use, if any@@ -23981,6 +25544,7 @@ Backend-only flags (these can be set in the config file also).--azureblob-tenant string ID of the service principal's tenant. Also called its directory ID--azureblob-upload-concurrency int Concurrency for multipart uploads (default 16)--azureblob-upload-cutoff string Cutoff for switching to chunked upload (<= 256 MiB) (deprecated)+ --azureblob-use-arrow-list Use the Apache Arrow listing format--azureblob-use-az Use Azure CLI tool az for authentication--azureblob-use-copy-blob Whether to use the Copy Blob API when copying to the same storage account (default true)--azureblob-use-emulator Uses local storage emulator if provided as 'true'@@ -24113,6 +25677,12 @@ Backend-only flags (these can be set in the config file also).--doi-doi string The DOI or the doi.org URL--doi-doi-resolver-api-url string The URL of the DOI resolver API to use--doi-provider string DOI provider+ --dosya-api-key string Your API Key, get it from https://dosya.dev/profile under "API keys"+ --dosya-api-url string API base URL (default "https://api.dosya.dev")+ --dosya-description string Description of the remote+ --dosya-encoding Encoding The encoding for the backend (default Slash,LtGt,DoubleQuote,SingleQuote,BackSlash,Del,Ctl,LeftSpace,RightSpace,InvalidUtf8,Dot)+ --dosya-upload-concurrency int Concurrency for multipart uploads (default 4)+ --dosya-workspace-id string Your workspace ID--drime-access-token string API Access token--drime-chunk-size SizeSuffix Chunk size to use for uploading (default 5Mi)--drime-description string Description of the remote@@ -24335,6 +25905,8 @@ Backend-only flags (these can be set in the config file also).--http-headers CommaSepList Set HTTP headers for all transactions--http-no-escape Do not escape URL metacharacters in path names--http-no-head Don't use HEAD requests+ --http-no-head-size SizeSuffix Set the reported file size for --http-no-head (default off)+ --http-no-head-time Time Set a default modification time for --http-no-head (default off)--http-no-slash Set this if the site doesn't end directories with /--http-url string URL of HTTP host to connect to--huaweidrive-auth-url string Auth server URL@@ -24394,6 +25966,17 @@ Backend-only flags (these can be set in the config file also).--jottacloud-token-url string Token server url--jottacloud-trashed-only Only show files that are in the trash--jottacloud-upload-resume-limit SizeSuffix Files bigger than this can be resumed if the upload fail's (default 10Mi)+ --kdrive-auth-url string Auth server URL+ --kdrive-chunk-size SizeSuffix Files above this size are uploaded in chunks of this size (default 20Mi)+ --kdrive-client-id string OAuth Client Id+ --kdrive-description string Description of the remote+ --kdrive-drive-id string ID of the kDrive to use+ --kdrive-encoding Encoding The encoding for the backend (default Slash,Del,Ctl,LeftSpace,RightSpace,InvalidUtf8,Dot)+ --kdrive-endpoint string The API endpoint to use+ --kdrive-root-folder-id string The folder to use as root+ --kdrive-token string OAuth Access Token as a JSON blob+ --kdrive-token-url string Token server url+ --kdrive-upload-concurrency int Concurrency for chunked uploads (default 4)--koofr-description string Description of the remote--koofr-encoding Encoding The encoding for the backend (default Slash,BackSlash,Del,Ctl,InvalidUtf8,Dot)--koofr-endpoint string The Koofr API endpoint to use@@ -24479,6 +26062,7 @@ Backend-only flags (these can be set in the config file also).--onedrive-root-folder-id string ID of the root folder--onedrive-server-side-across-configs Deprecated: use --server-side-across-configs instead--onedrive-tenant string ID of the service principal's tenant. Also called its directory ID+ --onedrive-tenant-api-version string The SharePoint API version to use with tenant_url (default "v2.0")--onedrive-tenant-url string The tenant URL for non-admin OneDrive access--onedrive-token string OAuth Access Token as a JSON blob--onedrive-token-url string Token server url@@ -24493,7 +26077,7 @@ Backend-only flags (these can be set in the config file also).--oos-decompress If set this will decompress gzip encoded objects--oos-description string Description of the remote--oos-disable-checksum Don't store MD5 checksum with object metadata- --oos-encoding Encoding The encoding for the backend (default Slash,InvalidUtf8,Dot)+ --oos-encoding Encoding The encoding for the backend (default Slash,CrLf,InvalidUtf8,Dot)--oos-endpoint string Endpoint for Object storage API--oos-leave-parts-on-error If true avoid calling abort upload on a failure, leaving all successfully uploaded parts for manual recovery--oos-max-upload-parts int Maximum number of parts in a multipart upload (default 10000)@@ -24688,7 +26272,7 @@ Backend-only flags (these can be set in the config file also).--sftp-disable-concurrent-reads If set don't use concurrent reads--sftp-disable-concurrent-writes If set don't use concurrent writes--sftp-disable-hashcheck Disable the execution of SSH commands to determine if remote file hashing is available- --sftp-encoding Encoding The encoding for the backend (default Slash,Del,Ctl,Dot)+ --sftp-encoding Encoding The encoding for the backend (default Slash,Dot)--sftp-hashes CommaSepList Comma separated list of supported checksum types--sftp-host string SSH host to connect to--sftp-host-key-algorithms SpaceSepList Space separated list of host key algorithms, ordered by preference@@ -24703,6 +26287,7 @@ Backend-only flags (these can be set in the config file also).--sftp-known-hosts-file string Optional path to known_hosts file--sftp-macs SpaceSepList Space separated list of MACs (message authentication code) algorithms, ordered by preference--sftp-md5sum-command string The command used to read MD5 hashes+ --sftp-multithread-upload Set this to use multi-thread (multi-connection) uploads--sftp-pass string SSH password, leave blank to use ssh-agent (obscured)--sftp-path-override string Override path used by SSH shell commands--sftp-pin-host-key Pin the server host key on first connection (Trust On First Use)@@ -24765,7 +26350,8 @@ Backend-only flags (these can be set in the config file also).--smb-port int SMB port number (default 445)--smb-spn string Service principal name--smb-use-kerberos Use Kerberos authentication- --smb-user string SMB username (default "$USER")+ --smb-user string SMB username+ --smb-workstation string Workstation name sent in NTLM authentication--storj-access-grant string Access grant--storj-api-key string API key--storj-description string Description of the remote@@ -24840,6 +26426,7 @@ Backend-only flags (these can be set in the config file also).--webdav-url string URL of http host to connect to--webdav-user string User name--webdav-vendor string Name of the WebDAV site/service/software you are using+ --yandex-app-folder Use the application folder as the root--yandex-auth-url string Auth server URL--yandex-client-credentials Use client credentials OAuth flow--yandex-client-id string OAuth Client Id@@ -25494,6 +27081,7 @@ Command line syntax--force Bypass --max-delete safety check and run the sync. Consider using with --verbose-h, --help help for bisync--ignore-listing-checksum Do not use checksums for listings (add --ignore-checksum to additionally skip post-copy checksum checks)+ --max-delete-renames-aware Exclude tracked renames from the --max-delete safety check (requires --track-renames)--max-lock Duration Consider lock files older than this to be expired (default: 0 (never expire)) (minimum: 2m) (default 0s)--no-cleanup Retain working files (useful for troubleshooting and testing).--no-slow-hash Ignore listing checksums only on backends where they are slow@@ -25871,6 +27459,25 @@ the check.Also see the all files changed check.+--max-delete-renames-aware++When --track-renames is in use, --max-delete-renames-aware can exclude+files that are guaranteed to match as tracked renames from the deletion+count. If the initial deletion count exceeds --max-delete, bisync+performs a rename matching preflight before starting either sync+direction. The run still aborts if the remaining unmatched deletions+exceed the limit.++The preflight uses the selected --track-renames-strategy and the+attributes stored in the normal bisync listings. A hash strategy forces+both listings to use a common hash and is incompatible with+--ignore-listing-checksum. The option only exempts matches when the+destination supports server-side moves or copies and the selected+strategy is supported by both paths. It requires --track-renames and is+disabled by default because hashing may add listing time. --force+bypasses both the safety check and this preflight. The preflight is also+disabled during --resync, when rename tracking is unavailable.+--filters-fileBy using rclone filter features you can exclude file types or directory@@ -26472,14 +28079,20 @@ The following backends have known issues that need more investigation:- TestHuaweiDrive (huaweidrive)- TestBisyncRemoteLocal/ext_paths- TestBisyncRemoteLocal/extended_filenames- - TestBisyncRemoteLocal/normalization- TestBisyncLocalRemote/ext_paths- TestBisyncLocalRemote/extended_filenames- - 4 more+ - TestBisyncLocalRemote/normalization+ - 3 more- TestPcloud (pcloud)- - TestBisyncRemoteRemote/check_access- - TestBisyncRemoteRemote/rmdirs-- Updated: 2026-07-31-010017+ - TestBisyncRemoteRemote/extended_filenames+- TestYandex (yandex)+ - TestBisyncRemoteLocal/all_changed+ - TestBisyncRemoteLocal/backupdir+ - TestBisyncRemoteLocal/basic+ - TestBisyncRemoteLocal/changes+ - TestBisyncRemoteLocal/check_access+ - 94 more+- Updated: 2026-11-06-010024The following backends either have not been tested recently or haveknown issues that are deemed unfixable for the time being:@@ -26495,6 +28108,8 @@ known issues that are deemed unfixable for the time being:- TestFilesCom (filescom)- TestImageKit (imagekit)- TestJottacloud (jottacloud)+- TestKdrive (kdrive)+ - TestBisyncRemoteRemote/normalization- TestLinkbox (linkbox)- TestMailru (mailru)- TestMega (mega)@@ -26507,7 +28122,8 @@ known issues that are deemed unfixable for the time being:- TestQuatrix (quatrix)- TestS3GCS (s3)- TestBisyncRemoteRemote/extended_filenames-- TestS3Rclone (s3)+ - TestBisyncRemoteLocal/extended_filenames+ - TestBisyncLocalRemote/extended_filenames- TestSFTPRsyncNet (sftp)- TestStorj (storj)- TestWebdavNextcloud (webdav)@@ -26585,10 +28201,11 @@ directory is to rename it to the same name on both sides. (As ofrclone v1.64, a --resync is no longer required after doing so, as bisyncwill automatically detect that Path1 and Path2 are in agreement.)-Note that although the flag --track-renames ensures that renamed/moved-files won't be deleted and uploaded again, they are still counted as-deleted files for purposes of the --max-delete flag (as this check-happens before the rename detection operation). See this issue.+By default, renamed or moved files are still counted as deleted files+for purposes of --max-delete, because the safety check happens before+rename detection. Use --max-delete-renames-aware together with+--track-renames to run a rename preflight and exclude guaranteed matches+from this count.--fast-list used by default@@ -27328,136 +28945,8 @@ Unison and synchronization in general.Changelog-v1.74.2--- Fixed an issue causing --conflict-loser pathname to produce- unexpected behavior if using a non-default --conflict-resolve value.--v1.74--- Added several missing rc parameters.-- Optional rc parameters are now truly optional.-- rc output now provides more structured information.--v1.71--- bisync is now officially released from beta.--- Fixed an issue causing hash type to be set incorrectly if path2 had- slow hashes and --no-slow-hash or --slow-hash-sync-only was in use.--- Fixed an issue causing errors when running concurrent bisync runs- through the rc.--v1.69.1--- Fixed an issue causing listings to not capture concurrent- modifications under certain conditions--v1.68--- Fixed an issue affecting backends that round modtimes to a lower- precision.--v1.67--- Added integration tests against all backends.--v1.66--- Copies and deletes are now handled in one operation instead of two--- --track-renames and --backup-dir are now supported--- Partial uploads known issue on local/ftp/sftp has been resolved- (unless using --inplace)--- Final listings are now generated from sync results, to avoid needing- to re-list--- Bisync is now much more resilient to changes that happen during a- bisync run, and far less prone to critical errors / undetected- changes--- Bisync is now capable of rolling a file listing back in cases of- uncertainty, essentially marking the file as needing to be rechecked- next time.--- A few basic terminal colors are now supported, controllable with- --color (AUTO|NEVER|ALWAYS)--- Initial listing snapshots of Path1 and Path2 are now generated- concurrently, using the same "march" infrastructure as check and- sync, for performance improvements and less risk of error.--- Fixed handling of unicode normalization and case insensitivity,- support for --fix-case, --ignore-case-sync,- --no-unicode-normalization--- --resync is now much more efficient (especially for users of- --create-empty-src-dirs)--- Google Docs (and other files of unknown size) are now supported- (with the same options as in sync)--- Equality checks before a sync conflict rename now fall back to- cryptcheck (when possible) or --download, instead of of --size-only,- when check is not available.--- Bisync no longer fails to find the correct listing file when configs- are overridden with backend-specific flags.--- Bisync now fully supports comparing based on any combination of- size, modtime, and checksum, lifting the prior restriction on- backends without modtime support.--- Bisync now supports a "Graceful Shutdown" mode to cleanly cancel a- run early without requiring --resync.--- New --recover flag allows robust recovery in the event of- interruptions, without requiring --resync.--- A new --max-lock setting allows lock files to automatically renew- and expire, for better automatic recovery when a run is interrupted.--- Bisync now supports auto-resolving sync conflicts and customizing- rename behavior with new --conflict-resolve, --conflict-loser, and- --conflict-suffix flags.--- A new --resync-mode flag allows more control over which version of a- file gets kept during a --resync.--- Bisync now supports --retries and --retries-sleep (when --resilient- is set.)--v1.64--- Fixed an issue causing dry runs to inadvertently commit filter- changes--- Fixed an issue causing --resync to erroneously delete empty folders- and duplicate files unique to Path2--- --check-access is now enforced during --resync, preventing data loss- in certain user error scenarios--- Fixed an issue causing bisync to consider more files than necessary- due to overbroad filters during delete operations--- Improved detection of false positive change conflicts (identical- files are now left alone instead of renamed)--- Added support for --create-empty-src-dirs--- Added experimental --resilient mode to allow recovery from- self-correctable errors--- Added new --ignore-listing-checksum flag to distinguish from- --ignore-checksum--- Performance improvements for large remotes--- Documentation and testing improvements+Changes to bisync are listed in the rclone changelog. The changelog this+page kept until v1.76 is archived on GitHub.Release signing@@ -27574,16 +29063,18 @@ Verify the hashesNow that we know the signatures on the hashes are OK we can verify thebinaries match the hashes, completing the verification.- $ sha256sum -c SHA256SUMS 2>&1 | grep OK+Extract the signed data with:++ $ gpg --output SHA256SUMS.signed --decrypt SHA256SUMS++Then check the binaries:++ $ sha256sum -c --ignore-missing SHA256SUMS.signedrclone-v1.63.1-windows-amd64.zip: OKOr do the check with rclone- $ rclone hashsum sha256 -C SHA256SUMS rclone-v1.63.1-windows-amd64.zip- 2023/09/11 10:53:58 NOTICE: SHA256SUMS: improperly formatted checksum line 0- 2023/09/11 10:53:58 NOTICE: SHA256SUMS: improperly formatted checksum line 1- 2023/09/11 10:53:58 NOTICE: SHA256SUMS: improperly formatted checksum line 49- 2023/09/11 10:53:58 NOTICE: SHA256SUMS: 4 warning(s) suppressed...+ $ rclone hashsum sha256 -C SHA256SUMS.signed rclone-v1.63.1-windows-amd64.zip= rclone-v1.63.1-windows-amd64.zip2023/09/11 10:53:58 NOTICE: Local file system at /tmp/check: 0 differences found2023/09/11 10:53:58 NOTICE: Local file system at /tmp/check: 1 matching files@@ -27943,8 +29434,10 @@ The S3 backend can be used with a number of different providers:- Alibaba Cloud (Aliyun) Object Storage System (OSS)- Ceph- China Mobile Ecloud Elastic Object Storage (EOS)+- Clever Cloud Cellar- Cloudflare R2- Arvan Cloud Object Storage (AOS)+- Autonomys Auto Drive- Bizfly Cloud Simple Storage- Cubbit DS3- DigitalOcean Spaces@@ -27959,6 +29452,7 @@ The S3 backend can be used with a number of different providers:- IBM COS S3- IDrive e2- Impossible Cloud+- Infomaniak Swiss Backup- Intercolo Object Storage- IONOS Cloud- Leviia Object Storage@@ -27967,6 +29461,7 @@ The S3 backend can be used with a number of different providers:- Magalu Object Storage- MEGA S4 Object Storage- Minio+- Neon Object Storage- Outscale- OVHcloud- Petabox@@ -27975,6 +29470,7 @@ The S3 backend can be used with a number of different providers:- Rabata Cloud Storage- RackCorp Object Storage- Rclone Serve S3+- RelAix.Storage- Scaleway- Scality (RING / ARTESCA)- Seagate Lyve Cloud@@ -28800,6 +30296,9 @@ glacier storage class you will see an error like below.2017/09/11 19:07:43 Failed to sync: failed to open source object: Object in GLACIER, restore first: path/to/file+The message shows the object's actual storage class, for example+DEEP_ARCHIVE.+In this case you need to restore the object(s) in question beforeaccessing object contents. The restore section below shows how to dothis with rclone.@@ -28868,14 +30367,15 @@ Additional Object Lock optionsStandard optionsHere are the Standard options specific to s3 (Amazon S3 Compliant-Storage Providers including AWS, Alibaba, ArvanCloud, BizflyCloud, Ceph,-ChinaMobile, Cloudflare, Cubbit, DigitalOcean, Dreamhost, Exaba, Fastly,-FileLu, FlashBlade, GCS, HCP, Hetzner, HuaweiOBS, IBMCOS, IDrive,-ImpossibleCloud, Intercolo, IONOS, Leviia, Liara, Linode, LyveCloud,-Magalu, Mega, Minio, Netease, Outscale, OVHcloud, Petabox, Qiniu,-Rabata, RackCorp, Rclone, Scaleway, Scality, SeaweedFS, Selectel,-Servercore, SpectraLogic, Storj, Synology, TencentCOS, US3, Wasabi,-Zadara, Zata, ZeroServices, Other).+Storage Providers including AWS, Alibaba, ArvanCloud, Autonomys,+BizflyCloud, Cellar, Ceph, ChinaMobile, Cloudflare, Cubbit,+DigitalOcean, Dreamhost, Exaba, Fastly, FileLu, FlashBlade, GCS, HCP,+Hetzner, HuaweiOBS, IBMCOS, IDrive, ImpossibleCloud, Infomaniak,+Intercolo, IONOS, Leviia, Liara, Linode, LyveCloud, Magalu, Mega, Minio,+Neon, Netease, Outscale, OVHcloud, Petabox, Qiniu, Rabata, RackCorp,+Rclone, RelAix, Scaleway, Scality, SeaweedFS, Selectel, Servercore,+SpectraLogic, Storj, Synology, TencentCOS, US3, Wasabi, Zadara, Zata,+ZeroServices, Other).--s3-provider@@ -28894,8 +30394,12 @@ Properties:- Alibaba Cloud Object Storage System (OSS) formerly Aliyun- "ArvanCloud"- Arvan Cloud Object Storage (AOS)+ - "Autonomys"+ - Autonomys Auto Drive- "BizflyCloud"- Bizfly Cloud Simple Storage+ - "Cellar"+ - Clever Cloud Cellar Object Storage- "Ceph"- Ceph Object Storage- "ChinaMobile"@@ -28930,6 +30434,8 @@ Properties:- IDrive e2- "ImpossibleCloud"- Impossible Cloud Object Storage+ - "Infomaniak"+ - Infomaniak Swiss Backup- "Intercolo"- Intercolo Object Storage- "IONOS"@@ -28948,6 +30454,8 @@ Properties:- MEGA S4 Object Storage- "Minio"- Minio Object Storage+ - "Neon"+ - Neon Object Storage- "Netease"- Netease Object Storage (NOS)- "Outscale"@@ -28964,6 +30472,8 @@ Properties:- RackCorp Object Storage- "Rclone"- Rclone S3 Server+ - "RelAix"+ - RelAix.Storage- "Scaleway"- Scaleway Object Storage- "Scality"@@ -29051,7 +30561,7 @@ Properties:- Config: region- Env Var: RCLONE_S3_REGION- Provider:- AWS,BizflyCloud,Ceph,Cloudflare,Cubbit,DigitalOcean,Dreamhost,Exaba,Fastly,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IBMCOS,ImpossibleCloud,Intercolo,IONOS,Leviia,LyveCloud,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,Rabata,RackCorp,Scaleway,Scality,SeaweedFS,Selectel,Servercore,Synology,Wasabi,Zadara,Zata,ZeroServices,Other+ AWS,BizflyCloud,Ceph,Cloudflare,Cubbit,DigitalOcean,Dreamhost,Exaba,Fastly,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IBMCOS,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,LyveCloud,Minio,Neon,Netease,Outscale,OVHcloud,Petabox,Qiniu,Rabata,RackCorp,Scaleway,Scality,SeaweedFS,Selectel,Servercore,Synology,Wasabi,Zadara,Zata,ZeroServices,Other- Type: string- Required: false- Examples:@@ -29189,6 +30699,9 @@ Properties:- "eu-central"- EU Central- Provider: Fastly+ - "eu-central-1"+ - EU Central 1+ - Provider: Fastly- "eu-south-1"- EU South 1- Provider: Fastly@@ -29240,50 +30753,104 @@ Properties:- "nbg1"- Nuremberg- Provider: Hetzner+ - "af-north-1"+ - AF-Cairo+ - Provider: HuaweiOBS- "af-south-1"- AF-Johannesburg- Provider: HuaweiOBS+ - "ap-southeast-1"+ - CN-Hong Kong+ - Provider: HuaweiOBS- "ap-southeast-2"- AP-Bangkok- Provider: HuaweiOBS- "ap-southeast-3"- AP-Singapore- Provider: HuaweiOBS- - "cn-east-3"- - CN East-Shanghai1+ - "ap-southeast-4"+ - AP-Jakarta+ - Provider: HuaweiOBS+ - "ap-southeast-5"+ - AP-Manila- Provider: HuaweiOBS- "cn-east-2"- CN East-Shanghai2- Provider: HuaweiOBS+ - "cn-east-3"+ - CN East-Shanghai1+ - Provider: HuaweiOBS+ - "cn-east-4"+ - CN East-Qingdao+ - Provider: HuaweiOBS+ - "cn-east-5"+ - CN East-Suzhou+ - Provider: HuaweiOBS- "cn-north-1"- CN North-Beijing1- Provider: HuaweiOBS+ - "cn-north-2"+ - CN North-Beijing2+ - Provider: HuaweiOBS- "cn-north-4"- CN North-Beijing4- Provider: HuaweiOBS+ - "cn-north-9"+ - CN North-Ulanqab+ - Provider: HuaweiOBS+ - "cn-north-11"+ - CN North-Hebei+ - Provider: HuaweiOBS+ - "cn-north-12"+ - CN North-Hohhot+ - Provider: HuaweiOBS- "cn-south-1"- CN South-Guangzhou- Provider: HuaweiOBS- - "ap-southeast-1"- - CN-Hong Kong+ - "cn-south-2"+ - CN South-Shenzhen- Provider: HuaweiOBS- - "sa-argentina-1"- - LA-Buenos Aires1+ - "cn-south-4"+ - CN South-Guangzhou-InvitationOnly- Provider: HuaweiOBS- - "sa-peru-1"- - LA-Lima1+ - "cn-southwest-2"+ - CN Southwest-Guiyang1+ - Provider: HuaweiOBS+ - "cn-southwest-3"+ - CN Southwest-Chongqing+ - Provider: HuaweiOBS+ - "eu-west-101"+ - EU-Dublin+ - Provider: HuaweiOBS+ - "la-north-2"+ - LA-Mexico City2+ - Provider: HuaweiOBS+ - "la-south-2"+ - LA-Santiago+ - Provider: HuaweiOBS+ - "me-east-1"+ - ME-Riyadh- Provider: HuaweiOBS- "na-mexico-1"- LA-Mexico City1- Provider: HuaweiOBS- - "sa-chile-1"- - LA-Santiago2+ - "ru-northwest-2"+ - RU-Moscow2+ - Provider: HuaweiOBS+ - "sa-argentina-1"+ - LA-Buenos Aires1- Provider: HuaweiOBS- "sa-brazil-1"- LA-Sao Paulo1- Provider: HuaweiOBS- - "ru-northwest-2"- - RU-Moscow2+ - "sa-chile-1"+ - LA-Santiago2+ - Provider: HuaweiOBS+ - "sa-peru-1"+ - LA-Lima1+ - Provider: HuaweiOBS+ - "tr-west-1"+ - TR-Istanbul- Provider: HuaweiOBS- "eu-central-2"- Frankfurt, Germany@@ -29306,6 +30873,10 @@ Properties:- "us-east-1"- New York, USA- Provider: ImpossibleCloud+ - "us-east-1"+ - Default region for compatibility.+ - All data is stored in Switzerland.+ - Provider: Infomaniak- "de-fra"- Frankfurt, Germany- Provider: Intercolo@@ -29327,6 +30898,18 @@ Properties:- "us-central-1"- Lenexa, USA- Provider: IONOS+ - "us-east-1"+ - US East (N. Virginia)+ - Provider: Neon,Petabox,Rabata+ - "us-east-2"+ - US East (Ohio)+ - Provider: Neon+ - "eu-central-1"+ - Europe (Frankfurt)+ - Provider: Neon,Petabox+ - "ap-southeast-1"+ - Asia Pacific (Singapore)+ - Provider: Neon,Petabox- "eu-west-2"- Paris, France- Provider: Outscale@@ -29384,15 +30967,6 @@ Properties:- "rbx-archive"- Roubaix, France (Cold Archive)- Provider: OVHcloud- - "us-east-1"- - US East (N. Virginia)- - Provider: Petabox,Rabata- - "eu-central-1"- - Europe (Frankfurt)- - Provider: Petabox- - "ap-southeast-1"- - Asia Pacific (Singapore)- - Provider: Petabox- "me-south-1"- Middle East (Bahrain)- Provider: Petabox@@ -29500,6 +31074,9 @@ Properties:- "pl-waw"- Warsaw, Poland- Provider: Scaleway+ - "it-mil"+ - Milan, Italy+ - Provider: Scaleway- "ru-1"- St. Petersburg- Provider: Selectel,Servercore@@ -29567,7 +31144,7 @@ Properties:- Config: endpoint- Env Var: RCLONE_S3_ENDPOINT- Provider:- AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,ChinaMobile,Cloudflare,Cubbit,DigitalOcean,Dreamhost,Exaba,Fastly,FileLu,FlashBlade,GCS,HCP,Hetzner,HuaweiOBS,IBMCOS,ImpossibleCloud,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Mega,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,Rabata,RackCorp,Rclone,Scaleway,Scality,SeaweedFS,Selectel,Servercore,SpectraLogic,Storj,Synology,TencentCOS,US3,Wasabi,Zadara,Zata,ZeroServices,Other+ AWS,Alibaba,ArvanCloud,Autonomys,BizflyCloud,Cellar,Ceph,ChinaMobile,Cloudflare,Cubbit,DigitalOcean,Dreamhost,Exaba,Fastly,FileLu,FlashBlade,GCS,HCP,Hetzner,HuaweiOBS,IBMCOS,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Mega,Minio,Neon,Netease,Outscale,OVHcloud,Petabox,Qiniu,Rabata,RackCorp,Rclone,RelAix,Scaleway,Scality,SeaweedFS,Selectel,Servercore,SpectraLogic,Storj,Synology,TencentCOS,US3,Wasabi,Zadara,Zata,ZeroServices,Other- Type: string- Required: false- Examples:@@ -29653,12 +31230,18 @@ Properties:- "s3.ir-tbz-sh1.arvanstorage.ir"- Tabriz Iran (Shahriar)- Provider: ArvanCloud+ - "s3.auto-drive.autonomys.xyz"+ - Auto Drive mainnet S3 endpoint+ - Provider: Autonomys- "hn.ss.bfcplatform.vn"- Hanoi endpoint- Provider: BizflyCloud- "hcm.ss.bfcplatform.vn"- Ho Chi Minh endpoint- Provider: BizflyCloud+ - "cellar-c2.services.clever-cloud.com"+ - Paris, France+ - Provider: Cellar- "eos-wuxi-1.cmecloud.cn"- The default endpoint - a good choice if you are unsure.- East China (Suzhou)@@ -29796,6 +31379,9 @@ Properties:- "eu-central.object.fastlystorage.app"- EU Central- Provider: Fastly+ - "eu-central-1.object.fastlystorage.app"+ - EU Central 1+ - Provider: Fastly- "eu-south-1.object.fastlystorage.app"- EU South 1- Provider: Fastly@@ -29850,50 +31436,104 @@ Properties:- "nbg1.your-objectstorage.com"- Nuremberg- Provider: Hetzner+ - "obs.af-north-1.myhuaweicloud.com"+ - AF-Cairo+ - Provider: HuaweiOBS- "obs.af-south-1.myhuaweicloud.com"- AF-Johannesburg- Provider: HuaweiOBS+ - "obs.ap-southeast-1.myhuaweicloud.com"+ - CN-Hong Kong+ - Provider: HuaweiOBS- "obs.ap-southeast-2.myhuaweicloud.com"- AP-Bangkok- Provider: HuaweiOBS- "obs.ap-southeast-3.myhuaweicloud.com"- AP-Singapore- Provider: HuaweiOBS- - "obs.cn-east-3.myhuaweicloud.com"- - CN East-Shanghai1+ - "obs.ap-southeast-4.myhuaweicloud.com"+ - AP-Jakarta+ - Provider: HuaweiOBS+ - "obs.ap-southeast-5.myhuaweicloud.com"+ - AP-Manila- Provider: HuaweiOBS- "obs.cn-east-2.myhuaweicloud.com"- CN East-Shanghai2- Provider: HuaweiOBS+ - "obs.cn-east-3.myhuaweicloud.com"+ - CN East-Shanghai1+ - Provider: HuaweiOBS+ - "obs.cn-east-4.myhuaweicloud.com"+ - CN East-Qingdao+ - Provider: HuaweiOBS+ - "obs.cn-east-5.myhuaweicloud.com"+ - CN East-Suzhou+ - Provider: HuaweiOBS- "obs.cn-north-1.myhuaweicloud.com"- CN North-Beijing1- Provider: HuaweiOBS+ - "obs.cn-north-2.myhuaweicloud.com"+ - CN North-Beijing2+ - Provider: HuaweiOBS- "obs.cn-north-4.myhuaweicloud.com"- CN North-Beijing4- Provider: HuaweiOBS+ - "obs.cn-north-9.myhuaweicloud.com"+ - CN North-Ulanqab+ - Provider: HuaweiOBS+ - "obs.cn-north-11.myhuaweicloud.com"+ - CN North-Hebei+ - Provider: HuaweiOBS+ - "obs.cn-north-12.myhuaweicloud.com"+ - CN North-Hohhot+ - Provider: HuaweiOBS- "obs.cn-south-1.myhuaweicloud.com"- CN South-Guangzhou- Provider: HuaweiOBS- - "obs.ap-southeast-1.myhuaweicloud.com"- - CN-Hong Kong+ - "obs.cn-south-2.myhuaweicloud.com"+ - CN South-Shenzhen- Provider: HuaweiOBS- - "obs.sa-argentina-1.myhuaweicloud.com"- - LA-Buenos Aires1+ - "obs.cn-south-4.myhuaweicloud.com"+ - CN South-Guangzhou-InvitationOnly- Provider: HuaweiOBS- - "obs.sa-peru-1.myhuaweicloud.com"- - LA-Lima1+ - "obs.cn-southwest-2.myhuaweicloud.com"+ - CN Southwest-Guiyang1+ - Provider: HuaweiOBS+ - "obs.cn-southwest-3.myhuaweicloud.com"+ - CN Southwest-Chongqing+ - Provider: HuaweiOBS+ - "obs.eu-west-101.myhuaweicloud.eu"+ - EU-Dublin+ - Provider: HuaweiOBS+ - "obs.la-north-2.myhuaweicloud.com"+ - LA-Mexico City2+ - Provider: HuaweiOBS+ - "obs.la-south-2.myhuaweicloud.com"+ - LA-Santiago+ - Provider: HuaweiOBS+ - "obs.me-east-1.myhuaweicloud.com"+ - ME-Riyadh- Provider: HuaweiOBS- "obs.na-mexico-1.myhuaweicloud.com"- LA-Mexico City1- Provider: HuaweiOBS- - "obs.sa-chile-1.myhuaweicloud.com"- - LA-Santiago2+ - "obs.ru-northwest-2.myhuaweicloud.com"+ - RU-Moscow2+ - Provider: HuaweiOBS+ - "obs.sa-argentina-1.myhuaweicloud.com"+ - LA-Buenos Aires1- Provider: HuaweiOBS- "obs.sa-brazil-1.myhuaweicloud.com"- LA-Sao Paulo1- Provider: HuaweiOBS- - "obs.ru-northwest-2.myhuaweicloud.com"- - RU-Moscow2+ - "obs.sa-chile-1.myhuaweicloud.com"+ - LA-Santiago2+ - Provider: HuaweiOBS+ - "obs.sa-peru-1.myhuaweicloud.com"+ - LA-Lima1+ - Provider: HuaweiOBS+ - "obs.tr-west-1.myhuaweicloud.com"+ - TR-Istanbul- Provider: HuaweiOBS- "s3.us.cloud-object-storage.appdomain.cloud"- US Cross Region Endpoint@@ -30102,6 +31742,18 @@ Properties:- "us-east-1.storage.impossibleapi.net"- New York, USA- Provider: ImpossibleCloud+ - "s3.swiss-backup01.infomaniak.com"+ - Swiss Backup Location 01+ - Provider: Infomaniak+ - "s3.swiss-backup02.infomaniak.com"+ - Swiss Backup Location 02+ - Provider: Infomaniak+ - "s3.swiss-backup03.infomaniak.com"+ - Swiss Backup Location 03+ - Provider: Infomaniak+ - "s3.swiss-backup04.infomaniak.com"+ - Swiss Backup Location 04+ - Provider: Infomaniak- "de-fra.i3storage.com"- Frankfurt, Germany- Provider: Intercolo@@ -30206,38 +31858,47 @@ Properties:- "br-ne1.magaluobjects.com"- Fortaleza, CE (BR), br-ne1- Provider: Magalu- - "s3.eu-amsterdam.megas4.com"- - Mega S4 Amsterdam+ - "s3.eu-luxembourg-1.megas4.com"+ - Mega S4 Luxembourg 1+ - Provider: Mega+ - "s3.eu-luxembourg-2.megas4.com"+ - Mega S4 Luxembourg 2+ - Provider: Mega+ - "s3.eu-amsterdam-1.megas4.com"+ - Mega S4 Amsterdam 1- Provider: Mega- - "s3.eu-luxembourg.megas4.com"- - Mega S4 Luxembourg+ - "s3.eu-amsterdam-2.megas4.com"+ - Mega S4 Amsterdam 2- Provider: Mega- - "s3.eu-paris.megas4.com"- - Mega S4 Paris+ - "s3.eu-paris-1.megas4.com"+ - Mega S4 Paris 1- Provider: Mega- - "s3.eu-barcelona.megas4.com"- - Mega S4 Barcelona+ - "s3.eu-paris-2.megas4.com"+ - Mega S4 Paris 2- Provider: Mega- - "s3.ca-montreal.megas4.com"- - Mega S4 Montreal+ - "s3.eu-barcelona-1.megas4.com"+ - Mega S4 Barcelona 1- Provider: Mega- - "s3.ca-vancouver.megas4.com"- - Mega S4 Vancouver+ - "s3.eu-barcelona-2.megas4.com"+ - Mega S4 Barcelona 2- Provider: Mega- - "s3.ap-tokyo.megas4.com"- - Mega S4 Tokyo+ - "s3.ca-montreal-1.megas4.com"+ - Mega S4 Montreal 1- Provider: Mega- - "s3.eu-central-1.s4.mega.io"- - Mega S4 eu-central-1 (Amsterdam, legacy)+ - "s3.ca-montreal-2.megas4.com"+ - Mega S4 Montreal 2- Provider: Mega- - "s3.eu-central-2.s4.mega.io"- - Mega S4 eu-central-2 (Bettembourg, legacy)+ - "s3.ca-vancouver-1.megas4.com"+ - Mega S4 Vancouver 1- Provider: Mega- - "s3.ca-central-1.s4.mega.io"- - Mega S4 ca-central-1 (Montreal, legacy)+ - "s3.ca-vancouver-2.megas4.com"+ - Mega S4 Vancouver 2- Provider: Mega- - "s3.ca-west-1.s4.mega.io"- - Mega S4 ca-west-1 (Vancouver, legacy)+ - "s3.ap-tokyo-1.megas4.com"+ - Mega S4 Tokyo 1+ - Provider: Mega+ - "s3.ap-tokyo-2.megas4.com"+ - Mega S4 Tokyo 2- Provider: Mega- "oos.eu-west-2.outscale.com"- Outscale EU West 2 (Paris)@@ -30404,6 +32065,12 @@ Properties:- "nz.s3.rackcorp.com"- Auckland (New Zealand) Endpoint- Provider: RackCorp+ - "s3-gw1.relaix.net"+ - Aachen-1, Germany+ - Provider: RelAix+ - "s3-gw2.relaix.net"+ - Aachen-2, Germany+ - Provider: RelAix- "s3.nl-ams.scw.cloud"- Amsterdam Endpoint- Provider: Scaleway@@ -30413,6 +32080,9 @@ Properties:- "s3.pl-waw.scw.cloud"- Warsaw Endpoint- Provider: Scaleway+ - "s3.it-mil.scw.cloud"+ - Milan Endpoint+ - Provider: Scaleway- "localhost:8333"- SeaweedFS S3 localhost- Provider: SeaweedFS@@ -30650,7 +32320,7 @@ Properties:- Config: location_constraint- Env Var: RCLONE_S3_LOCATION_CONSTRAINT- Provider:- AWS,ArvanCloud,Ceph,ChinaMobile,DigitalOcean,Dreamhost,Exaba,GCS,HCP,Hetzner,IBMCOS,ImpossibleCloud,LyveCloud,Minio,Netease,Qiniu,Rabata,RackCorp,Scality,SeaweedFS,Synology,Wasabi,Zata,ZeroServices,Other+ AWS,ArvanCloud,Ceph,ChinaMobile,DigitalOcean,Dreamhost,Exaba,GCS,HCP,Hetzner,IBMCOS,ImpossibleCloud,Infomaniak,LyveCloud,Minio,Netease,Qiniu,Rabata,RackCorp,Scality,SeaweedFS,Synology,Wasabi,Zata,ZeroServices,Other- Type: string- Required: false- Examples:@@ -31033,7 +32703,7 @@ Properties:- Config: acl- Env Var: RCLONE_S3_ACL- Provider:- AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IBMCOS,IDrive,ImpossibleCloud,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,Scaleway,Scality,SeaweedFS,TencentCOS,US3,Wasabi,Zata,ZeroServices,Other+ AWS,Alibaba,ArvanCloud,BizflyCloud,Cellar,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IBMCOS,IDrive,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,RelAix,Scaleway,Scality,SeaweedFS,TencentCOS,US3,Wasabi,Zata,ZeroServices,Other- Type: string- Required: false- Examples:@@ -31041,37 +32711,37 @@ Properties:- Owner gets FULL_CONTROL.- No one else has access rights (default).- Provider:- AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,Scaleway,Scality,SeaweedFS,US3,Wasabi,Zata,ZeroServices,Other+ AWS,Alibaba,ArvanCloud,BizflyCloud,Cellar,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,RelAix,Scaleway,Scality,SeaweedFS,US3,Wasabi,Zata,ZeroServices,Other- "public-read"- Owner gets FULL_CONTROL.- The AllUsers group gets READ access.- Provider:- AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,Scaleway,Scality,SeaweedFS,TencentCOS,US3,Wasabi,Zata,ZeroServices,Other+ AWS,Alibaba,ArvanCloud,BizflyCloud,Cellar,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,RelAix,Scaleway,Scality,SeaweedFS,TencentCOS,US3,Wasabi,Zata,ZeroServices,Other- "public-read-write"- Owner gets FULL_CONTROL.- The AllUsers group gets READ and WRITE access.- Granting this on a bucket is generally not recommended.- Provider:- AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,Scaleway,Scality,SeaweedFS,TencentCOS,Wasabi,Zata,ZeroServices,Other+ AWS,Alibaba,ArvanCloud,BizflyCloud,Cellar,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,RelAix,Scaleway,Scality,SeaweedFS,TencentCOS,Wasabi,Zata,ZeroServices,Other- "authenticated-read"- Owner gets FULL_CONTROL.- The AuthenticatedUsers group gets READ access.- Provider:- AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,Scaleway,Scality,SeaweedFS,TencentCOS,Wasabi,Zata,ZeroServices,Other+ AWS,Alibaba,ArvanCloud,BizflyCloud,Cellar,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,RelAix,Scaleway,Scality,SeaweedFS,TencentCOS,Wasabi,Zata,ZeroServices,Other- "bucket-owner-read"- Object owner gets FULL_CONTROL.- Bucket owner gets READ access.- If you specify this canned ACL when creating a bucket,Amazon S3 ignores it.- Provider:- AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,Scaleway,Scality,SeaweedFS,TencentCOS,Wasabi,Zata,ZeroServices,Other+ AWS,Alibaba,ArvanCloud,BizflyCloud,Cellar,Ceph,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,RelAix,Scaleway,Scality,SeaweedFS,TencentCOS,Wasabi,Zata,ZeroServices,Other- "bucket-owner-full-control"- Both the object owner and the bucket owner get FULL_CONTROLover the object.- If you specify this canned ACL when creating a bucket,Amazon S3 ignores it.- Provider:- AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,Scaleway,Scality,SeaweedFS,TencentCOS,Wasabi,Zata,ZeroServices,Other+ AWS,Alibaba,ArvanCloud,BizflyCloud,Cellar,Ceph,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,RelAix,Scaleway,Scality,SeaweedFS,TencentCOS,Wasabi,Zata,ZeroServices,Other- "private"- Owner gets FULL_CONTROL.- No one else has access rights (default).@@ -31275,14 +32945,15 @@ Properties:Advanced optionsHere are the Advanced options specific to s3 (Amazon S3 Compliant-Storage Providers including AWS, Alibaba, ArvanCloud, BizflyCloud, Ceph,-ChinaMobile, Cloudflare, Cubbit, DigitalOcean, Dreamhost, Exaba, Fastly,-FileLu, FlashBlade, GCS, HCP, Hetzner, HuaweiOBS, IBMCOS, IDrive,-ImpossibleCloud, Intercolo, IONOS, Leviia, Liara, Linode, LyveCloud,-Magalu, Mega, Minio, Netease, Outscale, OVHcloud, Petabox, Qiniu,-Rabata, RackCorp, Rclone, Scaleway, Scality, SeaweedFS, Selectel,-Servercore, SpectraLogic, Storj, Synology, TencentCOS, US3, Wasabi,-Zadara, Zata, ZeroServices, Other).+Storage Providers including AWS, Alibaba, ArvanCloud, Autonomys,+BizflyCloud, Cellar, Ceph, ChinaMobile, Cloudflare, Cubbit,+DigitalOcean, Dreamhost, Exaba, Fastly, FileLu, FlashBlade, GCS, HCP,+Hetzner, HuaweiOBS, IBMCOS, IDrive, ImpossibleCloud, Infomaniak,+Intercolo, IONOS, Leviia, Liara, Linode, LyveCloud, Magalu, Mega, Minio,+Neon, Netease, Outscale, OVHcloud, Petabox, Qiniu, Rabata, RackCorp,+Rclone, RelAix, Scaleway, Scality, SeaweedFS, Selectel, Servercore,+SpectraLogic, Storj, Synology, TencentCOS, US3, Wasabi, Zadara, Zata,+ZeroServices, Other).--s3-bucket-acl@@ -31302,7 +32973,7 @@ Properties:- Config: bucket_acl- Env Var: RCLONE_S3_BUCKET_ACL- Provider:- AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IBMCOS,IDrive,ImpossibleCloud,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Mega,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,Scaleway,SeaweedFS,Servercore,TencentCOS,US3,Wasabi,Zata,ZeroServices,Other+ AWS,Alibaba,ArvanCloud,BizflyCloud,Cellar,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IBMCOS,IDrive,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Mega,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,RelAix,Scaleway,SeaweedFS,Servercore,TencentCOS,US3,Wasabi,Zata,ZeroServices,Other- Type: string- Required: false- Examples:@@ -31341,7 +33012,7 @@ Properties:- Config: sse_customer_algorithm- Env Var: RCLONE_S3_SSE_CUSTOMER_ALGORITHM-- Provider: AWS,Ceph,ChinaMobile,Minio+- Provider: AWS,Cellar,Ceph,ChinaMobile,Minio- Type: string- Required: false- Examples:@@ -31355,13 +33026,13 @@ Properties:To use SSE-C you may provide the secret encryption key used toencrypt/decrypt your data.-Alternatively you can provide --sse-customer-key-base64.+Alternatively you can provide --s3-sse-customer-key-base64.Properties:- Config: sse_customer_key- Env Var: RCLONE_S3_SSE_CUSTOMER_KEY-- Provider: AWS,Ceph,ChinaMobile,Minio+- Provider: AWS,Cellar,Ceph,ChinaMobile,Minio- Type: string- Required: false- Examples:@@ -31373,13 +33044,13 @@ Properties:If using SSE-C you must provide the secret encryption key encoded inbase64 format to encrypt/decrypt your data.-Alternatively you can provide --sse-customer-key.+Alternatively you can provide --s3-sse-customer-key.Properties:- Config: sse_customer_key_base64- Env Var: RCLONE_S3_SSE_CUSTOMER_KEY_BASE64-- Provider: AWS,Ceph,ChinaMobile,Minio+- Provider: AWS,Cellar,Ceph,ChinaMobile,Minio- Type: string- Required: false- Examples:@@ -31398,7 +33069,7 @@ Properties:- Config: sse_customer_key_md5- Env Var: RCLONE_S3_SSE_CUSTOMER_KEY_MD5-- Provider: AWS,Ceph,ChinaMobile,Minio+- Provider: AWS,Cellar,Ceph,ChinaMobile,Minio- Type: string- Required: false- Examples:@@ -32502,6 +34173,46 @@ arguments.These can be run on a running backend using the rc commandbackend/command.+link++Generate a signed link with response header overrides.++ rclone backend link remote: [options] [<arguments>+]++This command generates a signed download link for one file, with+optional overrides for the HTTP response headers. Pass the file path as+a separate argument, relative to the remote path.++Usage examples:++ rclone backend link s3:bucket path/to/file -o expire=1h+ rclone backend link s3:bucket path/to/file -o response-expires="Thu, 01 Jan 1970 00:00:00 GMT"+ rclone backend link s3:bucket path/to/file -o response-content-disposition='attachment; filename="download.txt"'++The link expires after 7 days by default. Use -o expire=1h to change its+lifetime. Durations must be at least 1 second; values above 7 days are+limited to 7 days, as with rclone link.++The response-expires option sets the HTTP Expires response header, not+the lifetime of the signed link. It must be an HTTP date, such as+Thu, 01 Jan 1970 00:00:00 GMT.++The overrides are included in the signature and must not be changed in+the returned URL. They do not modify the object's stored metadata.++Options:++- "expire": How long the link will be valid (default 7d, maximum 7d).+- "response-cache-control": Set the Cache-Control response header.+- "response-content-disposition": Set the Content-Disposition response+ header.+- "response-content-encoding": Set the Content-Encoding response+ header.+- "response-content-language": Set the Content-Language response+ header.+- "response-content-type": Set the Content-Type response header.+- "response-expires": Set the Expires response header to an HTTP date.+restoreRestore objects from GLACIER or INTELLIGENT-TIERING archive tier.@@ -32756,15 +34467,15 @@ AWS Directory BucketsFrom rclone v1.69 Directory Buckets are supported.-You will need to set the directory_buckets = true config parameter or-use --s3-directory-buckets.+You will need to set the directory_bucket = true config parameter or use+--s3-directory-bucket.Note that rclone cannot yet:- Create directory buckets- List directory buckets-See the --s3-directory-buckets flag for more info+See the --s3-directory-bucket flag for more infoAWS Snowball Edge@@ -33005,6 +34716,91 @@ This will leave the config file looking like this.server_side_encryption =storage_class =+Autonomys Auto Drive++Auto Drive is permanent storage on the Autonomys Network, with an+S3-compatible API. Deleting an object hides it, but its data and every+earlier version are kept. Buckets are created by the first upload into+them.++Use your Auto Drive API key as the access_key_id, and any value as the+secret_access_key.++Here is an example of making a configuration. First run:++ rclone config++This will guide you through an interactive setup process.++ No remotes found, make a new one?+ n) New remote+ s) Set configuration password+ q) Quit config+ n/s/q> n++ Enter name for new remote.+ name> autodrive++ Option Storage.+ Type of storage to configure.+ Storage> s3++ Option provider.+ Choose your S3 provider.+ provider> Autonomys++ Option env_auth.+ Get AWS credentials from runtime (environment variables or EC2/ECS meta data if no env vars).+ Only applies if access_key_id and secret_access_key is blank.+ env_auth> false++ Option access_key_id.+ AWS Access Key ID.+ access_key_id> YOUR_API_KEY++ Option secret_access_key.+ AWS Secret Access Key (password).+ secret_access_key> ANY_VALUE++ Option endpoint.+ Endpoint for S3 API.+ Choose a number from below, or type in your own value.+ 1 / Auto Drive mainnet S3 endpoint+ \ (s3.auto-drive.autonomys.xyz)+ endpoint> 1++ Option bucket_object_lock_enabled.+ Enable Object Lock when creating new buckets.+ Enter a boolean value (true or false). Press Enter for the default (false).+ bucket_object_lock_enabled>++ Edit advanced config?+ y) Yes+ n) No (default)+ y/n> n++ Configuration complete.+ Options:+ - type: s3+ - provider: Autonomys+ - access_key_id: YOUR_API_KEY+ - secret_access_key: ANY_VALUE+ - endpoint: s3.auto-drive.autonomys.xyz+ Keep this "autodrive" remote?+ y) Yes this is OK (default)+ e) Edit this remote+ d) Delete this remote+ y/e/d> y++The resulting configuration file should look like:++ [autodrive]+ type = s3+ provider = Autonomys+ access_key_id = YOUR_API_KEY+ secret_access_key = ANY_VALUE+ endpoint = s3.auto-drive.autonomys.xyz+BizflyCloudBizfly Cloud Simple Storage is an S3-compatible service with regions in@@ -33061,6 +34857,12 @@ version of rclone before v1.59 then you may need to supply the parameter--s3-upload-cutoff 0 or put this in the config file as upload_cutoff 0to work around a bug which causes uploading of small files to fail.+Rclone defaults sign_accept_encoding to false for the Ceph provider.+That avoids SignatureDoesNotMatch errors when a reverse proxy (for+example Cloudflare) rewrites the Accept-Encoding request header. You can+override this with --s3-sign-accept-encoding / sign_accept_encoding if+needed.+Note also that Ceph sometimes puts / in the passwords it gives users. Ifyou read the secret access key using the command line tools you will geta JSON blob with the / escaped as \/. Make sure you only write / in the@@ -33332,6 +35134,110 @@ This will guide you through an interactive setup process.d) Delete this remotey/e/d> y+Clever Cloud Cellar++Cellar is the S3-compatible object storage service of Clever Cloud,+hosted in France. Credentials and the endpoint are shown in the Cellar+add-on dashboard, or in the CELLAR_ADDON_KEY_ID, CELLAR_ADDON_KEY_SECRET+and CELLAR_ADDON_HOST environment variables of applications linked to+it.++Cellar does not apply Object Lock headers sent when initiating a+multipart upload. When setting Object Lock retention or legal hold+options, also set --s3-object-lock-set-after-upload: the lock is then+applied with separate API calls after the upload, so the object is not+protected by these settings until those calls succeed.++Cellar supports server-side encryption with customer-provided keys+(SSE-C), set with sse_customer_algorithm and sse_customer_key_base64. It+rejects server-side copies of SSE-C objects unless they are done as+multipart copies, so use --no-update-modtime to avoid modification time+updates, and expect server-side moves and renames of encrypted objects+smaller than --s3-copy-cutoff to fail. See the Cellar SSE-C+documentation.++First run:++ rclone config++ This will guide you through an interactive setup process:++ n) New remote+ n/s/q> n++ Enter name for new remote.+ name> cellar++ Option Storage.+ Type of storage to configure.+ Choose a number from below, or type in your own value.+ [snip]+ XX / Amazon S3 Compliant Storage Providers including AWS, ..., Cellar, ... and others+ \ (s3)+ Storage> s3++ Option provider.+ Choose your S3 provider.+ [snip]+ XX / Clever Cloud Cellar Object Storage+ \ (Cellar)+ provider> Cellar++ Option env_auth.+ Get AWS credentials from runtime (environment variables or EC2/ECS meta data if no env vars).+ 1 / Enter AWS credentials in the next step.+ \ (false)+ 2 / Get AWS credentials from the environment (env vars or IAM).+ \ (true)+ env_auth> 1++ Option access_key_id.+ AWS Access Key ID.+ access_key_id> ACCESS_KEY++ Option secret_access_key.+ AWS Secret Access Key (password).+ secret_access_key> SECRET_KEY++ Option endpoint.+ Endpoint for S3 API.+ 1 / Paris, France+ \ (cellar-c2.services.clever-cloud.com)+ endpoint> 1++ Option acl.+ Canned ACL used when creating buckets and storing or copying objects.+ [snip]+ acl> 1++ Option bucket_object_lock_enabled.+ Enable Object Lock when creating new buckets.+ bucket_object_lock_enabled>++ Edit advanced config?+ y/n> n++ Configuration complete.+ Options:+ - type: s3+ - provider: Cellar+ - access_key_id: ACCESS_KEY+ - secret_access_key: SECRET_KEY+ - endpoint: cellar-c2.services.clever-cloud.com+ - acl: private+ Keep this "cellar" remote?+ y/e/d> y++This will leave the config file looking like this.++ [cellar]+ type = s3+ provider = Cellar+ access_key_id = ACCESS_KEY+ secret_access_key = SECRET_KEY+ endpoint = cellar-c2.services.clever-cloud.com+ acl = private+Cloudflare R2Cloudflare R2 Storage allows developers to store large amounts of@@ -34583,6 +36489,104 @@ This will leave the config file looking like this.endpoint = eu-central-2.storage.impossibleapi.netacl = private+Infomaniak Swiss Backup++Infomaniak Swiss Backup is a cloud backup space in Switzerland that you+can use however you want with the script or software of your choice that+is compatible with Swift, S3 or SFTP.++Prerequisites++Create and setup your Swiss Backup product following this documentation.+Memorize the generated Endpoint, Access Key, and Secret Key:++Configuration++To configure access to an S3 location on Swiss Backup, follow the steps+below:++1. Run rclone config and select n for a new remote.++ rclone config+ No remotes found, make a new one?+ n) New remote+ s) Set configuration password+ q) Quit config+ n/s/q> n++2. Give the name of the configuration. For example, name it+ 'swissbackup'.++ name> swissbackup++3. Select s3 storage.++ Type of storage to configure.+ XX / Amazon S3 Compliant Storage Providers including AWS, ... - Tier 1+ \ (s3)+ Storage> s3++4. Select Infomaniak provider.++ Choose your S3 provider.+ xx / Infomaniak Swiss Backup+ \ (Infomaniak)+ provider> Infomaniak++5. Enter your Access Key and Secret Key. You can generate new ones by+ clicking 'Generate new keys' on your Swiss Backup product dashboard.++ Option env_auth.+ 1 / Enter AWS credentials in the next step+ \ (false)+ 2 / Get AWS credentials from the environment (env vars or IAM)+ \ (true)+ env_auth> 1++ AWS Access Key ID.+ access_key_id> {YOUR-ACCESS-ID}++ AWS Secret Access Key (password)+ secret_access_key> {YOUR-ACCESS-SECRET}++6. Select us-east-1 region for compatibility.++ Region to connect to.+ / Default region for compatibility.+ 1 | All data is stored in Switzerland.+ \ (us-east-1)+ region> 1++7. Select the endpoint corresponding to the one generated with your+ Access and Secret Keys.++ Endpoint for S3 API.+ 1 / Swiss Backup Location 01+ \ (s3.swiss-backup01.infomaniak.com)+ 2 / Swiss Backup Location 02+ \ (s3.swiss-backup02.infomaniak.com)+ 3 / Swiss Backup Location 03+ \ (s3.swiss-backup03.infomaniak.com)+ 4 / Swiss Backup Location 04+ \ (s3.swiss-backup04.infomaniak.com)+ endpoint> 1++8. Adjust other parameters and advanced config at your convenience.+ Leave blank if not sure.++9. You should obtain something like this.++ Configuration complete.+ Options:+ - type: s3+ - provider: Infomaniak+ - access_key_id: xxxxxxxxxx+ - secret_access_key: xxxxxxxxxx+ - region: us-east-1+ - endpoint: s3.swiss-backup01.infomaniak.com++10. Press y to validate. Then q to quit.+Intercolo Object StorageIntercolo Object Storage offers GDPR-compliant, transparently priced,@@ -35222,6 +37226,10 @@ This will leave the config file looking like this.secret_access_key = SECRET_ACCESS_KEYendpoint = eu-central-1.linodeobjects.com+Linode Object Storage is Ceph-backed. Rclone defaults+sign_accept_encoding to false for this provider so SigV4 requests do not+fail with SignatureDoesNotMatch (see the Ceph section).+MagaluHere is an example of making a Magalu Object Storage configuration.@@ -35501,6 +37509,150 @@ So once set up, for example, to copy files into a bucketrclone copy /path/to/files minio:bucket+Neon Object Storage++Neon Object Storage is S3 compatible object storage in a Neon project.+Each branch of the project has its own copy of the buckets and its own+S3 endpoint. The endpoint looks like+https://<branch_id>.storage.<cell>.<region>.aws.neon.tech.++In the Neon Console, click Connect, open the Storage tab and choose+Parameters only. Use AWS_ACCESS_KEY_ID, AWS_SECRET_ACCESS_KEY,+AWS_ENDPOINT_URL_S3 and AWS_REGION as access_key_id, secret_access_key,+endpoint and region. To make a separate credential, use Settings >+Credentials or run+neon credentials create --scope storage:read --scope storage:write.++A credential works on the branch it was made on and on that branch's+child branches. To use a child branch, change only endpoint.++Neon also has these limits (see the S3 compatibility docs):++- The maximum object size is 5 GiB for a single upload and 10 GiB for+ a multipart upload.+- Neon rejects object keys containing control characters with+ 400 InvalidURI. To store files with these names, add Ctl,Del to the+ encoding, for example encoding = Slash,InvalidUtf8,Dot,Ctl,Del.+- Neon doesn't support ACLs over the S3 API. Leave acl and bucket_acl+ empty and set bucket access (private or public_read) in Neon.+- Neon stores versioning and lifecycle configurations but doesn't+ enforce them.+- Neon may throttle heavy use with 503 SlowDown errors. rclone retries+ these.++Here is an example of making a Neon Object Storage configuration. First+run:++ rclone config++This will guide you through an interactive setup process.++ No remotes found, make a new one?+ n) New remote+ s) Set configuration password+ q) Quit config+ n/s/q> n++ Enter name for new remote.+ name> neon++ Option Storage.+ Type of storage to configure.+ Choose a number from below, or type in your own value.+ [snip]+ XX / Amazon S3 Compliant Storage Providers including AWS, ... Neon, ...+ \ (s3)+ [snip]+ Storage> s3++ Option provider.+ Choose your S3 provider.+ Choose a number from below, or type in your own value.+ Press Enter to leave empty.+ [snip]+ XX / Neon Object Storage+ \ (Neon)+ [snip]+ provider> Neon++ Option env_auth.+ Get AWS credentials from runtime (environment variables or EC2/ECS meta data if no env vars).+ Only applies if access_key_id and secret_access_key is blank.+ Choose a number from below, or type in your own boolean value (true or false).+ Press Enter for the default (false).+ 1 / Enter AWS credentials in the next step.+ \ (false)+ 2 / Get AWS credentials from the environment (env vars or IAM).+ \ (true)+ env_auth>++ Option access_key_id.+ AWS Access Key ID.+ Leave blank for anonymous access or runtime credentials.+ Enter a value. Press Enter to leave empty.+ access_key_id> NEON_ACCESS_KEY_ID++ Option secret_access_key.+ AWS Secret Access Key (password).+ Leave blank for anonymous access or runtime credentials.+ Enter a value. Press Enter to leave empty.+ secret_access_key> NEON_SECRET_ACCESS_KEY++ Option region.+ Region to connect to.+ Leave blank if you are using an S3 clone and you don't have a region.+ Choose a number from below, or type in your own value.+ Press Enter to leave empty.+ 1 / US East (N. Virginia)+ \ (us-east-1)+ 2 / US East (Ohio)+ \ (us-east-2)+ 3 / Europe (Frankfurt)+ \ (eu-central-1)+ 4 / Asia Pacific (Singapore)+ \ (ap-southeast-1)+ region> us-east-2++ Option endpoint.+ Endpoint for S3 API.+ Required when using an S3 clone.+ Enter a value. Press Enter to leave empty.+ endpoint> https://<branch_id>.storage.<cell>.<region>.aws.neon.tech++ Option bucket_object_lock_enabled.+ Enable Object Lock when creating new buckets.+ Enter a boolean value (true or false). Press Enter for the default (false).+ bucket_object_lock_enabled>++ Edit advanced config?+ y) Yes+ n) No (default)+ y/n> n++ Configuration complete.+ Options:+ - type: s3+ - provider: Neon+ - access_key_id: NEON_ACCESS_KEY_ID+ - secret_access_key: NEON_SECRET_ACCESS_KEY+ - region: us-east-2+ - endpoint: https://<branch_id>.storage.<cell>.<region>.aws.neon.tech+ Keep this "neon" remote?+ y) Yes this is OK (default)+ e) Edit this remote+ d) Delete this remote+ y/e/d> y++This will leave the config file looking like this.++ [neon]+ type = s3+ provider = Neon+ access_key_id = NEON_ACCESS_KEY_ID+ secret_access_key = NEON_SECRET_ACCESS_KEY+ region = us-east-2+ endpoint = https://<branch_id>.storage.<cell>.<region>.aws.neon.tech+Netease NOSFor Netease NOS configure as per the configurator rclone config setting@@ -36579,6 +38731,127 @@ this:Note that setting use_multipart_uploads = false is to work around a bugwhich will be fixed in due course.+RelAix.Storage++RelAix.Storage is S3 compatible object storage from RelAix Networks,+operated in its own data centres in Aachen, Germany. There are two+locations, Aachen-1 and Aachen-2, each with its own endpoint. Buckets+and data stay at the location of the endpoint you use.++You get an access key and a secret key from RelAix Networks. Then run:++ rclone config++This will guide you through an interactive setup process.++ No remotes found, make a new one?+ n) New remote+ s) Set configuration password+ q) Quit config+ n/s/q> n++ Enter name for new remote.+ name> my-relaix++ Option Storage.+ Type of storage to configure.+ Choose a number from below, or type in your own value.+ [snip]+ XX / Amazon S3 Compliant Storage Providers including AWS, ...+ \ (s3)+ [snip]+ Storage> s3++ Option provider.+ Choose your S3 provider.+ Choose a number from below, or type in your own value.+ Press Enter to leave empty.+ [snip]+ XX / RelAix.Storage+ \ (RelAix)+ [snip]+ provider> RelAix++ Option env_auth.+ Get AWS credentials from runtime (environment variables or EC2/ECS meta data if no env vars).+ Only applies if access_key_id and secret_access_key is blank.+ Choose a number from below, or type in your own boolean value (true or false).+ Press Enter for the default (false).+ 1 / Enter AWS credentials in the next step.+ \ (false)+ 2 / Get AWS credentials from the environment (env vars or IAM).+ \ (true)+ env_auth>++ Option access_key_id.+ AWS Access Key ID.+ Leave blank for anonymous access or runtime credentials.+ Enter a value. Press Enter to leave empty.+ access_key_id> ACCESS_KEY++ Option secret_access_key.+ AWS Secret Access Key (password).+ Leave blank for anonymous access or runtime credentials.+ Enter a value. Press Enter to leave empty.+ secret_access_key> SECRET_KEY++ Option endpoint.+ Endpoint for S3 API.+ Required when using an S3 clone.+ Choose a number from below, or type in your own value.+ Press Enter to leave empty.+ 1 / Aachen-1, Germany+ \ (s3-gw1.relaix.net)+ 2 / Aachen-2, Germany+ \ (s3-gw2.relaix.net)+ endpoint> 1++ Option acl.+ Canned ACL used when creating buckets and storing or copying objects.+ [snip]+ acl>++ Option bucket_object_lock_enabled.+ Enable Object Lock when creating new buckets.+ Enter a boolean value (true or false). Press Enter for the default (false).+ bucket_object_lock_enabled>++ Edit advanced config?+ y) Yes+ n) No (default)+ y/n> n++ Configuration complete.+ Options:+ - type: s3+ - provider: RelAix+ - access_key_id: ACCESS_KEY+ - secret_access_key: SECRET_KEY+ - endpoint: s3-gw1.relaix.net+ Keep this "my-relaix" remote?+ y) Yes this is OK (default)+ e) Edit this remote+ d) Delete this remote+ y/e/d> y++This will leave the config file looking like this.++ [my-relaix]+ type = s3+ provider = RelAix+ access_key_id = ACCESS_KEY+ secret_access_key = SECRET_KEY+ endpoint = s3-gw1.relaix.net++Object Lock can only be enabled when a bucket is created. To create+buckets with Object Lock, set bucket_object_lock_enabled = true before+running rclone mkdir.++To lock objects on upload with object_lock_mode, also set+object_lock_set_after_upload = true. RelAix.Storage does not yet apply+Object Lock headers sent with multipart uploads, so rclone needs to set+the lock in a separate call after the upload.+ScalewayScaleway The Object Storage platform allows you to store anything from@@ -38356,6 +40629,23 @@ Rclone works fastest with large squashfs block sizes. For example:mksquashfs 100files 100files.sqfs -comp zstd -b 1M+Symlinks++Symlinks stored in an archive are skipped unless the -l/--links flag is+supplied, as they are on the other backends. With the flag they are+listed with a .rclonelink suffix and read back as the path they point+at, the same convention the local backend uses, so they can be copied+back as symlinks:++ $ rclone lsl --links :archive:/tmp/test.zip+ 5 2026-09-28 14:16:59.000000000 src/file.txt+ 8 2026-09-28 14:16:59.000000000 src/link.txt.rclonelink+ $ rclone cat --links :archive:/tmp/test.zip/src/link.txt.rclonelink+ file.txt+ $ rclone copy --links :archive:/tmp/test.zip /tmp/out+ $ ls -l /tmp/out/src/link.txt+ lrwxr-xr-x 1 user user 8 Sep 28 14:16 /tmp/out/src/link.txt -> file.txt+LimitationsFiles in the archive backend are read only. It isn't possible to create@@ -39090,6 +41380,10 @@ endpoint rewrites the requests for authentication, e.g., in CloudflareWorkers, this header needs to be handled properly. Leave blank if youwant to use the endpoint provided by Backblaze.+Rclone will not follow a redirect from this endpoint which downgrades+HTTPS to HTTP, and will not send the "Authorization" header or the SSE-C+key headers once a redirect has left the host and port of this endpoint.+The URL provided here SHOULD have the protocol and SHOULD NOT have atrailing slash or specify the /file/bucket subpath as rclone willrequest files with "{download_url}/file/{bucket_name}/{path}".@@ -39189,7 +41483,7 @@ Properties:To use SSE-C, you may provide the secret encryption key encoded in aUTF-8 compatible string to encrypt/decrypt your data-Alternatively you can provide --sse-customer-key-base64.+Alternatively you can provide --b2-sse-customer-key-base64.Properties:@@ -39206,7 +41500,7 @@ Properties:To use SSE-C, you may provide the secret encryption key encoded inBase64 format to encrypt/decrypt your data-Alternatively you can provide --sse-customer-key.+Alternatively you can provide --b2-sse-customer-key.Properties:@@ -39473,7 +41767,8 @@ Invalid refresh tokenAccording to the box docs:- Each refresh_token is valid for one use in 60 days.+ A Refresh Token is valid for 60 days and can be used to obtain a new+ Access Token and Refresh Token only once.This means that if you@@ -42443,6 +44738,15 @@ This means that- filenames with the same name will encrypt the same- filenames which start the same won't have a common prefix+A version string of the form -vYYYY-MM-DD-HHMMSS-NNN on the end of a+file name (as added by --b2-versions / --s3-versions) is left in plain+text so that versioned files can be found. Directory names are encrypted+in full. Rclone before v1.76 left such a suffix in plain text on+directory names too, so a directory named like this created by an older+rclone will appear in listings with a warning but can't be opened or+removed until renamed on the underlying remote to the name given in the+warning.+This uses a 32 byte key (256 bits) and a 16 byte (128 bits) IV both ofwhich are derived from the user password.@@ -43032,6 +45336,269 @@ will default to those currently in use.It doesn't return anything.+dosya.dev++dosya.dev is a cloud file storage, sharing, and synchronization platform+with workspace-based multi-tenancy, file versioning, and team+collaboration features.++Configuration++Here is an example of how to make a remote called remote. First run:++ rclone config++This will guide you through an interactive setup process:++ No remotes found, make a new one?+ n) New remote+ s) Set configuration password+ q) Quit config+ n/s/q> n++ Enter name for new remote.+ name> remote++ Option Storage.+ Type of storage to configure.+ Choose a number from below, or type in your own value.+ XX / dosya.dev+ \ (dosya)+ Storage> dosya++ Option api_key.+ Your API Key, get it from https://dosya.dev/profile under "API keys".+ Enter a value.+ api_key> dos_YOURAPIKEY++ Option workspace_id.+ Your workspace ID.++ Leave this blank to use the one workspace the API key can reach, which+ is what a key restricted to a single workspace does. If the key can+ reach several, rclone will refuse and list their IDs for you to choose+ from.+ Enter a value. Press Enter to leave empty.+ workspace_id>++ Edit advanced config?+ y) Yes+ n) No (default)+ y/n> n++ Configuration complete.+ Options:+ - type: dosya+ - api_key: dos_YOURAPIKEY+ Keep this "remote" remote?+ y) Yes this is OK (default)+ e) Edit this remote+ d) Delete this remote+ y/e/d> y++Once configured you can then use rclone like this (replace remote with+the name you gave your remote):++List directories and files in the top level of your dosya workspace++ rclone lsf remote:++To copy a local directory to a dosya directory called backup++ rclone copy /home/source remote:backup++The dosya backend uploads large files in parts and stores them as single+files that remain accessible through the web app and other clients. If+you previously used a WebDAV remote with a chunker overlay for large+uploads, you can use this backend without the overlay.++Getting your API key++1. Log in to dosya.dev+2. Go to Profile > API keys+3. Create a new API key+4. Copy the key (starts with dos_)++A key can be restricted to a single workspace when you create it, which+is the simplest thing to do for rclone: workspace_id can then be left+out of the config entirely.++Workspaces++Leave workspace_id blank and rclone asks which workspaces the key can+reach. A key restricted to one workspace, and an account that has only+one, both resolve without any configuration. If the key can reach+several, rclone refuses and lists their IDs so you can set the one you+want:++ this API key can reach 2 workspaces, so workspace_id must be set to one of:+ ws_AAAAAAAA (Work)+ ws_BBBBBBBB (Personal)++To access several workspaces at once, create a separate remote for each:++ rclone config create work dosya api_key=dos_XXX workspace_id=ws_WORK+ rclone config create personal dosya api_key=dos_XXX workspace_id=ws_PERSONAL++Then use them independently:++ rclone ls work:+ rclone copy work:docs/ personal:backup/++Modification times and hashes++dosya.dev supports modification times with a resolution of 1 second. The+modification time is sent with the upload, so a copied or synced file+keeps its original timestamp. Modification times cannot be changed after+upload without re-uploading the file.++dosya.dev supports SHA-256 checksums. The server computes the hash of a+single-shot (small file) upload from the bytes it receives, so+rclone check and hash-based sync work without downloading. Large files+uploaded in multiple parts have no server-side hash (the parts are never+re-read to hash the whole object), the same limitation the S3 backend+has for multipart objects; those objects report an empty hash.++Restricted filename characters++In addition to the default restricted characters set the following+characters are also replaced:++ Character Value Replacement+ ----------- ------- -------------+ " 0x22 "+ ' 0x27 '+ < 0x3C <+ > 0x3E >+ \ 0x5C \++Invalid UTF-8 bytes will also be replaced, as they can't be used in JSON+strings.++Public Links++dosya.dev supports rclone link to create public share links to files.+You can use the --expire flag to specify how long the link should be+valid.++ rclone link remote:path/to/file.txt+ rclone link --expire 7d remote:path/to/file.txt++Standard options++Here are the Standard options specific to dosya (dosya.dev).++--dosya-api-key++Your API Key, get it from https://dosya.dev/profile under "API keys".++Properties:++- Config: api_key+- Env Var: RCLONE_DOSYA_API_KEY+- Type: string+- Required: true++--dosya-workspace-id++Your workspace ID.++Leave this blank to use the one workspace the API key can reach, which+is what a key restricted to a single workspace does. If the key can+reach several, rclone will refuse and list their IDs for you to choose+from.++Properties:++- Config: workspace_id+- Env Var: RCLONE_DOSYA_WORKSPACE_ID+- Type: string+- Required: false++Advanced options++Here are the Advanced options specific to dosya (dosya.dev).++--dosya-api-url++API base URL.++Properties:++- Config: api_url+- Env Var: RCLONE_DOSYA_API_URL+- Type: string+- Default: "https://api.dosya.dev"++--dosya-upload-concurrency++Concurrency for multipart uploads.++This is the number of parts of the same file that are uploaded+concurrently. The first part is always uploaded on its own, because the+server creates the upload when it receives it.++Each part in flight is held in memory, so memory use grows with this+value times the part size the server chooses.++If you are uploading large files over high-speed links and these uploads+do not fully utilize your bandwidth, then increasing this may help to+speed up the transfers.++Properties:++- Config: upload_concurrency+- Env Var: RCLONE_DOSYA_UPLOAD_CONCURRENCY+- Type: int+- Default: 4++--dosya-encoding++The encoding for the backend.++See the encoding section in the overview for more info.++Properties:++- Config: encoding+- Env Var: RCLONE_DOSYA_ENCODING+- Type: Encoding+- Default:+ Slash,LtGt,DoubleQuote,SingleQuote,BackSlash,Del,Ctl,LeftSpace,RightSpace,InvalidUtf8,Dot++--dosya-description++Description of the remote.++Properties:++- Config: description+- Env Var: RCLONE_DOSYA_DESCRIPTION+- Type: string+- Required: false++Limitations++dosya.dev only supports filenames up to 255 characters in length.++Empty files (0 bytes) cannot be uploaded to dosya.dev.++Upload requires knowing the file size in advance, so streaming uploads+of unknown-size files are not supported.++Each workspace limits the number of uploads a user can have in flight at+once (5 by default, configurable in the workspace settings). When+--transfers is set higher than that limit, rclone waits for a free+upload slot rather than failing the transfer, retrying every 5 seconds+for up to 10 minutes.++Deleting files and directories++rclone deletes files and directories permanently, in a single request+(DELETE ?permanent=1), bypassing the workspace trash. Removing a+directory also asks the server whether it is empty (?if_empty=1) rather+than listing it first.+DrimeDrime is a cloud storage and transfer service focused on fast, resilient@@ -44862,6 +47429,33 @@ files in the directory.rclone sync --interactive /home/local/directory remote:dir+Restricted filename characters++In addition to the default restricted characters set the following+characters are also replaced:++ Character Value Replacement+ ----------- ------- -------------+ \ 0x5C \++File names can also not end with the following characters. These only+get replaced if they are the last character in the name:++ Character Value Replacement+ ----------- ------- -------------+ SP 0x20 ␠+ LF 0x0A ␊+ CR 0x0D ␍+ HT 0x09 ␉+ VT 0x0B ␋++Invalid UTF-8 bytes will also be replaced.++Files.com treats the replacements / and \ as if they were / and \, and+the names . and .. as if they were . and .., and rejects them with+Invalid path. This means that files with / or \ in their names, or which+are called . or .., can't be stored on Files.com.+HashesIn December 2024 files.com started supporting more checksums.@@ -47749,7 +50343,7 @@ usage is using the flag form --drive-size-as-quota when doing rclonels/lsl/lsf/lsjson/etc only.If you do use this flag for syncing (not recommended) then you will need-to use --ignore size also.+to use --ignore-size also.Properties:@@ -48609,7 +51203,7 @@ Here is how to create your own Google Drive client ID for rclone:You should now see the three scopes on your Data access page. Nowpress save at the bottom!-6. After adding scopes, click Audience Scroll down and click "+ Add+6. After adding scopes, click Audience. Scroll down and click "+ Addusers". Add yourself as a test user and press save.7. Go to Overview on the left panel, click "Create OAuth client".@@ -48625,6 +51219,15 @@ Here is how to create your own Google Drive client ID for rclone:9. Go to "Audience" and then click "PUBLISH APP" button and confirm.Add yourself as a test user if you haven't already.+ (If the "PUBLISH APP" button is greyed out, Google now requires a+ homepage URL and a privacy policy URL before publishing, even for a+ personal single-user app. Go to "Branding" in the left panel, fill+ in "Application home page" and "Application privacy policy link" - a+ free GitHub Pages site is fine for these if you don't have a domain+ of your own - then "Authorized domains" with the domain you used,+ and click Save. Return to "Audience" and the "PUBLISH APP" button+ should now be available.)+10. Provide the noted client ID and client secret to rclone.11. Run the web-based authorization flow from within rclone config, by@@ -49792,12 +52395,18 @@ Development Guide".Enable Drive Application Switch-1. Log in to the AppGallery Connect website and click "My Projects".-2. Find your project in the project list, select the application that- needs to access Drive in the project, and click "API Management".-3. In the "API Management" area of the "Project Settings" page, click- the switch on the right side of Drive Kit to ensure the switch is in- the open state.+1. Log in to AppGallery Connect with your Huawei account.+2. Create a project or select an existing project.+3. Add a web application to the project.+4. Open "Project settings" for the application and copy the OAuth 2.0+ Client ID and Client Secret from the "General" tab.+5. Open "API Management" for the project and enable "Cloud Drive".+6. Add http://127.0.0.1:53682/ as a redirect URI for the OAuth 2.0+ Client ID.++The older Huawei Developer Console may show some of the same credential+information, but AppGallery Connect is the place to create the project,+application, API permission, and redirect URI used by rclone.Sign Huawei Cloud Drive Privacy Notice and User Agreement@@ -49813,23 +52422,12 @@ Service Development Guide".Key steps summary:-1. Log into the Huawei Developer Console with your Huawei account.--2. Create a new project or select an existing project.--3. Go to "Manage APIs" and enable the "Drive Kit" API.--4. Click on "Credentials" in the left side panel.--5. Click "Create Credentials" and choose "OAuth client ID".--6. Choose "Web application" as the application type.--7. Add http://localhost:53682/ to the "Authorized redirect URIs".--8. Click "Create" and note down the Client ID and Client Secret.--9. Provide the noted client ID and client secret to rclone during+1. Create or select a project in AppGallery Connect.+2. Add a web application to the project.+3. Enable Cloud Drive in API Management.+4. Configure http://127.0.0.1:53682/ as the OAuth redirect URI.+5. Copy the OAuth 2.0 Client ID and Client Secret from the application.+6. Provide the client ID and client secret to rclone duringconfiguration.Scopes@@ -50985,11 +53583,13 @@ will be resolved relative to this url, according to the URL standard.This means with remote url https://beta.rclone.org/branch and path fix,the resolved URL will be https://beta.rclone.org/branch/fix, while withpath /fix the resolved URL will be https://beta.rclone.org/fix as the-absolute path is resolved from the root of the domain.+absolute path is resolved from the root of the domain. The path can't+change the scheme, host or user of the configured url, so a path such as+//other.example.com/fix is rejected.If the path following the remote: ends with / it will be assumed topoint to a directory. If the path does not end with /, then a HEAD-request is sent and the response used to decide if it it is treated as a+request is sent and the response used to decide if it is treated as afile or a directory (run with -vv to see details). When --http-no-headis specified, a path without ending / is always assumed to be a file. Ifrclone incorrectly assumes the path is a file, the solution is to@@ -51158,6 +53758,13 @@ For example, to set a Cookie use 'Cookie,name=value', orYou can set multiple headers, e.g.'"Cookie","name=value","Authorization","xxx"'.+The headers are only sent to the host in the configured URL. If the+server redirects to another host (including a subdomain or a different+port) the headers are not sent to it, or to any further hop in that+redirect chain. When headers are set, a redirect from https to http is+refused as it would send them in cleartext, even with+--allow-https-downgrade.+Properties:- Config: headers@@ -51212,6 +53819,44 @@ Properties:- Type: bool- Default: false+--http-no-head-size++Set the reported file size for --http-no-head.++When HEAD requests are disabled, rclone normally reports an unknown file+size. Set this option to a non-negative value to report that size for+every file, including 0. Leave it unset (the default, shown as "off") to+keep reporting an unknown size.++The configured size is authoritative. If it does not match the actual+file size, transfers may fail their size check and retry unless+--ignore-size is used. It can also make multi-thread transfers request+incorrect ranges and make mounted files unreadable past the reported+size.++Properties:++- Config: no_head_size+- Env Var: RCLONE_HTTP_NO_HEAD_SIZE+- Type: SizeSuffix+- Default: off++--http-no-head-time++Set a default modification time for --http-no-head.++When HEAD requests are disabled, rclone normally reports an unset+modification time. Set this option to report the specified time for+every file instead. Changing the configured time later makes every file+appear modified to a sync.++Properties:++- Config: no_head_time+- Env Var: RCLONE_HTTP_NO_HEAD_TIME+- Type: Time+- Default: off+--http-descriptionDescription of the remote.@@ -53127,6 +55772,322 @@ and folders which may cause Copy, Move and DirMove operations topreviously deleted paths to fail. Emptying the trash should help in suchcases.+kDrive++Paths are specified as remote:path++Paths may be as deep as required, e.g. remote:directory/subdirectory.++Configuration++The initial setup for kDrive involves getting an OAuth2 token which you+need to do in your browser. rclone config walks you through it.++kDrive uses the PKCE OAuth2 flow (RFC 7636), which means no+client_secret is required. The client_id is optional and defaults to+rclone's built-in application credentials.++Here is an example of how to make a remote called remote. First run:++ rclone config++This will guide you through an interactive setup process:++ No remotes found, make a new one?+ n) New remote+ s) Set configuration password+ q) Quit config+ n/s/q> n+ name> remote+ Type of storage to configure.+ Choose a number from below, or type in your own value+ [snip]+ XX / Infomaniak kDrive+ \ "kdrive"+ [snip]+ Storage> kdrive++ Option client_id.+ OAuth Client Id.+ Leave blank normally.+ Enter a value. Press Enter to leave empty.+ client_id>++ Option drive_id.+ ID of the kDrive to use.+ When showing a folder on kdrive, you can find the drive_id here:+ https://ksuite.infomaniak.com/{account_id}/kdrive/app/drive/{drive_id}/files/...+ Enter a value. Press Enter to leave empty.+ drive_id> 0654321++ Use web browser to automatically authenticate rclone with remote?+ * Say Y if the machine running rclone has a web browser you can use+ * Say N if running rclone on a (remote) machine without web browser access+ If not sure try Y. If Y failed, try N.+ y/n> y++ Waiting for code...+ Got code++ Configuration complete.+ Options:+ - type: kdrive+ - drive_id: 0654321++ Keep this "remote" remote?+ y) Yes this is OK+ e) Edit this remote+ d) Delete this remote+ y/e/d> y++Once configured you can then use rclone like this (replace remote with+the name you gave your remote):++List directories in top level of your rclone++ rclone lsd remote:++Note that by default the top level directory of a private kDrive storage+is your "Private" folder. When used with a Pro account, a second folder+"Common documents" is also available, which can be used as the root by+setting root_folder_id to common. These top-level folders cannot be+modified.++List all the files in your kDrive++ rclone ls remote:++To copy a local directory to a kDrive directory called backup++ rclone copy /home/source remote:backup++Modification times and hashes++kDrive allows modification times to be set on objects accurate to 1+second. These will be used to detect whether objects need syncing or+not.++kDrive supports the XXH3 checksum hash, so you can use the --checksum+flag. XXH3 is only available for files uploaded as a single chunk: files+uploaded in chunks (by rclone when larger than --kdrive-chunk-size, or+by the kDrive apps which chunk files at their own sizes) have a "nested"+hash which can't be compared with a plain XXH3, and rclone reports no+hash for those files.++Restricted filename characters++In addition to the default restricted characters set the following+characters are also replaced:++ Character Value Replacement+ ------------ ------- -------------+ LeftSpace 0x20 ␠+ RightSpace 0x20 ␠++Invalid UTF-8 bytes will also be replaced, as they can't be used in JSON+strings.++Overwriting files++When rclone overwrites an existing file, kDrive keeps the replaced+content as a new version of the file. Old versions count against the+drive quota, so repeatedly overwriting large files grows the storage+usage. Manage old versions from the kDrive web app.++Uploading files of unknown size++kDrive can't upload files of unknown size, so rclone rcat and streaming+uploads are not supported.++Deleting files++Deleted files will be moved to the trash. Your subscription level will+determine how long items stay in the trash. rclone cleanup can be used+to empty the trash.++Root folder ID++You can set the root_folder_id for rclone. This is the directory+(identified by its numeric Folder ID) that rclone considers to be the+root of your kDrive drive.++Normally you will leave this blank and rclone will use your private+directory. You can also use one of the convenience shortcuts:++- private: your user private directory (default)+- common: the kDrive common directory, shared among users++Or the numeric Folder ID of any directory in your drive. The root of the+drive only contains the Private and Common documents folders, so a+numeric ID is needed to use any other directory as root. In order to+find the Folder ID, list the parent directory of the folder you wish+rclone to use with rclone lsf; the first column of the output is the+folder ID:++Example:++ $ rclone lsf --dirs-only -Fip --csv remote:Documents+ 456,Sub folder/+ 789,Photos/++So if the folder you want rclone to use is "Photos/", then use the+returned id from rclone lsf command (ex. 789) as the root_folder_id+variable value in the config file.++Standard options++Here are the Standard options specific to kdrive (Infomaniak kDrive).++--kdrive-client-id++OAuth Client Id.++Leave blank normally.++Properties:++- Config: client_id+- Env Var: RCLONE_KDRIVE_CLIENT_ID+- Type: string+- Required: false++--kdrive-drive-id++ID of the kDrive to use.++When showing a folder on kdrive, you can find the drive_id here:+https://ksuite.infomaniak.com/{account_id}/kdrive/app/drive/{drive_id}/files/...++Properties:++- Config: drive_id+- Env Var: RCLONE_KDRIVE_DRIVE_ID+- Type: string+- Required: false++Advanced options++Here are the Advanced options specific to kdrive (Infomaniak kDrive).++--kdrive-token++OAuth Access Token as a JSON blob.++Properties:++- Config: token+- Env Var: RCLONE_KDRIVE_TOKEN+- Type: string+- Required: false++--kdrive-auth-url++Auth server URL.++Leave blank to use the provider defaults.++Properties:++- Config: auth_url+- Env Var: RCLONE_KDRIVE_AUTH_URL+- Type: string+- Required: false++--kdrive-token-url++Token server url.++Leave blank to use the provider defaults.++Properties:++- Config: token_url+- Env Var: RCLONE_KDRIVE_TOKEN_URL+- Type: string+- Required: false++--kdrive-encoding++The encoding for the backend.++See the encoding section in the overview for more info.++Properties:++- Config: encoding+- Env Var: RCLONE_KDRIVE_ENCODING+- Type: Encoding+- Default: Slash,Del,Ctl,LeftSpace,RightSpace,InvalidUtf8,Dot++--kdrive-root-folder-id++The folder to use as root.++You may use either one of the convenience shortcuts [private or common]+or an explicit folder ID.++- private: your user private directory+- common: the kDrive common directory, shared among users++Properties:++- Config: root_folder_id+- Env Var: RCLONE_KDRIVE_ROOT_FOLDER_ID+- Type: string+- Required: false++--kdrive-chunk-size++Files above this size are uploaded in chunks of this size.++The kDrive API accepts chunks between 1 MB and 1 GB, up to 10,000 chunks+per file and a total file size of 1 TiB. The chunk size is raised+automatically for very large files to stay within the 10,000 chunks+limit.++Properties:++- Config: chunk_size+- Env Var: RCLONE_KDRIVE_CHUNK_SIZE+- Type: SizeSuffix+- Default: 20Mi++--kdrive-upload-concurrency++Concurrency for chunked uploads.++Properties:++- Config: upload_concurrency+- Env Var: RCLONE_KDRIVE_UPLOAD_CONCURRENCY+- Type: int+- Default: 4++--kdrive-endpoint++The API endpoint to use.++Leave blank normally. There is no reason to change the endpoint except+for internal use or beta-testing.++Properties:++- Config: endpoint+- Env Var: RCLONE_KDRIVE_ENDPOINT+- Type: string+- Required: false++--kdrive-description++Description of the remote.++Properties:++- Config: description+- Env Var: RCLONE_KDRIVE_DESCRIPTION+- Type: string+- Required: false+KoofrPaths are specified as remote:path@@ -53624,6 +56585,14 @@ LimitationsInvalid UTF-8 bytes will also be replaced, as they can't be used in JSONstrings.+Linkbox converts file extensions to lower case on upload, so a file+uploaded as file.TXT is stored as file.txt.++Linkbox renames an uploaded file to name(1) if its name is the start of+the name of a file already in the same directory, for example when+uploading apple to a directory containing applebanana. rclone then can't+find the file it uploaded and the upload fails with object not found.+Mail.ru CloudMail.ru Cloud is a cloud storage provided by a Russian internet company@@ -54909,6 +57878,23 @@ and tags when --metadata is enabled (or when using --metadata-set /Notes: - Rclone ignores reserved x-ms-* keys (except x-ms-tags) for usermetadata.+Hierarchical namespace++Storage accounts with hierarchical namespace enabled (Azure Data Lake+Storage Gen2) have real directories. On these accounts rclone supports+empty directories, creating them with rclone mkdir and removing them+with rclone rmdir, whatever the --azureblob-directory-markers setting.++Note that deleting the last file in a directory leaves the directory+behind. Use rclone rmdirs or --delete-empty-src-dirs to remove empty+directories.++Rclone reads whether the storage account has a hierarchical namespace+when it first uses the remote and saves it as the hns option in the+config file. Set --azureblob-hns to override this.++Blob tags (x-ms-tags metadata) aren't supported on these accounts.+PerformanceWhen uploading large files, increasing the value of@@ -55132,14 +58118,14 @@ be explicitly specified using exactly one of the msi_object_id,msi_client_id, or msi_mi_res_id parameters.If none of msi_object_id, msi_client_id, or msi_mi_res_id is set, this-is is equivalent to using env_auth.+is equivalent to using env_auth.Federated Identity CredentialsIf these variables are set, rclone will authenticate with federatedidentity.-- tenant: tenant ID of of the storage+- tenant: tenant ID of the storage- client_id: client ID of the application the user will authenticateto storage- msi_client_id: managed identity client ID of the application the@@ -55656,6 +58642,52 @@ Properties:- Type: int- Default: 5000+--azureblob-use-arrow-list++Use the Apache Arrow listing format.++If set, directory listings are fetched using the ListBlobs Apache Arrow+response format instead of XML. Arrow responses are smaller and much+cheaper to parse, making listings of large containers several times+faster. Combine with "list_parallelism" for the biggest gains.++"Blob Listing with Apache Arrow" is in public preview at Microsoft and+is only supported on flat namespace accounts. On accounts with a+hierarchical namespace (ADLS Gen2), or where the feature is otherwise+unavailable, the server returns XML and the listing transparently falls+back to the normal XML path (logged at debug level).++Properties:++- Config: use_arrow_list+- Env Var: RCLONE_AZUREBLOB_USE_ARROW_LIST+- Type: bool+- Default: false++--azureblob-list-parallelism++Number of parallel shards to list a directory with.++If set greater than 1, the blob name keyspace of each directory is split+into ranges which are listed concurrently using the Arrow+startFrom/endBefore range parameters. This can dramatically speed up+listing containers with millions of objects, for both recursive (ListR)+and single directory listings. Speed keeps improving up to a parallelism+of around 30.++This has no effect unless "use_arrow_list" is also set, as Arrow is the+only listing path that supports server-side name ranges. If the account+does not support range listing (e.g. it has a hierarchical namespace)+the listing falls back to sequential. The default of 0 (or 1) lists+sequentially.++Properties:++- Config: list_parallelism+- Env Var: RCLONE_AZUREBLOB_LIST_PARALLELISM+- Type: int+- Default: 0+--azureblob-access-tierAccess tier of blob: hot, cool, cold or archive.@@ -55762,6 +58794,9 @@ creates an empty object ending with "/", to persist the folder.This object also has the metadata "hdi_isfolder = true" to conform tothe Microsoft standard.+This is not needed on storage accounts with a hierarchical namespace+(see the hns option) as these always support empty directories.+Properties:- Config: directory_markers@@ -55769,6 +58804,23 @@ Properties:- Type: bool- Default: false+--azureblob-hns++Set if the storage account has a hierarchical namespace.++Storage accounts with hierarchical namespace enabled (Azure Data Lake+Storage Gen2) have real directories which can be empty.++If this is unset then rclone will read it from the storage account and+save it in the config file.++Properties:++- Config: hns+- Env Var: RCLONE_AZUREBLOB_HNS+- Type: Tristate+- Default: unset+--azureblob-no-check-containerIf set, don't attempt to check the container exists or create it.@@ -56233,14 +59285,14 @@ be explicitly specified using exactly one of the msi_object_id,msi_client_id, or msi_mi_res_id parameters.If none of msi_object_id, msi_client_id, or msi_mi_res_id is set, this-is is equivalent to using env_auth.+is equivalent to using env_auth.Federated Identity CredentialsIf these variables are set, rclone will authenticate with federatedidentity.-- tenant: tenant ID of of the storage+- tenant: tenant ID of the storage- client_id: client ID of the application the user will authenticateto storage- msi_client_id: managed identity client ID of the application the@@ -56626,7 +59678,7 @@ Properties:Upload chunk size.Note that this is stored in memory and there may be up to-"--transfers" * "--azurefile-upload-concurrency" chunks stored at once+"--transfers" * "--azurefiles-upload-concurrency" chunks stored at oncein memory.Properties:@@ -56648,7 +59700,7 @@ and these uploads do not fully utilize your bandwidth, then increasingthis may help to speed up the transfers.Note that chunks are stored in memory and there may be up to-"--transfers" * "--azurefile-upload-concurrency" chunks stored at once+"--transfers" * "--azurefiles-upload-concurrency" chunks stored at oncein memory.Properties:@@ -56730,6 +59782,9 @@ Paths are specified as remote:pathPaths may be as deep as required, e.g. remote:directory/subdirectory.+Note: OneDrive for Business and SharePoint users need rclone v1.76 or+later from April 1, 2027 - see Pre-authenticated URLs.+ConfigurationThe initial setup for OneDrive involves getting a token from Microsoft@@ -56955,7 +60010,7 @@ Non-admin access to Business OneDriveIf you do not have admin access to your organization's OneDrive forBusiness, you can still connect by manually providing the SharePointtenant URL and drive ID. This works by overriding the base API URL from-the standard Microsoft Graph endpoint to the SharePoint v2.0 endpoint.+the standard Microsoft Graph endpoint to the SharePoint API endpoint.Steps to manually obtain credentials@@ -56971,6 +60026,12 @@ Steps to manually obtain credentials".driveUrl": "{tenant_url}/v2.0/drives/{drive_id}",".driveAccessToken": "access_token={access_token}"+ In some guest access scenarios SharePoint may also expose a matching+ v2.1 URL and token pair:++ ".driveUrlV21": "{tenant_url}/v2.1/drives/{drive_id}",+ ".driveAccessTokenV21": "access_token={access_token}"+Rclone configurationUse the extracted values to configure your remote:@@ -56981,6 +60042,12 @@ Use the extracted values to configure your remote:tenant_url = {tenant_url}drive_type = business+If you use a v2.1 token, set the matching SharePoint API version:++ tenant_api_version = v2.1++If tenant_api_version is omitted, rclone defaults to v2.0.+Since the exact expiry time cannot be determined from web traffic, setthe expiry to a future date. Note that the token will eventually expireand you will need to repeat the process to obtain a new one.@@ -57024,10 +60091,11 @@ This can be useful with rclone mount and rclone rc vfs/refreshrecursive=true) to very quickly fill the mount with information aboutall the files.-The API used for the recursive listing (ListR) only supports listing-from the root of the drive. This will become increasingly inefficient-the further away you get from the root as rclone will have to discard-files outside of the directory you are using.+Rclone asks for the delta listing of the directory being listed. Some+drives only support delta listings from the root of the drive, in which+case rclone lists the whole drive and discards the files outside of the+directory you are using, which is increasingly inefficient the further+away from the root you get.Some commands (like rclone lsf -R) will use ListR by default - you canturn this off with --disable ListR if you need to.@@ -57219,7 +60287,7 @@ Properties:The tenant URL for non-admin OneDrive access.-Set this to your SharePoint tenant URL to use the SharePoint v2.0 API+Set this to your SharePoint tenant URL to use the SharePoint APIendpoint instead of the standard Microsoft Graph API. This allowsaccessing business OneDrive without admin consent.@@ -57236,6 +60304,22 @@ Properties:- Type: string- Required: false+--onedrive-tenant-api-version++The SharePoint API version to use with tenant_url.++Set this to the SharePoint API version matching the browser-extracted+access token. For example, use v2.1 with a driveAccessTokenV21 token.++This only applies when tenant_url is set.++Properties:++- Config: tenant_api_version+- Env Var: RCLONE_ONEDRIVE_TENANT_API_VERSION+- Type: string+- Default: "v2.0"+--onedrive-chunk-sizeChunk size to upload files with - must be multiple of 320k (327,680@@ -57370,17 +60454,14 @@ Properties:Remove all versions on modifying operations.-Onedrive for business creates versions when rclone uploads new files-overwriting an existing one and when it sets the modification time.+Onedrive creates versions when rclone uploads new files overwriting an+existing one and when it sets the modification time.These versions take up space out of the quota.This flag checks for versions after file upload and setting modificationtime and removes all but the last version.-NB Onedrive personal can't currently delete versions so don't use this-flag there.-Properties:- Config: no_versions@@ -57395,8 +60476,8 @@ Permanently delete files on removal.Normally files will get sent to the recycle bin on deletion. Settingthis flag causes them to be permanently deleted. Use with care.-OneDrive personal accounts do not support the permanentDelete API, it-only applies to OneDrive for Business and SharePoint document libraries.+This works with OneDrive for Business, SharePoint document libraries,+and OneDrive personal accounts, including free accounts.Properties:@@ -57442,13 +60523,17 @@ Properties:- Creates a read-write link to the item.- "embed"- Creates an embeddable link to the item.+ - Only available in OneDrive personal.--onedrive-link-passwordSet the password for links created by the link command.-At the time of writing this only works with OneDrive personal paid-accounts.+At the time of writing this works with OneDrive for Business and+OneDrive personal paid accounts.++OneDrive personal free accounts can't set a password or an expiry time+(with --expire) on links.Properties:@@ -57542,20 +60627,11 @@ Setting this flag speeds up these things greatly:rclone size onedrive:rclone rc vfs/refresh recursive=true-However the delta listing API only works at the root of the drive. If-you use it not at the root then it recurses from the root and discards-all the data that is not under the directory you asked for. So it will-be correct but may not be very efficient.--This is why this flag is not set as the default.--As a rule of thumb if nearly all of your data is under rclone's root-directory (the root/directory in onedrive:root/directory) then using-this flag will be a big performance win. If your data is mostly not-under the root then using this flag will be a big performance loss.--It is recommended if you are mounting your onedrive at the root (or near-the root when using crypt) and using rclone rc vfs/refresh.+Rclone asks for the delta listing of the directory being listed. If the+drive only supports delta listings at the root of the drive (as some+older OneDrive for Business and SharePoint drives do) then rclone lists+from the root and discards all the data that is not under the directory+you asked for. So it will be correct but may not be very efficient.Properties:@@ -57750,8 +60826,7 @@ Here are the possible system metadata items for the onedrive backend.Name Help Type Example Read Only------------------------------- ---------------------------------- ----------- -------------------------------------- --------------------btime Time of file birth (creation) with RFC 3339 2006-01-02T15:04:05Z N- S accuracy (mS for OneDrive- Personal).+ S accuracy.content-type The MIME type of the file. string text/plain Y@@ -57778,8 +60853,7 @@ Here are the possible system metadata items for the onedrive backend.the item contains malware.mtime Time of last modification with S RFC 3339 2006-01-02T15:04:05Z N- accuracy (mS for OneDrive- Personal).+ accuracy.package-type If present, indicates that this string oneNote Yitem is a package instead of a@@ -57806,11 +60880,9 @@ Here are the possible system metadata items for the onedrive backend.organization, or users.shared-time Time when the item was shared, RFC 3339 2006-01-02T15:04:05Z Y- with S accuracy (mS for OneDrive- Personal).+ with S accuracy.- utime Time of upload with S accuracy (mS RFC 3339 2006-01-02T15:04:05Z Y- for OneDrive Personal).+ utime Time of upload with S accuracy. RFC 3339 2006-01-02T15:04:05Z Y------------------------------------------------------------------------------------------------------------------------------------------See the metadata docs for more info.@@ -57906,11 +60978,6 @@ Or you can set the no_versions parameter to true and rclone will removeversions after operations which create new versions. This takes extratransactions so only enable it if you need it.-Note At the time of writing Onedrive Personal creates versions (but not-for setting the modification time) but the API for removing them returns-"API not found" so cleanup and no_versions should not be used on-Onedrive Personal.-Disabling versioningStarting October 2018, users will no longer be able to disable@@ -57961,10 +61028,36 @@ or --dry-run which is a great way to see what it would do.rclone cleanup --interactive remote:path/subdir # interactively remove all old version for path/subdirrclone cleanup remote:path/subdir # unconditionally remove all old version for path/subdir-NB Onedrive personal can't currently delete versions-Troubleshooting+Pre-authenticated URLs++Microsoft is retiring pre-authenticated (tempauth) URLs in OneDrive for+Business and SharePoint from April 1, 2027.++Upgrade to rclone v1.76 or later before April 1, 2027. Earlier versions+rely on pre-authenticated URLs, so after that date their downloads,+chunked uploads and server-side copies will fail with OneDrive for+Business and SharePoint. Rclone v1.76 and later keep working with the+existing configuration, apart from the client credentials flow with a+client secret (see below).++Rclone uses these URLs when downloading files, when uploading files+bigger than --onedrive-upload-cutoff in chunks and when doing+server-side copies. When SharePoint returns them without+pre-authentication rclone authenticates them with a SharePoint token+which it gets using the existing login, so no reconfiguration is needed.+OneDrive Personal is not affected.++If rclone can't get a SharePoint token, or SharePoint refuses it, you+will see the error SharePoint refused to authenticate a request. In this+case rclone uploads files smaller than 250 MiB in a single request+instead and copies files by downloading and uploading them, but bigger+files can't be uploaded and downloads may fail.++SharePoint does not accept tokens from the client credentials flow which+were got with a client secret, so this affects client_credentials users.+Excessive throttling or blocked on SharePointIf you experience excessive throttling or is being blocked on SharePoint@@ -58020,7 +61113,8 @@ account. You can't do much about it, maybe write an email to youradmins.However, there are other ways to interact with your OneDrive account.-Have a look at the WebDAV backend: https://rclone.org/webdav/#sharepoint+Have a look at the WebDAV backend:+https://rclone.org/webdav/#sharepoint-onlineinvalid_grant (AADSTS50076)@@ -58056,6 +61150,25 @@ a workaround:3. The shortcut will appear in My files, you can access it with rclone,it behaves like a normal folder/file. [in_my_files] [rclone_mount]+Listing such a shortcut may still fail with++ invalidRequest: The provided drive id appears to be malformed, or does not represent a valid drive.++as the drive it points to can't always be read with the drive ID the API+supplies. Personal Vault is a shortcut of the same kind and fails to+list in the same way, though it reports+invalidResourceId: ObjectHandle is Invalid.++A normal listing reports the error for that folder and carries on, so+the rest of the drive is still listed, and excluding the folder (for+example with --exclude "/Personal Vault/**") avoids the error.++A recursive listing with --fast-list stops at the first error, so+listing the root of a drive which has one of these shortcuts in it will+fail. Filters don't help here as the folder is listed by the backend+before the filters are applied - use --disable ListR (or don't use+--fast-list) on such a drive.+Live Photos uploaded from iOS (small video clips in .heic files)The iOS OneDrive app introduced upload and storage of Live Photos in@@ -58899,7 +62012,7 @@ Properties:- Config: encoding- Env Var: RCLONE_OOS_ENCODING- Type: Encoding-- Default: Slash,InvalidUtf8,Dot+- Default: Slash,CrLf,InvalidUtf8,Dot--oos-leave-parts-on-error@@ -60069,6 +63182,7 @@ that being:- Oracle Cloud Storage- Blomp Cloud Storage- IBM Bluemix Cloud ObjectStorage Swift+- Infomaniak Swiss BackupPaths are specified as remote:container (or remote: for the lsdcommand.) You may put subdirectories in too, e.g.@@ -60194,8 +63308,8 @@ files in the container.Configuration from an OpenStack credentials file-An OpenStack credentials file typically looks something something like-this (without the comments)+An OpenStack credentials file typically looks something like this+(without the comments)export OS_AUTH_URL=https://a.provider.net/v2.0export OS_TENANT_ID=ffffffffffffffffffffffffffffffff@@ -60364,6 +63478,14 @@ Properties:- OVH- "https://authenticate.ain.net"- Blomp Cloud Storage+ - "https://swiss-backup01.infomaniak.com/identity/v3"+ - Infomaniak Swiss Backup Location 01+ - "https://swiss-backup02.infomaniak.com/identity/v3"+ - Infomaniak Swiss Backup Location 02+ - "https://swiss-backup03.infomaniak.com/identity/v3"+ - Infomaniak Swiss Backup Location 03+ - "https://swiss-backup04.infomaniak.com/identity/v3"+ - Infomaniak Swiss Backup Location 04--swift-user-id@@ -60744,7 +63866,7 @@ Request" error rather than a more sensible error when the authenticationfails for Swift.So this most likely means your username / password is wrong. You can-investigate further with the --dump-bodies flag.+investigate further with the --dump bodies flag.This may also be caused by specifying the region when you shouldn't have(e.g. OVH).@@ -61321,6 +64443,8 @@ Properties:Files bigger than this will be cached on disk to calculate hash ifrequired.+Files of unknown size are always cached on disk.+Properties:- Config: hash_memory_limit@@ -63028,6 +66152,12 @@ characters are also replaced:Invalid UTF-8 bytes will also be replaced, as they can't be used in JSONstrings.+Seafile stores file and directory names in Unicode NFC form whatever+form they are uploaded in, so rclone normalizes names to NFC before+sending them. Listings therefore always show NFC names, even for files+created with decomposed (NFD) names as macOS produces them. Library+names are not normalized as Seafile keeps them exactly as entered.+Seafile and rclone linkRclone supports generating share links for non-encrypted libraries only.@@ -63649,6 +66779,22 @@ mod_sftp). If you are using one of these servers, you can set the optionset_modtime = false in your RClone backend configuration to disable thisbehaviour.+Restricted filename characters++SFTP servers on Unix-like systems can store any character in a file name+except NUL and /, so by default only these are replaced:++ Character Value Replacement+ ----------- ------- -------------+ NUL 0x00 ␀+ / 0x2F /++If the server runs on Windows, which can't store many more characters,+add the Windows restricted characters with --sftp-encoding, as in the+FTP example:++ --sftp-encoding "Slash,LtGt,DoubleQuote,Colon,Question,Asterisk,Pipe,BackSlash,Ctl,RightSpace,RightPeriod,InvalidUtf8,Dot"+About commandThe about command returns the total space, free space, and used space on@@ -64041,7 +67187,7 @@ Properties:- Config: encoding- Env Var: RCLONE_SFTP_ENCODING- Type: Encoding-- Default: Slash,Del,Ctl,Dot+- Default: Slash,Dot--sftp-set-modtime@@ -64288,6 +67434,33 @@ Properties:- Type: bool- Default: false+--sftp-multithread-upload++Set this to use multi-thread (multi-connection) uploads.++Normally rclone uploads a single large file over one SFTP connection.+With this set rclone uploads large files over several SFTP connections+at once (one per chunk), which greatly improves throughput on high+latency links, the same way multi-thread downloads already work. Each+connection writes a different, non-overlapping byte range of the file,+so this requires a server that allows several handles to the same file+with writes at arbitrary offsets, which OpenSSH does.++This is off by default because many SFTP servers only accept sequential+writes (for example some object storage backed SFTP gateways) and would+fail large uploads with "truncate failed" or "invalid offset" style+errors.++This is ignored (uploads stay single-threaded) when+--sftp-disable-concurrent-writes or --sftp-connections is in use.++Properties:++- Config: multithread_upload+- Env Var: RCLONE_SFTP_MULTITHREAD_UPLOAD+- Type: bool+- Default: false+--sftp-idle-timeoutMax time before closing idle connections.@@ -64569,7 +67742,7 @@ found in this paper.SFTP isn't supported under plan9 until this issue is fixed.Note that since SFTP isn't HTTP based the following flags don't work-with it: --dump-headers, --dump-bodies, --dump-auth.+with it: --dump headers, --dump bodies, --dump auth.Note that --timeout and --contimeout are both supported.@@ -64980,12 +68153,14 @@ Properties:SMB username.+Leave blank to use the current user name.+Properties:- Config: user- Env Var: RCLONE_SMB_USER- Type: string-- Default: "$USER"+- Required: false--smb-port@@ -65061,6 +68236,23 @@ Advanced optionsHere are the Advanced options specific to smb (SMB / CIFS).+--smb-workstation++Workstation name sent in NTLM authentication.++The NetBIOS name sent as the client workstation in the NTLM AUTHENTICATE+message. Some servers (e.g. domain-joined Windows servers with+per-account workstation restrictions or NTLM allow-lists) require this+to match a specific name. Leave blank to send no workstation name (the+default).++Properties:++- Config: workstation+- Env Var: RCLONE_SMB_WORKSTATION+- Type: string+- Required: false+--smb-idle-timeoutMax time before closing idle connections.@@ -66626,7 +69818,9 @@ Likewise plain WebDAV does not support hashes, however when used withFastmail Files, ownCloud or Nextcloud rclone will support SHA1 and MD5hashes. Depending on the exact version of ownCloud or Nextcloud hashesmay appear on all objects, or only on objects which had a hash uploaded-with them.+with them. With Nextcloud, rclone asks the server to calculate the SHA1+of uploads which had no hash to send, such as streamed uploads, and+after setting the modification time, which discards the stored hash.Standard options@@ -66760,6 +69954,13 @@ For example, to set a Cookie use 'Cookie,name=value', orYou can set multiple headers, e.g.'"Cookie","name=value","Authorization","xxx"'.+The headers are only sent to the host in the configured URL. If the+server redirects to another host (including a subdomain or a different+port) the headers are not sent to it, or to any further hop in that+redirect chain. The exception is the directory listing and file download+requests when auth_redirect is set, as that keeps all credentials across+redirects.+Properties:- Config: headers@@ -67165,6 +70366,17 @@ The default restricted characters set are replaced.Invalid UTF-8 bytes will also be replaced, as they can't be used in JSONstrings.+Application folder++If you registered your own OAuth application with Yandex and only+granted it the "Application Folder" permission+(cloud_api:disk.app_folder) instead of full disk access, then you must+set the app_folder option to true. Yandex Disk requires paths to be+addressed with an app:/ prefix instead of the usual disk:/ prefix when+using this restricted scope, and rclone will return a 403 Forbidden+error on every request if this option isn't set correctly for the token+in use.+Standard optionsHere are the Standard options specific to yandex (Yandex Disk).@@ -67309,6 +70521,26 @@ Properties:- Type: Duration- Default: 0s+--yandex-app-folder++Use the application folder as the root.++If you registered your own OAuth application with Yandex with only the+"Application Folder" permission (cloud_api:disk.app_folder) instead of+full disk access, then rclone needs to address paths with the "app:/"+prefix instead of the usual "disk:/" prefix, otherwise all requests fail+with a 403 Forbidden error.++Set this to true if your OAuth token only has access to the application+folder.++Properties:++- Config: app_folder+- Env Var: RCLONE_YANDEX_APP_FOLDER+- Type: bool+- Default: false+--yandex-descriptionDescription of the remote.@@ -67961,6 +71193,9 @@ and6 b/two6 b/one+Symlinks which point to a parent directory would loop forever, so rclone+reports them as errors and skips them.+--local-links, --links, -lNormally rclone will ignore symlinks or junction points (which behave@@ -68009,6 +71244,9 @@ Copying them back with '-l'├── file1 -> ./file4└── file2 -> /home/user/file3+A .rclonelink file whose contents are too long to be a path (more than+128 KiB) is refused rather than turned into a symlink.+However, if copied back without '-l'$ rclone copyto remote:/tmp/a/ /tmp/b/@@ -68308,6 +71546,14 @@ consume more storage than having just one.) However, for use cases wheredata redundancy is preferable, --local-no-clone can be used to disablecloning and force "deep" copies.+Cloning also preserves macOS-specific metadata such as extended+attributes, which a normal copy does not, so this metadata is lost when+cloning is disabled.++If the destination file already exists, cloning replaces it with a new+file, even with --inplace, so hard links to it are broken. Use+--local-no-clone to update existing files in place.+Currently, cloning is only supported when using APFS on macOS (supportfor other platforms may be added in the future.)@@ -68397,10 +71643,10 @@ Properties:Make out-of-space errors fatal during transfers.-When enabled, an ENOSPC error during a write returns a fatal error so-that rclone aborts rather than retrying the operation. Useful for backup-scripts that should halt loudly on a full disk rather than spin-retrying.+When enabled, an out of space error while writing, creating a directory,+or creating a file returns a fatal error so that rclone aborts rather+than retrying the operation. Useful for backup scripts that should halt+loudly on a full disk rather than spin retrying.Properties:@@ -68498,9 +71744,10 @@ Metadata is supported on files and directories.When restoring metadata with --metadata rclone applies the "mode", "uid"and "gid" from the source. These come from the source remote which maynot be trusted, so restoring metadata as root from an untrusted source-can change file ownership and is not recommended. The setuid, setgid and-sticky bits are not restored by default - see the---local-metadata-restore-special-bits flag.+can change file ownership and is not recommended. A user other than root+keeps its own uid and sets the gid only to a group it is a member of, as+rsync does. The setuid, setgid and sticky bits are not restored by+default - see the --local-metadata-restore-special-bits flag.Here are the possible system metadata items for the local backend.@@ -68564,6 +71811,1018 @@ Options:Changelog+v1.76.0 - 2026-11-06++See commits++- New backends+ - Dosya (Kaya)+ - Infomaniak kDrive (Guillaume MALLERET)+- New S3 Providers+ - Autonomys Auto Drive (Jim Counter)+ - Clever Cloud Cellar (David Legrand)+ - Infomaniak Swiss Backup (Louis)+ - Neon Object Storage (Rishi Raj Jain)+ - RelAix.Storage (mygrexit)+- New commands+ - index: Write static directory listings into a remote (Nick+ Craig-Wood)+- Security+ - build+ - Disable Multipath TCP listeners by default to mitigate+ CVE-2026-98432 (Nick Craig-Wood)+ - Fix multiple CVEs including CVE-2026-111943 by upgrading to+ go1.27.3 (Nick Craig-Wood)+ - Update golang.org/x/net to v0.62.0 to fix multiple CVEs+ (Nick Craig-Wood)+ - Update github.com/andybalholm/brotli to v1.2.7 to fix+ CVE-2026-111497 (Nick Craig-Wood)+ - fshttp: Refuse HTTPS to HTTP redirects unless+ --allow-https-downgrade GHSA-m97j-7x65-wh93 CVE-PENDING (Nick+ Craig-Wood)+ - lib/http: Stop --htpasswd and --user revealing which usernames+ exist through timing GHSA-hwj7-wp9g-wrv9 CVE-PENDING (Nick+ Craig-Wood)+ - local+ - Fix user metadata escaping the root through a planted+ symlink GHSA-r5wv-vqqh-2r9c CVE-PENDING (Nick Craig-Wood)+ - Fix MkdirMetadata finding directories outside the root+ through a planted symlink (Nick Craig-Wood)+ - Fix Remove and Rmdir deleting outside the root through a+ planted symlink (Nick Craig-Wood)+ - Fix DirMove creating the destination parent through a+ planted symlink (Nick Craig-Wood)+ - netstorage: Fix API auth headers leaking to other hosts on+ redirect GHSA-m97j-7x65-wh93 CVE-PENDING (Nick Craig-Wood)+ - rc+ - Limit request size when the rc server has no authentication+ GHSA-f8p4-8g4j-q3m2 CVE-PENDING (Nick Craig-Wood)+ - Reject unauthorised and unknown calls before reading the+ request GHSA-f8p4-8g4j-q3m2 CVE-PENDING (Nick Craig-Wood)+ - seafile: Fix HTTPS to HTTP redirects leaking credentials+ GHSA-m97j-7x65-wh93 CVE-PENDING (Nick Craig-Wood)+ - serve: Make auth proxy revocation take effect within 5 minutes+ GHSA-vmgx-g9c5-669p CVE-PENDING (Nick Craig-Wood)+ - serve sftp+ - Fix auth proxy letting a public key be used as a password+ GHSA-66qv-66mg-9h74 CVE-PENDING (Nick Craig-Wood)+ - Fix auth proxy running for public keys the client can't use+ GHSA-66qv-66mg-9h74 CVE-PENDING (Nick Craig-Wood)+ - swift+ - Fix auth token leaking to other hosts or over plaintext on+ redirect GHSA-m97j-7x65-wh93 CVE-PENDING (Nick Craig-Wood)+ - Fix password being resent to another host on auth redirect+ GHSA-m97j-7x65-wh93 CVE-PENDING (Nick Craig-Wood)+ - ulozto: Fix auth tokens leaking on redirects from download links+ GHSA-3xpx-hjg2-vrjh CVE-PENDING (Nick Craig-Wood)+- New Features+ - accounting+ - Stop --max-transfer overshooting the limit (Nick Craig-Wood)+ - Add "Updated dirs" count of directory modtime/metadata+ updates to the stats (Nick Craig-Wood)+ - batcher: Log when committing uploads has finished (phatlc)+ - bisync+ - Make --max-delete aware of --track-renames (Morax)+ - Fall back to --download when a checksum is missing (nielash)+ - Reject --max-delete above 100 (nielash)+ - build+ - Update to go1.27 and make go1.26 the minimum required+ version (Nick Craig-Wood)+ - Update all dependencies (Nick Craig-Wood)+ - cmd: Complete enum flag values from their choices (Demiao Huang)+ - config: Only run --password-command once when using --daemon+ (phatlc)+ - doc fixes (ashwinibailangadi, ferrumclaudepilgrim, frodleTian,+ FTCHD, gomitrah, Nick Craig-Wood, qodelicious, Satvik Mishra,+ ZainnQureshii)+ - docker: Make container user UID and GID configurable via build+ ARGs (Vaibhav Mashal)+ - filter+ - Speed up large --include-from lists of directories (kaii9)+ - Make [!...] in globs a negated character class (kevin9327)+ - Make [^...] in globs exclude / and document it (kevin9327)+ - fs+ - Reject sizes too large for SizeSuffix instead of silently+ disabling the limit (KBS)+ - Stop writing config for on the fly and environment variable+ remotes into the config file (Nick Craig-Wood)+ - Reject durations like 300y that overflow instead of wrapping+ (kevin9327)+ - Accept multi-unit relative times like 1d12h in Time options+ (kevin)+ - Reject NaN sizes and CountSuffix values that overflow+ (kevin9327)+ - Reject --bwlimit times without a colon or with signed digits+ (kevin9327)+ - Reject NaN durations instead of wrapping (Soujas K)+ - Add MIME type for JPEG XL (.jxl) files (Seth Westphal)+ - gui+ - Take the GUI out of beta (Nick Craig-Wood)+ - Update embedded release to 1.2.1 (Nick Craig-Wood)+ - hash: Add CRC-32C (Castagnoli) support (Dirk Petersen)+ - index: Add partial runs which only re-index changed directories+ (Nick Craig-Wood)+ - lib/http+ - Return 416 for ranges --rc-serve and serve restic can't+ satisfy (Rohit Behera)+ - Replace github.com/abbot/go-http-auth with built in htpasswd+ support (Nick Craig-Wood)+ - Accept the Basic auth scheme in any case (kevin)+ - lib/proxy: Send the decoded user and password in+ Proxy-Authorization (kevin)+ - lib/transform: Make titlecase capitalise words instead of+ upper-casing them (kevin)+ - ncdu: Show progress and the item count when deleting a selection+ (Lev Devaev)+ - operations+ - Check checksums in rcat with known size (phatlc)+ - Speed up checksum by hashing files in parallel (Teddy+ Tennant)+ - Fall back to a single stream when a multi-thread source+ ignores ranges (Morax)+ - Show directory operations in the --progress display (Nick+ Craig-Wood)+ - Add operations/getfile remote control API endpoint (Hakan+ İSMAİL)+ - Make --immutable work with copyto, moveto and single file+ copy (phatlc)+ - Respect --backup-dir in deletefile (Cao Yuhang)+ - rc+ - Deprecate --rc-web-gui flag (Nick Craig-Wood)+ - Add sync report parameters to sync/sync, sync/copy and+ sync/move (Nick Craig-Wood)+ - rest: Reject negative sizes in HTTP response headers (Yousef+ Issa)+ - serve http+ - Sort directory listings in the browser (Nick Craig-Wood)+ - Redesign the directory listing page (Nick Craig-Wood)+ - serve s3+ - Report the object's modtime as Last-Modified after a PUT+ (kevin9327)+ - Support changing object tags (Nick Craig-Wood)+ - Quote the ETags of objects in listings (Nick Craig-Wood)+ - Support listing with delimiters other than "/" (Nick+ Craig-Wood)+ - Support directory markers and refuse file/directory+ conflicts (Nick Craig-Wood)+ - Make the ETag of multipart uploads agree with HEAD and+ support partNumber (Nick Craig-Wood)+ - Make conditional writes atomic against other writes through+ the server (Nick Craig-Wood)+ - Make the ETag returned on upload agree with HEAD for any+ --etag-hash (Nick Craig-Wood)+ - List empty directories as directory markers (Nick+ Craig-Wood)+ - Allow copying directory markers (Nick Craig-Wood)+ - Remove the directories made for a failed upload (Nick+ Craig-Wood)+ - Keep the MD5 ETag of objects on remotes without MD5 (Nick+ Craig-Wood)+ - Don't read directories needlessly when listing with a+ delimiter (Nick Craig-Wood)+ - Don't compute the MD5 of multipart uploads unless the ETag+ is MD5 (Nick Craig-Wood)+ - List the directory markers put for directories holding other+ keys (Nick Craig-Wood)+ - Let a failed CompleteMultipartUpload be retried (Nick+ Craig-Wood)+ - List objects with the storage class they were put with (Nick+ Craig-Wood)+ - Add --directory-markers to list every directory as a marker+ (Nick Craig-Wood)+ - Remove a directory with its last key after its marker is+ deleted (Nick Craig-Wood)+ - Make --no-cleanup a deprecated alias of --directory-markers+ (Nick Craig-Wood)+ - Document which S3 features are supported and the design+ goals (Nick Craig-Wood)+ - serve webdav+ - Sort directory listings in the browser (Nick Craig-Wood)+ - Redesign the directory listing page (Nick Craig-Wood)+ - size: Include source in JSON output (Jiacheng Xu)+ - sync: Add pattern-based transfer ordering (Shurong Cao)+ - touch: Parse documented timestamp formats (Nicholas Velten)+- Bug Fixes+ - accounting: Fix totalBytes counting server-side copies and moves+ twice (ilya lavrenov)+ - bisync+ - Fix conflict winner missing from listings when Path2 wins+ (nielash)+ - Fix --conflict-resolve comparing a zero size or modtime+ (nielash)+ - Fix endless retry loop for skipped files (nielash)+ - Fix listings ignoring a modtime less than 1s older (nielash)+ - Fix rc ignoring global --dry-run and --max-delete settings+ (nielash)+ - Fix rc ignoring legacy backupdir1/2 and invalid values+ (nielash)+ - Fix internal error when a conflict rename reuses deleted+ names (TastyHeadphones)+ - fs: Fix durations like 1d12h being rejected (Rohit Behera)+ - lib/encoder: Fix decoding a trailing character after an escaped+ quote (kevin9327)+ - list: Fix ext sort error sometimes being ignored at the end of a+ listing (Nick Craig-Wood)+ - ncdu: Update to tcell v3 which fixes duplicated keys on Windows+ (Nick Craig-Wood)+ - operations+ - Fix making a directory with modtime asking twice with+ --interactive (Nick Craig-Wood)+ - Fix case- or Unicode-only renames deleting the file+ (nielash)+ - Fix moving a file onto its own name deleting it (nielash)+ - Fix cat --offset before the start of a file printing its end+ (b84330808)+ - Fix oversized getfile tail requests (唐僧洗发用飘柔)+ - rc: Fix LogLevel in _config being ignored (Nick Craig-Wood)+ - rest: Fix retries reading the request body after the request has+ finished (Nick Craig-Wood)+ - serve dlna: Fix crash on Browse with a negative StartingIndex or+ RequestedCount (phatlc)+ - serve restic: Fix temporary errors confusing restic into leaving+ locks behind (Azhar)+ - serve s3+ - Update to gofakes3 v0.0.8 which fixes 500 errors with go1.27+ (Nick Craig-Wood)+ - Upgrade gofakes3 to v0.0.11 to fix IsTruncated and Docker+ Distribution Registry (DEVN)+ - Fix buckets being undeletable after an aborted multipart+ upload (Nick Craig-Wood)+ - Fix CopyObject metadata including the source's and an mtime+ (Nick Craig-Wood)+ - serve sftp: Fix rmdir deleting a file instead of returning an+ error (phatlc)+ - sync+ - Fix directory modtimes being set twice (Nick Craig-Wood)+ - Fix --delete-empty-src-dirs disabled by a prior error in a+ reused stats group (ghwany)+ - walk: Fix duplicate directories with --fast-list when+ directories are listed after their contents (Nick Craig-Wood)+- Mount+ - Fix VFS instance leaks on mount failures and unmounts (Hakan+ İSMAİL)+- VFS+ - Preserve shrinking truncate on uncached files (47th)+ - Fix crash renaming a directory while the dir cache expires+ (Shreyansh Jain)+ - Fix virtual entries leak in removed directory (Biagio+ Peccerillo)+ - Fix renamed items listed twice after a case or Unicode-only+ match (nielash)+ - Fix uploads never starting while a dirty file keeps being read+ (Neil Cawse)+- Local+ - Speed up default checksummed copies by writing in larger blocks+ (Dominik Sander)+ - Make out of space errors fatal during multi-thread transfers+ (ferrumclaudepilgrim)+ - Stop --copy-links following symlink loops (phatlc)+ - Limit the size of symlink targets read from .rclonelink files+ (Nick Craig-Wood)+ - Fix server-side copy on macOS when the destination exists (Ayush+ Sharma)+ - Set only the ownership a non-root user may set (Dirk Petersen)+- Crypt+ - Fix nil pointer panic in ObjectInfo.Hash (fixfon)+- Archive+ - Read symlinks stored in zip and squashfs archives (Lev Devaev)+ - Fix reading an archive in the root of a remote (Oshino Penguin)+- Azure Blob+ - Enable on Solaris (acoeur)+ - Fix directories on storage accounts with hierarchical namespace+ (Nick Craig-Wood)+ - Fix NewObject on a directory returning "is not a regular file"+ (Nick Craig-Wood)+ - Fix listing a file as a directory on hierarchical namespace+ accounts (Nick Craig-Wood)+ - Fix sync at the root failing on soft deleted containers (Ludovic+ Bondon)+- Box+ - Reuse multipart upload part buffers via the global pool (Nick+ Craig-Wood)+ - Fix ~ and environment variables not being expanded in+ box_config_file (Nick Craig-Wood)+- Chunker+ - Clean stale chunks after server-side overwrite+ (hahahahahayesyeseys)+- Compress+ - Buffer small uploads in the global memory pool (Nick Craig-Wood)+ - Stop buffering the compressibility check output in memory (Nick+ Craig-Wood)+- Drime+ - Fix endless listing of directories with more than 20,000 entries+ (Maksym Tymoshyk)+- Drive+ - Fetch metadata permissions concurrently (Loi Nguyen)+ - Reuse resumable-upload chunk buffers via the global memory pool+ (SillyZir)+ - Stop a fast-list error from hanging the sync (Sash)+- Dropbox+ - Reuse upload chunk buffers via the global memory pool (Nick+ Craig-Wood)+- Filelu+ - Reuse multipart upload buffers via the global pool and retry+ failed parts (Nick Craig-Wood)+ - Encode multipart init paths (wjc)+- FTP+ - Discard connections that fail with non-FTP errors (Daniel Bae)+ - Fix ranged reads failing when the server answers 425 to an early+ close (Robert Kubiś)+- Hasher+ - Keep cached hash when only the modtime changes (Aniket)+- Hidrive+ - Stop copying every upload chunk a second time (Nick Craig-Wood)+ - Buffer the start of each upload from the global memory pool+ (Nick Craig-Wood)+ - Reuse chunked upload buffers via the global memory pool (Nick+ Craig-Wood)+ - Buffer small file updates from the global memory pool (Nick+ Craig-Wood)+ - Fix truncated file left behind when an upload hits+ --max-transfer (Nick Craig-Wood)+- HTTP+ - Add metadata defaults for --http-no-head (Giridhar)+- Huaweidrive+ - Build small uploads in pooled memory instead of a growing buffer+ (Nick Craig-Wood)+ - Reuse resumable-upload chunk buffers via the global pool (Nick+ Craig-Wood)+ - Fix creating, copying and renaming files with non-ASCII names+ (Nick Craig-Wood)+- Iclouddrive+ - Mark the backend as case insensitive (berce)+- Imagekit+ - Stop multipart writers after failed uploads (Florian Josef+ Altendorfer)+- Jottacloud+ - Buffer MD5 pre-reads in the global memory pool (Nick Craig-Wood)+ - Fix server-side copy onto a trashed file failing as corrupted+ (Nick Craig-Wood)+ - Fix server-side copy onto a trashed file keeping the old name+ case (Nick Craig-Wood)+ - Fix authentication with whitelabel services elkjop and elko+ (albertony)+ - Fix traditional authentication failing with invalid_client+ (albertony)+- Linkbox+ - Buffer the hashed 10 MiB file prefix in the global memory pool+ (Nick Craig-Wood)+- Mailru+ - Buffer speedup hashing in the global memory pool and fix upload+ retries (Nick Craig-Wood)+- Onedrive+ - OneDrive for Business and SharePoint users must upgrade to+ v1.76+ before April 1, 2027 (Nick Craig-Wood)+ - Authenticate SharePoint downloads, uploads and copies after+ tempauth retirement (Nick Craig-Wood)+ - Return an error if a ranged download doesn't return the+ requested range (Nick Craig-Wood)+ - Reuse upload chunk buffers via the global memory pool (Nick+ Craig-Wood)+ - Allow --onedrive-upload-cutoff up to the documented limit of 250+ MiB (Can Arslan)+ - Add configurable SharePoint tenant API version (Mattias Michaux)+ - Make --onedrive-delta quicker when not listing from the drive+ root (Nick Craig-Wood)+ - Keep the delta token when a ChangeNotify poll fails+ (xiaozou-wine)+ - Show the size of directories in listings (GingerAdonis)+ - Poll async copy status with the configured HTTP client (Soujas+ K)+- Opendrive+ - Reuse upload chunk buffers via the global memory pool (Nick+ Craig-Wood)+- Oracleobjectstorage+ - Add server-side object moves (tomaszni)+ - Encode carriage returns and line feeds (tomaszni)+ - Purge objects in batches (tomaszni)+ - Retry 409 ConcurrentObjectUpdate errors (berce)+- Pcloud+ - Stop multipart writers after failed uploads (Florian Josef+ Altendorfer)+- Pikpak+ - Share upload chunk buffers with the global memory pool (Nick+ Craig-Wood)+ - Use pooled memory for the gcid hash buffer and spool+ unknown-size uploads to disk (Nick Craig-Wood)+ - Fix downloads that return an HTTP error page as the file+ (Cestercian)+ - Stop multipart writers after failed uploads (Florian Josef+ Altendorfer)+ - Fix uploads failing when the server is slow to complete them+ (Nick Craig-Wood)+- Putio+ - Reuse upload chunk buffers via the global memory pool (Nick+ Craig-Wood)+- S3+ - Avoid buffering Object Lock uploads when the source MD5 is known+ (Nick Craig-Wood)+ - Add backend link command with signed response header overrides+ (eliotee)+ - Add Fastly EU Central 1 region (Leon Brocard)+ - Add the MD5 of multipart chunks to the debug log+ (interested.tortoise)+ - Report the real storage class when an object needs restoring+ (Dirk Petersen)+ - Speed up multi-thread uploads by checksumming chunks in 1 MiB+ reads (Roman Butor)+ - Add IT-MIL region to Scaleway (Alessio Margelli)+ - Add missing Huawei OBS regions (Enbiya Göral)+ - Fix rmdir of the root directory with --s3-directory-markers+ (Nick Craig-Wood)+ - Fix presigned URLs for GCS, Linode and Ceph (Leon Brocard)+- Seafile+ - Fix operations on files whose names are not in Unicode NFC form+ (Fred)+ - Stop multipart writers after failed uploads (Florian Josef+ Altendorfer)+ - Fix double slash in the URL when getting a 2FA token (Nick+ Craig-Wood)+- SFTP+ - Implement multi-thread uploads (Splainte)+ - Fix data race when using --sftp-ssh (Dirk Petersen)+ - Fix hash check for names starting with "-" with relative+ --sftp-path-override (Dirk Einecke)+ - Fix sync deleting files with control characters in their names+ (nielash)+ - Fix hashing files with a newline in their name on rsync.net+ (Nick Craig-Wood)+- Shade+ - Stop copying each upload chunk into a second heap buffer (Nick+ Craig-Wood)+- Smb+ - Add workstation field to smb backend (Kalin Stoyanov)+- Swift+ - Add Infomaniak Swiss Backup provider (Louis)+- WebDAV+ - Reject ranged responses which ignore the range (Morax)+ - Retry Nextcloud chunk uploads individually (Stochastic)+ - Fix Nextcloud public share access by sending X-Requested-With+ (drakeo338)+ - Fix memory leak uploading to Infinite Scale (jxj)+- Yandex+ - Add app_folder option to support cloud_api:disk.app_folder OAuth+ scope (Sanjay Kanth A)+- Zoho+ - Stop multipart writers after failed uploads (Florian Josef+ Altendorfer)++v1.75.2 - 2026-10-09++See commits++- Security+ - b2: Fix account token and SSE-C key leaking on redirects from+ --b2-download-url GHSA-gpw7-77cq-9f72 CVE-2026-116196 (Nick+ Craig-Wood)+ - build+ - Fix multiple CVEs by upgrading to go1.26.9 (Nick Craig-Wood)+ - CVE-2026-97032: net/http: HTTP/2 server crash due to+ HPACK encoder race+ - CVE-2026-78659: net/http: HTTP/2 server memory+ exhaustion due to Trailer headers+ - CVE-2026-97031: crypto/tls: reject malformed ECH outer+ extension references+ - CVE-2026-94444: cmd/go: checksum bypass for+ golang.org/fips140+ - CVE-2026-94447: cmd/go: checksum database bypass for+ golang.org/toolchain+ - CVE-2026-94448: html/template: reset context tracking on+ consecutive template expressions+ - CVE-2026-97030: html/template: recognize |yield| as+ regexp preceder keyword+ - CVE-2026-94440: net/textproto, mime/multipart: memory+ limit bypass when parsing MIME headers+ - CVE-2026-56866: net/http: HTTP/1 client connection+ desynchronization after CONNECT rejection+ - CVE-2026-94439: net/http: HTTP/1 server connection+ desynchronization after 2xx CONNECT response+ - CVE-2026-78669: net/http: excessive CPU consumption from+ repeated initial window changes+ - CVE-2026-78660: net/http: HTTP/2 transport accepts+ malformed framing-related headers+ - CVE-2026-56857: os: Root.Mkdir(All) can follow junctions+ out of the root on Windows+ - CVE-2026-78667: net/http: lack of limit on size of+ parsed Range headers+ - CVE-2026-78663: net/http: double flow control refund on+ HTTP/2 server streams+ - Update golang.org/x/net to v0.60.0 to fix multiple CVEs+ (Nick Craig-Wood)+ - CVE-2026-78659: net/http: HTTP/2 server memory+ exhaustion due to Trailer headers+ - CVE-2026-78669: net/http: excessive CPU consumption from+ repeated initial window changes+ - CVE-2026-78663: net/http: double flow control refund on+ HTTP/2 server streams+ - CVE-2026-97032: net/http: HTTP/2 server crash due to+ HPACK encoder race+ - Update google.golang.org/grpc to fix CVE-2026-84445+ (dependabot[bot])+ - CVE-2026-84445: xds: server panic on a request missing+ both :authority and Host headers+ - fshttp: Drop --header values over HTTP after HTTPS downgrade+ GHSA-rrwf-2qr6-p832 CVE-PENDING (Nick Craig-Wood)+ - http: Stop a path starting with // moving the remote to another+ host GHSA-xqmq-v4wj-rw5m CVE-2026-116192 (Nick Craig-Wood)+ - lib/http: Require authentication for OPTIONS requests+ GHSA-gj73-fh6v-92fj CVE-2026-116193 (Nick Craig-Wood)+ - local: Fix .rclonelink names escaping the root with --links+ GHSA-55qp-jrwr-x827 CVE-2026-116195 (Nick Craig-Wood)+ - netstorage+ - Stop a path starting with // moving the remote to another+ host GHSA-xqmq-v4wj-rw5m CVE-2026-116192 (Nick Craig-Wood)+ - Stop a listing resume token moving signed requests to+ another host GHSA-xqmq-v4wj-rw5m CVE-2026-116192 (Nick+ Craig-Wood)+ - pcloud+ - Fix OAuth config accepting callbacks without state+ GHSA-hvpr-p4pv-4f46 CVE-2026-116194 (Nick Craig-Wood)+ - Only accept pCloud hosts from the hostname in the OAuth+ callback GHSA-hvpr-p4pv-4f46 (Nick Craig-Wood)+ - selfupdate+ - Fix hash check trusting unsigned data in the signed+ SHA256SUMS GHSA-w33h-3qgq-8mv8 CVE-2026-116189 (Nick+ Craig-Wood)+ - Fix release verification instructions trusting unsigned+ hashes GHSA-w33h-3qgq-8mv8 (Nick Craig-Wood)+ - serve nfs: Fix path traversal by refusing unsafe file names in+ the VFS GHSA-44pm-2q5p-fm63 CVE-2026-116191 (Nick Craig-Wood)+ - serve s3+ - Fix presigned URL being turned into a copy of any object+ GHSA-w3h7-jxxq-vfrp CVE-PENDING (Nick Craig-Wood)+ - Stop auth proxy users listing or using each other's+ multipart uploads and metadata GHSA-35g6-9fjv-rmx5+ CVE-2026-116187 (Nick Craig-Wood)+ - Fix multipart buffer limits being bypassed by overflowing+ part sizes GHSA-qjw8-8hmp-c3m6 CVE-2026-116188 (Nick+ Craig-Wood)+ - Add --multipart-max-uploads to limit uploads in progress+ (Nick Craig-Wood)+ - Add --multipart-streaming-buffer-total to limit memory+ across uploads (Nick Craig-Wood)+ - Add --metadata-max-objects to limit the metadata kept in+ memory (Nick Craig-Wood)+ - serve webdav: Fix path traversal above the served directory+ GHSA-44pm-2q5p-fm63 CVE-2026-116191 (Nick Craig-Wood)+ - webdav: Fix headers option leaking to other hosts on redirect+ GHSA-3rqf-7h69-gg5v CVE-2026-116190 (Nick Craig-Wood)+- Bug Fixes+ - accounting+ - Stop averaging when the last check finishes (Kunpeng Xie)+ - Keep speed averaging active when transfers overlap (kaii9)+ - dedupe: Fix rename mode giving up after 100 names and make it+ faster (Nick Craig-Wood)+ - doc fixes (Dhevenddra K G, enkvadrat, GhostCoder6969, James+ Truitt, maximilize, n4n5, Nick Craig-Wood, PSR94, Rohit Behera,+ shaurya, subomi)+ - docker: Fix files uploaded with the wrong mime type (hsdfat)+ - fs+ - Fix about showing a negative total when a quota reaches the+ int64 maximum (youdie006)+ - Fix bandwidth limits below 1 KiB being multiplied by 1024+ (Roland)+ - Fix --bwlimit timetables written out of order using the+ wrong limit (Rohit Behera)+ - Make BwTimetable.Set replace the timetable instead of+ appending to it (youdie006)+ - Retry "network is unreachable" and "network is down" errors+ (nielash)+ - fserrors: Fix out of space detection on Windows+ (ferrumclaudepilgrim)+ - lib/rest: Limit the size of HTTP response bodies read into+ memory (Nick Craig-Wood)+ - log: Fix race when adding a log output while logging (Nick+ Craig-Wood)+ - oauthutil: Fix Renew.Shutdown nil timer panic and race+ (TastyHeadphones)+ - operations+ - Stop --copy-dest replacing files with --immutable (phatlc)+ - Fix ignored error in rcat probe reads (Wang Chencheng)+ - Fix hang when deleting files and a fatal error occurs (Nick+ Craig-Wood)+ - Fix hang and lost context when moving a directory file by+ file (Nick Craig-Wood)+ - Don't sleep for a Retry-After error when the transfer is+ cancelled (Nick Craig-Wood)+ - rc+ - Make job/status and job/list require authentication (Nick+ Craig-Wood)+ - Fix rc API accepting an out of range number and overflowing+ 64 bits (KBS)+ - Reject out of range integer parameters instead of truncating+ them (Nick Craig-Wood)+ - Fix large numeric parameters being rejected on 32 bit builds+ (Jeremy Schoemaker)+ - selfupdate: Fix --version X.Y depending on ./ in the download+ site links (Nick Craig-Wood)+ - serve+ - Fix --auth-proxy stopping VFS cache uploads after 5 minutes+ (Nick Craig-Wood)+ - Fix --auth-proxy backends never being shut down after+ simultaneous logins (Nick Craig-Wood)+ - serve dlna: Log unescaped paths (phatlc)+ - serve docker: Fix volume path being lost when the plugin+ restarts (phatlc)+ - serve ftp+ - Fix transfers failing after 5 minutes with --auth-proxy+ (Nick Craig-Wood)+ - Fix VFS leak when the server fails to start (Nick+ Craig-Wood)+ - serve http+ - Fix downloads failing after 5 minutes with --auth-proxy+ (Nick Craig-Wood)+ - Escape filename in zip download Content-Disposition header+ (Harsh Raj Singhania)+ - serve restic+ - Fix repositories named data (kaii9)+ - Prevent concurrent append-only overwrites (Vasek Sraier)+ - serve s3+ - Apply the multipart upload limits and --metadata-max-objects+ to each user separately (Nick Craig-Wood)+ - Stream in-order multipart parts without buffering them (Nick+ Craig-Wood)+ - Start the upload to the backend on the first multipart part+ (Nick Craig-Wood)+ - Don't buffer re-uploads of multipart parts already streamed+ (Nick Craig-Wood)+ - Count buffered multipart parts in whole memory pages (Nick+ Craig-Wood)+ - Time out multipart parts waiting for buffer space (Nick+ Craig-Wood)+ - Limit the memory requests can use (Nick Craig-Wood)+ - Forget expired multipart uploads completely (Nick+ Craig-Wood)+ - Forget the metadata of deleted objects (Nick Craig-Wood)+ - List objects lazily so paging a deep hierarchy is fast (Nick+ Craig-Wood)+ - Fix multipart upload corruption if the backend write of an+ early part fails (Nick Craig-Wood)+ - Fix uploads failing when the auth proxy expires the user's+ backend (Nick Craig-Wood)+ - Fix stale metadata being returned for objects changed+ elsewhere (Nick Craig-Wood)+ - Fix --etag-hash auto crashing or using the wrong hash with+ --auth-proxy (Nick Craig-Wood)+ - Fix CopyObject of a missing object not returning NoSuchKey+ (Nick Craig-Wood)+ - Fix log messages with bad format strings (Nick Craig-Wood)+ - Document memory use and bring the docs up to date (Nick+ Craig-Wood)+ - serve webdav+ - Fix transfers failing after 5 minutes with --auth-proxy+ (Nick Craig-Wood)+ - Fix crash when the server fails to start (Nick Craig-Wood)+ - Escape filename in zip download Content-Disposition header+ (Harsh Raj Singhania)+ - vfscache: Fix hang when the cache cleaner is disabled+ (ferrumclaudepilgrim)+- Mount+ - Fix --daemon failing when --rc is enabled (somaz)+ - Ignore com.apple.* xattrs on the macOS FSKit backend (Eugene)+- VFS+ - Fix reads returning zeros after a file changes during handle+ caching (Neil Cawse)+ - Stop setting a file's modtime from discarding its cached data+ (Neil Cawse)+ - Fix crash reading the metadata of a file which is being written+ (Nick Craig-Wood)+ - Fix a VFS being reused while it is being shut down (Nick+ Craig-Wood)+ - Fix AddVirtual ignoring isDir (phatlc)+ - Stop --links objects called "...rclonelink" appearing as ".." in+ directory listings (Nick Craig-Wood)+- Local+ - Fix directories named *.rclonelink being treated as links with+ --links (Nick Craig-Wood)+- Archive+ - Fix corrupt listings when listing a zip directory more than once+ (Nick Craig-Wood)+ - Fix zip file entries named for a directory causing confusion+ (Nick Craig-Wood)+ - Fix listing entries with a leading slash as if they were in the+ root (Nick Craig-Wood)+ - Fix crash when creating archive to stdout (Vladimir Babin)+ - Fix a squashfs path to a single file listing its whole directory+ (Nick Craig-Wood)+- Azure Blob+ - Use the released Azure SDK for Apache Arrow listing (Nick+ Craig-Wood)+- Box+ - Fix log messages with bad format strings (Nick Craig-Wood)+- Chunker+ - Fix panic listing a directory after an interrupted upload with+ meta_format none (Nick Craig-Wood)+- Compress+ - Fix crash on ranged reads when gzip metadata is corrupted+ (Acts1631)+- Drime+ - Fix server-side copy over an existing file leaving a "name (1)"+ copy (Nick Craig-Wood)+ - Fix deleted files and directories still being listed with+ hard_delete (Nick Craig-Wood)+- Drive+ - Document the Branding step needed to publish your own client_id+ (Sanjay Kanth A)+- Dropbox+ - Fix shared folder mount for roots nested more than one level+ deep (phatlc)+ - Match shared-folder and received-file names case-insensitively+ (phatlc)+- Filescom+ - Fix corrupted uploads after a retried upload error (Nick+ Craig-Wood)+ - Document that names with / or in can't be stored (Nick+ Craig-Wood)+- HTTP+ - Fix crash when the server fails to start (Nick Craig-Wood)+- Iclouddrive+ - Fix potential crash looking up items (Nick Craig-Wood)+- Imagekit+ - Fix potential crash in rmdir and purge (Nick Craig-Wood)+- Internxt+ - Fix lookups of files starting with a dot and dropped uploads+ (jzunigax2)+ - Fix sync with --backup-dir or --suffix deleting the backed up+ file (Nick Craig-Wood)+ - Fix server-side moves failing with "Not Found" or "already+ exists" (Nick Craig-Wood)+ - Fix "directory not empty" and stale directories after moves and+ deletes (Nick Craig-Wood)+ - Fix server-side directory move failing after a gateway timeout+ (Nick Craig-Wood)+- Onedrive+ - Update docs for versions, links and time precision on personal+ accounts (Nick Craig-Wood)+ - Document that shared with me shortcuts may fail to list on+ personal (Nick Craig-Wood)+ - Update --onedrive-hard-delete docs to mention personal account+ support (foecmke)+- Oracleobjectstorage+ - Fix SSE-C server-side copies (tomaszni)+ - Upload empty streams without multipart (tomaszni)+- Pixeldrain+ - Fix corrupted uploads after a retried upload error (Nick+ Craig-Wood)+- Premiumizeme+ - Fix uploading files with ";" in their names (Nick Craig-Wood)+- S3+ - Disable signing Accept-Encoding for Ceph and Linode (Aditya)+ - Fix version-at listings with URL encoded keys (jxj)+ - Fix storage class missing from --s3-versions listings (Dirk+ Petersen)+- Seafile+ - Fix corrupted uploads after a retried upload error (Nick+ Craig-Wood)+- SFTP+ - Fix failed uploads leaving disk space in use on the server (Nick+ Craig-Wood)+- Shade+ - Fix potential crash in directory move (Nick Craig-Wood)+- Smb+ - Save the user name in the config even if it matches the current+ user (phatlc)+- WebDAV+ - Fix duplicated listing entries after retried PROPFIND (ZRHann)+ - Fix "XML syntax error" on paths which don't exist with ownCloud+ 10.16 (Nick Craig-Wood)++v1.75.1 - 2026-09-04++See commits++- Security+ - archive+ - Fix zip slip path traversal in untrusted zip files+ GHSA-66hp-wgxq-6f5q CVE-2026-88014 (Nick Craig-Wood)+ - Hide any archive entry which escapes the directory being+ listed GHSA-66hp-wgxq-6f5q (Nick Craig-Wood)+ - Reject unsafe entry names when mounting squashfs images+ GHSA-66hp-wgxq-6f5q (Nick Craig-Wood)+ - Fix zip subdirectory root matching sibling directories+ GHSA-66hp-wgxq-6f5q (Nick Craig-Wood)+ - Fix zip entry named "." hiding every other file+ GHSA-66hp-wgxq-6f5q (Nick Craig-Wood)+ - Fix "directory not found" for archive paths containing "./"+ or "//" GHSA-66hp-wgxq-6f5q (Nick Craig-Wood)+ - build+ - Fix multiple CVEs by upgrading to go1.26.6 (Nick Craig-Wood)+ - CVE-2026-56860: net/url: quadratic complexity in+ resolvePath+ - CVE-2026-56858: html/template: JavaScript regexp context+ tracking+ - CVE-2026-56862: crypto/tls: limit handshake messages+ accepted post-handshake+ - CVE-2026-56853: net/http: apply ReadHeaderTimeout to+ unencrypted HTTP/2 check+ - CVE-2026-56859: encoding/xml: recursion depth guard+ during decode+ - CVE-2026-33818: encoding/asn1: enforce maximum recursion+ depth+ - CVE-2026-46600: net: panic parsing an invalid SVCB or+ HTTPS RR in dnsmessage+ - CVE-2026-39821: net/http: reject ASCII-only+ Punycode-encoded labels in idna+ - Update golang.org/x/crypto to v0.56.0 to fix multiple CVEs+ (Nick Craig-Wood)+ - CVE-2026-56854: ssh: source-address critical option not+ enforced for non-public-key auth callbacks+ - CVE-2026-78662: ssh: a malicious peer could flood an+ undecided channel's incoming requests, deadlocking the+ connection+ - CVE-2026-56855: ssh: a malicious peer could send crafted+ messages on an established channel, deadlocking the+ connection+ - Update golang.org/x/image to v0.45.0 to fix CVE-2026-46603+ (Nick Craig-Wood)+ - CVE-2026-46603: excessive memory allocation during VP8L+ decoding+ - fs: Confine directory listing entries that escape the root+ GHSA-3vxh-3pcx-9m8q CVE-2026-93986 GHSA-38xv-hf3p-h7mq+ CVE-2026-88046 (Nick Craig-Wood)+ - fshttp: Don't send --header values to other hosts on redirect+ GHSA-486v-q2wf-fp2r CVE-2026-88013 (Nick Craig-Wood)+ - http: Don't leak configured headers to other hosts or over+ plaintext on redirect GHSA-486v-q2wf-fp2r CVE-2026-88013 (Nick+ Craig-Wood)+ - lib/rest: Check HTTPS downgrades against the original request on+ redirect GHSA-486v-q2wf-fp2r CVE-2026-88013 (Nick Craig-Wood)+ - local+ - Fix dir metadata escaping the root through a planted symlink+ GHSA-f8g7-2xjc-7mfh CVE-2026-88016 (Nick Craig-Wood)+ - Fix btime escaping the root via a planted symlink+ GHSA-f8g7-2xjc-7mfh CVE-2026-88016 (Nick Craig-Wood)+ - Fix panic on Range request past the end of a symlink+ GHSA-p6m2-r3w9-mpxw CVE-2026-88015 (Nick Craig-Wood)+ - serve docker+ - Reject volume names that escape the base directory+ GHSA-p6vx-hf7p-98j6 CVE-2026-93987 (Nick Craig-Wood)+ - Reject volume names resolving to the base directory itself+ GHSA-p6vx-hf7p-98j6 CVE-2026-93987 (Nick Craig-Wood)+ - Re-derive volume mountpoint from name when restoring state+ GHSA-p6vx-hf7p-98j6 CVE-2026-93987 (Nick Craig-Wood)+ - serve ftp: Fix auth-proxy sessions sharing credentials by+ username GHSA-c476-6w5q-jw77 CVE-2026-88017 (Nick Craig-Wood)+ - serve s3+ - Fix memory exhaustion from client-declared multipart part+ size GHSA-2p48-j3qc-rx9f CVE-2026-88045 (Nick Craig-Wood)+ - Reject bogus multipart part sizes in the reorder buffer+ GHSA-2p48-j3qc-rx9f (Nick Craig-Wood)+ - Fix auth proxy accepting any request signed with an empty+ secret GHSA-xwwr-4h3p-r22c CVE-2026-88018 (Nick Craig-Wood)+ - NB the auth proxy protocol for serve s3 has changed -+ the proxy program is now given the access key ID as user+ and must return the secret as _secret_access_key+ - Fix each server accepting the --auth-key credentials of all+ the others (Nick Craig-Wood)+ - Fix misleading anonymous access log when using an auth proxy+ via rc GHSA-p569-5gjg-9cmj CVE-2026-88044 (Nick Craig-Wood)+ - serve sftp: Fix auth proxy configured via rc being silently+ ignored GHSA-p569-5gjg-9cmj CVE-2026-88044 (Nick Craig-Wood)+- Bug Fixes+ - accounting+ - Fix memory leak on long-running rcd (nielash)+ - Fix memory leak from stats groups on long-running rcd+ (nielash)+ - Fix bwlimit burst overflow (Rayan Salhab)+ - bisync+ - Fix memory leak when running via the rc (nielash)+ - Fix failed transfers of empty files being recorded as synced+ (Nick Craig-Wood)+ - build: Make go1.26 the minimum required version as needed by+ golang.org/x/crypto v0.56.0 (Nick Craig-Wood)+ - config: Redact env var config values in logs (Pastalikek65)+ - doc fixes (Anton Karpov, CAOShurong, Dean Chen, Nick Craig-Wood,+ Recoordinate, Rodrigo Rodrigues, Shantanav Mukherjee, shaurya)+ - lib/batcher: Prevent commits racing shutdown (Loi Nguyen)+ - lib/transform: Fix panic in truncate_keep_extension (VXNCXNX)+ - multipart: Fix chunked uploads storing truncated objects when+ the source ends early (Nick Craig-Wood)+ - operations: Fix silent truncation of streaming uploads whose+ source ends early (Nick Craig-Wood)+ - serve+ - Fix VFS instance leaks on server startup failures and+ shutdown (Hakan İSMAİL)+ - Pass the client IP address to the auth proxy+ (am-at-enrollvb)+ - serve http: Prevent scrolling to the top on page reload (Sune+ Mølgaard)+ - serve nfs: Fix EIO when creating symlinks with --vfs-links+ (SillyZir)+ - serve s3+ - Fix failed uploads deleting or corrupting the object at the+ key (Nick Craig-Wood)+ - Fix crash when a multipart upload is aborted while a part is+ uploading (Nick Craig-Wood)+ - Fix modtime not being set when only mtime metadata is+ supplied on PUT (Nick Craig-Wood)+ - Upload all multipart uploads via the VFS so they obey+ --bwlimit and show in stats (Nick Craig-Wood)+ - Reserve the .rclone_temp_ prefix for temporary objects (Nick+ Craig-Wood)+ - Clean up abandoned multipart uploads after+ --multipart-expiry (Nick Craig-Wood)+ - vfscache+ - Fix reader deadlock when the item size drops below the read+ offset (Dave)+ - Fix log message growing without bound on repeated write+ errors (Vijay Misal)+ - walk: Stop directory traversal when the context is cancelled+ (Rahman Yilmaz)+- VFS+ - Synchronize poll updates with shutdown (Loi Nguyen)+ - Make poll shutdown lifecycle deterministic (Loi Nguyen)+- Crypt+ - Fix hash mismatches with no_data_encryption on backends which+ check upload hashes (Nick Craig-Wood)+ - Fix directory names which look like versioned file names+ (TowyTowy)+ - Warn about directories with legacy version-like encrypted names+ (Nick Craig-Wood)+- Azure Blob+ - Fix Entra ID server-side copy source authentication (Edward+ Klesel)+ - Fix spurious vfs cache corruption errors during chunked reads+ (Nick Craig-Wood)+- Azurefiles+ - Fix zero padded files being created when the source ends early+ (Nick Craig-Wood)+- Box+ - Fix truncated files being uploaded successfully when the source+ ends early (Rohit Behera)+- Compress+ - Fix corrupted objects being created when the source ends early+ (Nick Craig-Wood)+- Drive+ - Don't list trashed files when removing a directory into the+ trash (alliasgher)+- Dropbox+ - Preserve Paper export paths on lookup (Loi Nguyen)+ - Fix context cancellation (e.g. --max-duration limit) not+ stopping in-flight requests (debaditya)+ - Fix chunked uploads of truncated files never finishing (Nick+ Craig-Wood)+ - Don't retry chunked upload requests when the upload has been+ cancelled (Nick Craig-Wood)+ - Decode received shared-file names (Sanjay Kanth A)+ - Fix ChangeNotify when the root's case differs from Dropbox's+ (Loi Nguyen)+- Filelu+ - Fix truncated files being uploaded successfully when the source+ ends early (Nick Craig-Wood)+ - Fix duplicate root path during multipart folder creation+ (kingston125)+- Huaweidrive+ - Fix truncated files being uploaded successfully when the source+ ends early (Rohit Behera)+- Iclouddrive+ - Fix uploads into an app container failing with 412 (Christian De+ Santis)+- Internetarchive+ - Fix corrupted files being created when the source ends early+ (Nick Craig-Wood)+- Internxt+ - Persist rotated token returned by the user info call+ (0rangeSeaW0lf)+- Onedrive+ - Fix 403 Forbidden for configuration personal onedrive (machsix)+ - Fall back to manual drive ID entry when drive listing fails+ (SillyZir)+ - Don't retry multipart upload chunk on 404 (upload session not+ found) (water)+- Overview+ - Fix "internal error: no overview data found" on 32 bit+ architectures (Nick Craig-Wood)+- Pikpak+ - Fix truncated files being created when the source ends early+ (Nick Craig-Wood)+ - Fix truncated single part uploads reported as ok when source+ ends early (Nick Craig-Wood)+- Protondrive+ - Fix files uploaded with v1.75.0 not being readable in the Proton+ apps (Nick Craig-Wood)+ - Fix corrupted uploads after a retried upload error (Nick+ Craig-Wood)+- Quatrix+ - Fix chunk upload retries and fix memory leak (Nick Craig-Wood)+- S3+ - Update Mega endpoints (Nick Craig-Wood)+ - Treat UploadPart success without ETag as retryable error+ (CAOShurong)+ - Fix server side copy failing with --s3-no-head-object (Anatoly+ Tarnavsky)+- Sia+ - Fix corrupted files being created when the source ends early+ (Nick Craig-Wood)+- Smb+ - Reuse the upload connection for SetModTime (alliasgher)+- WebDAV+ - Fix SetModTime failing and hashes missing on Nextcloud (Nick+ Craig-Wood)+- Yandex+ - Fix truncated files being uploaded successfully when the source+ ends early (Rohit Behera)+v1.75.0 - 2026-07-31See commits@@ -68573,42 +72832,42 @@ See commits- Zero Services (ZERO-Z3)- Security- archive: Don't crash on malformed squashfs images- GHSA-6jcg-q3wp-x2f4 CVE-PENDING (Nick Craig-Wood)+ GHSA-6jcg-q3wp-x2f4 CVE-2026-79775 (Nick Craig-Wood)- ftp: Fix ftp command injection when encoding doesn't include- CRLF GHSA-8c48-q9wj-3w37 CVE-PENDING (Nick Craig-Wood)+ CRLF GHSA-8c48-q9wj-3w37 CVE-2026-71311 (Nick Craig-Wood)- lib/http: Use TLS on all --addr listeners when --cert and --keyare set GHSA-mfvx-7rcj-9m5g (Nick Craig-Wood)- lib/proxy: Fix unbounded HTTP CONNECT headers causing OOM- GHSA-xhf4-832v-7xcr CVE-PENDING (Nick Craig-Wood)+ GHSA-xhf4-832v-7xcr CVE-2026-71310 (Nick Craig-Wood)- local: Stop source file names escaping the destination directory- GHSA-7p4m-qxvv-g567 CVE-PENDING (Nick Craig-Wood)+ GHSA-7p4m-qxvv-g567 CVE-2026-71313 (Nick Craig-Wood)- rc- Don't expose pprof debug handlers on an unauthenticated- server GHSA-mfvx-7rcj-9m5g CVE-PENDING (Nick Craig-Wood)+ server GHSA-mfvx-7rcj-9m5g CVE-2026-79776 (Nick Craig-Wood)- Require authentication to list the remotes with --rc-serveGHSA-mfvx-7rcj-9m5g (Nick Craig-Wood)- - Fix leaking stack traces on panics GHSA-gwfq-86j8-7qhv (Nick- Craig-Wood)+ - Fix leaking stack traces on panics GHSA-gwfq-86j8-7qhv+ CVE-2026-79777 (Nick Craig-Wood)- s3- Fix redirect credential leaks, reject HTTPS->HTTP and strip- secrets GHSA-8mxv-9xhp-86h4 (Nick Craig-Wood)+ secrets GHSA-8mxv-9xhp-86h4 CVE-2026-79780 (Nick Craig-Wood)- Strip S3 Express session token on cross-host redirects- GHSA-8mxv-9xhp-86h4 (Nick Craig-Wood)+ GHSA-8mxv-9xhp-86h4 CVE-2026-79780 (Nick Craig-Wood)- serve ftp: Use constant time comparison for password checkGHSA-mfvx-7rcj-9m5g (Nick Craig-Wood)- serve restic: Fix path traversal above the served directory- GHSA-45pq-889g-fcgh CVE-PENDING (Nick Craig-Wood)+ GHSA-45pq-889g-fcgh CVE-2026-71309 (Nick Craig-Wood)- serve sftp: Don't crash the whole server on a bad request- GHSA-6jcg-q3wp-x2f4 (Nick Craig-Wood)+ GHSA-6jcg-q3wp-x2f4 CVE-2026-79775 (Nick Craig-Wood)- sftp: Fix command injection via crafted filenames on PowerShell- remotes GHSA-2m8m-jhrm-w6j2 CVE-PENDING (Nick Craig-Wood)+ remotes GHSA-2m8m-jhrm-w6j2 CVE-2026-71312 (Nick Craig-Wood)- vfs: Don't crash the process if a backend panics on a background- goroutine GHSA-6jcg-q3wp-x2f4 (Nick Craig-Wood)+ goroutine GHSA-6jcg-q3wp-x2f4 CVE-2026-79775 (Nick Craig-Wood)- webdav- Fix HTTPS to HTTP redirects leaking credentials- GHSA-h4mf-4v27-hggj (Nick Craig-Wood)+ GHSA-h4mf-4v27-hggj CVE-2026-79779 (Nick Craig-Wood)- Tus: fix potential nil pointer crash GHSA-3x6r-wxxg-53vv- (Nick Craig-Wood)+ CVE-2026-79778 (Nick Craig-Wood)- Update google.golang.org/grpc to fix multiple security problems(Nick Craig-Wood)- New Features@@ -68954,7 +73213,7 @@ See commits- serve s3- Fix spurious 404 on HEAD/GET during VFS writeback (max)- Fix path traversal letting clients see files in the root- GHSA-8v25-v8p6-qf7v (Nick Craig-Wood)+ GHSA-8v25-v8p6-qf7v CVE-2026-79781 (Nick Craig-Wood)- serve webdav: Fix MOVE overwrite failing without Overwriteheader (Sanjay Santhanam)- serve/http: Fix --disable-zip so it works over rc (Nick@@ -68966,7 +73225,7 @@ See commits- Stop --links symlinks escaping the destination directoryCVE-2026-54572 (Nick Craig-Wood)- Don't restore setuid/setgid/sticky bits from metadata by default- GHSA-945v-v9p3-v5xw (Nick Craig-Wood)+ GHSA-945v-v9p3-v5xw CVE-2026-79783 (Nick Craig-Wood)- Drive- Warn when non-exportable Google documents are skipped (NickCraig-Wood)@@ -79253,6 +83512,143 @@ Contributors- Socialpranker [email protected]- phatlc [email protected]- Punya Jain [email protected]+- acoeur [email protected]+- SillyZir [email protected]+- am-at-enrollvb [email protected]+- Loi Nguyen [email protected]+- debaditya [email protected]+- alliasgher [email protected]+- Anton Karpov [email protected]+- Edward Klesel [email protected]+- Christian De Santis+- Rodrigo Rodrigues [email protected]+- Recoordinate [email protected]+- Jiacheng Xu [email protected]+- Teddy Tennant [email protected]+- Dean Chen [email protected]+- Pastalikek65 [email protected]+- Morax [email protected]+- Shurong Cao [email protected]+- Dave [email protected]+- max [email protected]+- Hakan İSMAİL [email protected]+- Shantanav Mukherjee [email protected]+- VXNCXNX [email protected]+- Rohit Behera [email protected]+- Rahman Yilmaz [email protected]+- Sanjay Kanth A [email protected]+- machsix [email protected]+- CAOShurong [email protected]+- shaurya [email protected]+- water [email protected]+- Rayan Salhab [email protected]+- Vijay Misal [email protected]+- Sune Mølgaard [email protected]+- Anatoly Tarnavsky [email protected]+- cyphercodes [email protected]+- shaurya [email protected]+- no-hup [email protected]+- Dhevenddra [email protected]+- Splainte [email protected]+- PSR94 [email protected]+- Shane McCarron [email protected]+- Dhevenddra K G [email protected]+- Vaibhav Mashal [email protected]+- Murat Topcu [email protected]+- tomaszni [email protected]+- eliotee [email protected]+- Aditya [email protected]+- Nicholas Velten [email protected]+- youdie006 [email protected]+- subomi [email protected]+- Can Arslan [email protected]+- Can Arslan [email protected]+- ZRHann [email protected]+- enkvadrat [email protected]+- Vladimir Babin [email protected]+- Eugene [email protected]+- Kunpeng Xie [email protected]+- foecmke [email protected]+- Vasek Sraier [email protected]+- Kalin Stoyanov [email protected]+- NytePlus [email protected]+- Jeremy Schoemaker [email protected]+- solunolab [email protected]+- Mattias Michaux [email protected]+- jxj [email protected]+- Harsh Raj Singhania+- Roland [email protected]+- interested.tortoise [email protected]+- hsdfat [email protected]+- Dirk Petersen [email protected]+- TastyHeadphones [email protected]+- Neil Cawse [email protected]+- Lev Devaev [email protected]+- GhostCoder6969 [email protected]+- bounty-agent [email protected]+- mygrexit [email protected]+- Kaya [email protected]+- somaz [email protected]+- xiaozou-wine [email protected]+- Roman Butor [email protected]+- kaii9 [email protected]+- Stochastic [email protected]+- James Truitt [email protected]+- firatkaya [email protected]+- kevin9327 [email protected]+- drakeo338 [email protected]+- Alessio Margelli [email protected]+- Ludovic Bondon [email protected]+- fmterrors [email protected]+- GingerAdonis [email protected]+- Satvik Mishra [email protected]+- wjc [email protected]+- Maksym Tymoshyk [email protected]+- Aniket [email protected]+- Oshino Penguin [email protected]+- Yousef Issa [email protected]+- ghwany [email protected]+- DEVN [email protected]+- 47th [email protected]+- Sash [email protected]+- Shreyansh Jain [email protected]+- Biagio Peccerillo [email protected]+- gomitrah [email protected]+- Jim Counter [email protected]+- Dirk Einecke [email protected]+- ASRodrigo1 [email protected]+- chaogeshuodiannao [email protected]+- David Legrand [email protected]+- Azhar [email protected]+- Daniel Bae [email protected]+- frodleTian [email protected]+- Cestercian [email protected]+- Ayush Sharma [email protected]+- Dirk Petersen [email protected]+- Codex [email protected]+- bunnysayzz [email protected]+- hahahahahayesyeseys [email protected]+- qodelicious [email protected]+- fixfon [email protected]+- berce [email protected]+- Fred [email protected]+- Florian Josef Altendorfer [email protected]+- ashwinibailangadi [email protected]+- ilya lavrenov [email protected]+- Soujas K [email protected]+- Robert Kubiś [email protected]+- b84330808 [email protected]+- Guillaume MALLERET [email protected]+- Louis [email protected]+- 唐僧洗发用飘柔 [email protected]+- Seth Westphal [email protected]+- ZainnQureshii [email protected]+- Rishi Raj Jain [email protected]+- Enbiya Göral [email protected]+- Florian Josef AltendorferContact the rclone projectdiff --git a/cmd/bisync/rc.md b/cmd/bisync/rc.mdindex dc3efd482..52bc24b27 100644--- a/cmd/bisync/rc.md+++ b/cmd/bisync/rc.md@@ -5,6 +5,8 @@ This takes the following parameters:- path1 (required) - (string) a remote directory string e.g. `drive:path1`- path2 (required) - (string) a remote directory string e.g. `drive:path2`- dryRun - (bool) dry-run mode+- maxDelete - (int) Safety check on maximum percentage of deleted files+allowed. If exceeded, the bisync run will abort. (default: 50%)- backupDir1 - (string) --backup-dir for Path1. Must be a non-overlapping path onthe same remote.- backupDir2 - (string) --backup-dir for Path2. Must be a non-overlapping path on@@ -34,6 +36,8 @@ unavailable. (warning: may be slow and use lots of data!)using with --verbose- ignoreListingChecksum - (bool) Do not use checksums for listings (add --ignore-checksum to additionally skip post-copy checksum checks)+- maxDeleteRenamesAware - (bool) Exclude tracked renames from the --max-delete+safety check (requires --track-renames)- maxLock - (Duration) Consider lock files older than this to be expired(default: 0 (never expire)) (minimum: 2m)- noCleanup - (bool) Retain working files (useful for troubleshooting anddiff --git a/docs/content/azureblob.md b/docs/content/azureblob.mdindex f341917d2..915b00775 100644--- a/docs/content/azureblob.md+++ b/docs/content/azureblob.md@@ -919,6 +919,52 @@ Properties:- Type: int- Default: 5000+#### --azureblob-use-arrow-list++Use the Apache Arrow listing format.++If set, directory listings are fetched using the ListBlobs Apache+Arrow response format instead of XML. Arrow responses are smaller and+much cheaper to parse, making listings of large containers several+times faster. Combine with "list_parallelism" for the biggest gains.++"Blob Listing with Apache Arrow" is in public preview at Microsoft and+is only supported on flat namespace accounts. On accounts with a+hierarchical namespace (ADLS Gen2), or where the feature is otherwise+unavailable, the server returns XML and the listing transparently+falls back to the normal XML path (logged at debug level).++Properties:++- Config: use_arrow_list+- Env Var: RCLONE_AZUREBLOB_USE_ARROW_LIST+- Type: bool+- Default: false++#### --azureblob-list-parallelism++Number of parallel shards to list a directory with.++If set greater than 1, the blob name keyspace of each directory is+split into ranges which are listed concurrently using the Arrow+startFrom/endBefore range parameters. This can dramatically speed up+listing containers with millions of objects, for both recursive+(ListR) and single directory listings. Speed keeps improving up to a+parallelism of around 30.++This has no effect unless "use_arrow_list" is also set, as Arrow is the+only listing path that supports server-side name ranges. If the+account does not support range listing (e.g. it has a hierarchical+namespace) the listing falls back to sequential. The default of 0 (or+1) lists sequentially.++Properties:++- Config: list_parallelism+- Env Var: RCLONE_AZUREBLOB_LIST_PARALLELISM+- Type: int+- Default: 0+#### --azureblob-access-tierAccess tier of blob: hot, cool, cold or archive.@@ -1022,6 +1068,9 @@ creates an empty object ending with "/", to persist the folder.This object also has the metadata "hdi_isfolder = true" to conform tothe Microsoft standard.++This is not needed on storage accounts with a hierarchical namespace+(see the hns option) as these always support empty directories.Properties:@@ -1031,6 +1080,24 @@ Properties:- Type: bool- Default: false+#### --azureblob-hns++Set if the storage account has a hierarchical namespace.++Storage accounts with hierarchical namespace enabled (Azure Data Lake+Storage Gen2) have real directories which can be empty.++If this is unset then rclone will read it from the storage account+and save it in the config file.+++Properties:++- Config: hns+- Env Var: RCLONE_AZUREBLOB_HNS+- Type: Tristate+- Default: unset+#### --azureblob-no-check-containerIf set, don't attempt to check the container exists or create it.diff --git a/docs/content/azurefiles.md b/docs/content/azurefiles.mdindex 370a21310..f6358a503 100644--- a/docs/content/azurefiles.md+++ b/docs/content/azurefiles.md@@ -736,7 +736,7 @@ Properties:Upload chunk size.Note that this is stored in memory and there may be up to-"--transfers" * "--azurefile-upload-concurrency" chunks stored at once+"--transfers" * "--azurefiles-upload-concurrency" chunks stored at oncein memory.Properties:@@ -758,7 +758,7 @@ links and these uploads do not fully utilize your bandwidth, thenincreasing this may help to speed up the transfers.Note that chunks are stored in memory and there may be up to-"--transfers" * "--azurefile-upload-concurrency" chunks stored at once+"--transfers" * "--azurefiles-upload-concurrency" chunks stored at oncein memory.Properties:diff --git a/docs/content/b2.md b/docs/content/b2.mdindex d6976e609..cf1d4fb66 100644--- a/docs/content/b2.md+++ b/docs/content/b2.md@@ -569,6 +569,11 @@ If the custom endpoint rewrites the requests for authentication,e.g., in Cloudflare Workers, this header needs to be handled properly.Leave blank if you want to use the endpoint provided by Backblaze.+Rclone will not follow a redirect from this endpoint which downgrades+HTTPS to HTTP, and will not send the "Authorization" header or the+SSE-C key headers once a redirect has left the host and port of this+endpoint.+The URL provided here SHOULD have the protocol and SHOULD NOT havea trailing slash or specify the /file/bucket subpath as rclone willrequest files with "{download_url}/file/{bucket_name}/{path}".@@ -665,7 +670,7 @@ Properties:To use SSE-C, you may provide the secret encryption key encoded in a UTF-8 compatible string to encrypt/decrypt your data-Alternatively you can provide --sse-customer-key-base64.+Alternatively you can provide --b2-sse-customer-key-base64.Properties:@@ -681,7 +686,7 @@ Properties:To use SSE-C, you may provide the secret encryption key encoded in Base64 format to encrypt/decrypt your data-Alternatively you can provide --sse-customer-key.+Alternatively you can provide --b2-sse-customer-key.Properties:diff --git a/docs/content/bisync.md b/docs/content/bisync.mdindex ea1ec7829..bbda67319 100644--- a/docs/content/bisync.md+++ b/docs/content/bisync.md@@ -1074,14 +1074,20 @@ The following backends have known issues that need more investigation:- `TestHuaweiDrive` (`huaweidrive`)- [`TestBisyncRemoteLocal/ext_paths`](https://pub.rclone.org/integration-tests/current/huaweidrive-cmd.bisync-TestHuaweiDrive-1.txt)- [`TestBisyncRemoteLocal/extended_filenames`](https://pub.rclone.org/integration-tests/current/huaweidrive-cmd.bisync-TestHuaweiDrive-1.txt)- - [`TestBisyncRemoteLocal/normalization`](https://pub.rclone.org/integration-tests/current/huaweidrive-cmd.bisync-TestHuaweiDrive-1.txt)- [`TestBisyncLocalRemote/ext_paths`](https://pub.rclone.org/integration-tests/current/huaweidrive-cmd.bisync-TestHuaweiDrive-1.txt)- [`TestBisyncLocalRemote/extended_filenames`](https://pub.rclone.org/integration-tests/current/huaweidrive-cmd.bisync-TestHuaweiDrive-1.txt)- - [4 more](https://pub.rclone.org/integration-tests/current/)+ - [`TestBisyncLocalRemote/normalization`](https://pub.rclone.org/integration-tests/current/huaweidrive-cmd.bisync-TestHuaweiDrive-1.txt)+ - [3 more](https://pub.rclone.org/integration-tests/current/)- `TestPcloud` (`pcloud`)- - [`TestBisyncRemoteRemote/check_access`](https://pub.rclone.org/integration-tests/current/pcloud-cmd.bisync-TestPcloud-1.txt)- - [`TestBisyncRemoteRemote/rmdirs`](https://pub.rclone.org/integration-tests/current/pcloud-cmd.bisync-TestPcloud-1.txt)-- Updated: 2026-07-31-010017+ - [`TestBisyncRemoteRemote/extended_filenames`](https://pub.rclone.org/integration-tests/current/pcloud-cmd.bisync-TestPcloud-1.txt)+- `TestYandex` (`yandex`)+ - [`TestBisyncRemoteLocal/all_changed`](https://pub.rclone.org/integration-tests/current/yandex-cmd.bisync-TestYandex-1.txt)+ - [`TestBisyncRemoteLocal/backupdir`](https://pub.rclone.org/integration-tests/current/yandex-cmd.bisync-TestYandex-1.txt)+ - [`TestBisyncRemoteLocal/basic`](https://pub.rclone.org/integration-tests/current/yandex-cmd.bisync-TestYandex-1.txt)+ - [`TestBisyncRemoteLocal/changes`](https://pub.rclone.org/integration-tests/current/yandex-cmd.bisync-TestYandex-1.txt)+ - [`TestBisyncRemoteLocal/check_access`](https://pub.rclone.org/integration-tests/current/yandex-cmd.bisync-TestYandex-1.txt)+ - [94 more](https://pub.rclone.org/integration-tests/current/)+- Updated: 2026-11-06-010024<!--- end list_failures - DO NOT EDIT THIS SECTION - use make commanddocs --->The following backends either have not been tested recently or have known issues@@ -1099,6 +1105,8 @@ that are deemed unfixable for the time being:- `TestFilesCom` (`filescom`)- `TestImageKit` (`imagekit`)- `TestJottacloud` (`jottacloud`)+- `TestKdrive` (`kdrive`)+ - `TestBisyncRemoteRemote/normalization`- `TestLinkbox` (`linkbox`)- `TestMailru` (`mailru`)- `TestMega` (`mega`)@@ -1111,7 +1119,8 @@ that are deemed unfixable for the time being:- `TestQuatrix` (`quatrix`)- `TestS3GCS` (`s3`)- `TestBisyncRemoteRemote/extended_filenames`-- `TestS3Rclone` (`s3`)+ - `TestBisyncRemoteLocal/extended_filenames`+ - `TestBisyncLocalRemote/extended_filenames`- `TestSFTPRsyncNet` (`sftp`)- `TestStorj` (`storj`)- `TestWebdavNextcloud` (`webdav`)diff --git a/docs/content/changelog.md b/docs/content/changelog.mdindex 32bc34014..04cd6268e 100644--- a/docs/content/changelog.md+++ b/docs/content/changelog.md@@ -6,6 +6,306 @@ description: "Rclone Changelog"# Changelog+## v1.76.0 - 2026-11-06++[See commits](https://github.com/rclone/rclone/compare/v1.75.0...v1.76.0)++- New backends+ - [Dosya](/dosya/) (Kaya)+ - [Infomaniak kDrive](/kdrive/) (Guillaume MALLERET)+- New S3 Providers+ - [Autonomys Auto Drive](/s3/#autonomys) (Jim Counter)+ - [Clever Cloud Cellar](/s3/#clever-cloud-cellar) (David Legrand)+ - [Infomaniak Swiss Backup](/s3/#infomaniak) (Louis)+ - [Neon Object Storage](/s3/#neon) (Rishi Raj Jain)+ - [RelAix.Storage](/s3/#relaix) (mygrexit)+- New commands+ - [index](/commands/rclone_index/): Write static directory listings into a remote (Nick Craig-Wood)+- Security+ - build+ - Disable Multipath TCP listeners by default to mitigate CVE-2026-98432 (Nick Craig-Wood)+ - Fix multiple CVEs including CVE-2026-111943 by upgrading to go1.27.3 (Nick Craig-Wood)+ - Update `golang.org/x/net` to v0.62.0 to fix multiple CVEs (Nick Craig-Wood)+ - Update `github.com/andybalholm/brotli` to v1.2.7 to fix CVE-2026-111497 (Nick Craig-Wood)+ - fshttp: Refuse HTTPS to HTTP redirects unless `--allow-https-downgrade` GHSA-m97j-7x65-wh93 CVE-PENDING (Nick Craig-Wood)+ - lib/http: Stop `--htpasswd` and `--user` revealing which usernames exist through timing GHSA-hwj7-wp9g-wrv9 CVE-PENDING (Nick Craig-Wood)+ - local+ - Fix user metadata escaping the root through a planted symlink GHSA-r5wv-vqqh-2r9c CVE-PENDING (Nick Craig-Wood)+ - Fix MkdirMetadata finding directories outside the root through a planted symlink (Nick Craig-Wood)+ - Fix Remove and Rmdir deleting outside the root through a planted symlink (Nick Craig-Wood)+ - Fix DirMove creating the destination parent through a planted symlink (Nick Craig-Wood)+ - netstorage: Fix API auth headers leaking to other hosts on redirect GHSA-m97j-7x65-wh93 CVE-PENDING (Nick Craig-Wood)+ - rc+ - Limit request size when the rc server has no authentication GHSA-f8p4-8g4j-q3m2 CVE-PENDING (Nick Craig-Wood)+ - Reject unauthorised and unknown calls before reading the request GHSA-f8p4-8g4j-q3m2 CVE-PENDING (Nick Craig-Wood)+ - seafile: Fix HTTPS to HTTP redirects leaking credentials GHSA-m97j-7x65-wh93 CVE-PENDING (Nick Craig-Wood)+ - serve: Make auth proxy revocation take effect within 5 minutes GHSA-vmgx-g9c5-669p CVE-PENDING (Nick Craig-Wood)+ - serve sftp+ - Fix auth proxy letting a public key be used as a password GHSA-66qv-66mg-9h74 CVE-PENDING (Nick Craig-Wood)+ - Fix auth proxy running for public keys the client can't use GHSA-66qv-66mg-9h74 CVE-PENDING (Nick Craig-Wood)+ - swift+ - Fix auth token leaking to other hosts or over plaintext on redirect GHSA-m97j-7x65-wh93 CVE-PENDING (Nick Craig-Wood)+ - Fix password being resent to another host on auth redirect GHSA-m97j-7x65-wh93 CVE-PENDING (Nick Craig-Wood)+ - ulozto: Fix auth tokens leaking on redirects from download links GHSA-3xpx-hjg2-vrjh CVE-PENDING (Nick Craig-Wood)+- New Features+ - accounting+ - Stop `--max-transfer` overshooting the limit (Nick Craig-Wood)+ - Add "Updated dirs" count of directory modtime/metadata updates to the stats (Nick Craig-Wood)+ - batcher: Log when committing uploads has finished (phatlc)+ - bisync+ - Make `--max-delete` aware of `--track-renames` (Morax)+ - Fall back to `--download` when a checksum is missing (nielash)+ - Reject `--max-delete` above 100 (nielash)+ - build+ - Update to go1.27 and make go1.26 the minimum required version (Nick Craig-Wood)+ - Update all dependencies (Nick Craig-Wood)+ - cmd: Complete enum flag values from their choices (Demiao Huang)+ - config: Only run `--password-command` once when using `--daemon` (phatlc)+ - doc fixes (ashwinibailangadi, ferrumclaudepilgrim, frodleTian, FTCHD, gomitrah, Nick Craig-Wood, qodelicious, Satvik Mishra, ZainnQureshii)+ - docker: Make container user UID and GID configurable via build ARGs (Vaibhav Mashal)+ - filter+ - Speed up large `--include-from` lists of directories (kaii9)+ - Make `[!...]` in globs a negated character class (kevin9327)+ - Make `[^...]` in globs exclude `/` and document it (kevin9327)+ - fs+ - Reject sizes too large for SizeSuffix instead of silently disabling the limit (KBS)+ - Stop writing config for on the fly and environment variable remotes into the config file (Nick Craig-Wood)+ - Reject durations like 300y that overflow instead of wrapping (kevin9327)+ - Accept multi-unit relative times like 1d12h in Time options (kevin)+ - Reject NaN sizes and CountSuffix values that overflow (kevin9327)+ - Reject `--bwlimit` times without a colon or with signed digits (kevin9327)+ - Reject NaN durations instead of wrapping (Soujas K)+ - Add MIME type for JPEG XL (.jxl) files (Seth Westphal)+ - gui+ - Take the GUI out of beta (Nick Craig-Wood)+ - Update embedded release to 1.2.1 (Nick Craig-Wood)+ - hash: Add CRC-32C (Castagnoli) support (Dirk Petersen)+ - index: Add partial runs which only re-index changed directories (Nick Craig-Wood)+ - lib/http+ - Return 416 for ranges `--rc-serve` and serve restic can't satisfy (Rohit Behera)+ - Replace `github.com/abbot/go-http-auth` with built in htpasswd support (Nick Craig-Wood)+ - Accept the Basic auth scheme in any case (kevin)+ - lib/proxy: Send the decoded user and password in Proxy-Authorization (kevin)+ - lib/transform: Make titlecase capitalise words instead of upper-casing them (kevin)+ - ncdu: Show progress and the item count when deleting a selection (Lev Devaev)+ - operations+ - Check checksums in rcat with known size (phatlc)+ - Speed up checksum by hashing files in parallel (Teddy Tennant)+ - Fall back to a single stream when a multi-thread source ignores ranges (Morax)+ - Show directory operations in the `--progress` display (Nick Craig-Wood)+ - Add `operations/getfile` remote control API endpoint (Hakan İSMAİL)+ - Make `--immutable` work with copyto, moveto and single file copy (phatlc)+ - Respect `--backup-dir` in deletefile (Cao Yuhang)+ - rc+ - Deprecate `--rc-web-gui` flag (Nick Craig-Wood)+ - Add sync report parameters to `sync/sync`, `sync/copy` and `sync/move` (Nick Craig-Wood)+ - rest: Reject negative sizes in HTTP response headers (Yousef Issa)+ - serve http+ - Sort directory listings in the browser (Nick Craig-Wood)+ - Redesign the directory listing page (Nick Craig-Wood)+ - serve s3+ - Report the object's modtime as Last-Modified after a PUT (kevin9327)+ - Support changing object tags (Nick Craig-Wood)+ - Quote the ETags of objects in listings (Nick Craig-Wood)+ - Support listing with delimiters other than "/" (Nick Craig-Wood)+ - Support directory markers and refuse file/directory conflicts (Nick Craig-Wood)+ - Make the ETag of multipart uploads agree with HEAD and support partNumber (Nick Craig-Wood)+ - Make conditional writes atomic against other writes through the server (Nick Craig-Wood)+ - Make the ETag returned on upload agree with HEAD for any `--etag-hash` (Nick Craig-Wood)+ - List empty directories as directory markers (Nick Craig-Wood)+ - Allow copying directory markers (Nick Craig-Wood)+ - Remove the directories made for a failed upload (Nick Craig-Wood)+ - Keep the MD5 ETag of objects on remotes without MD5 (Nick Craig-Wood)+ - Don't read directories needlessly when listing with a delimiter (Nick Craig-Wood)+ - Don't compute the MD5 of multipart uploads unless the ETag is MD5 (Nick Craig-Wood)+ - List the directory markers put for directories holding other keys (Nick Craig-Wood)+ - Let a failed CompleteMultipartUpload be retried (Nick Craig-Wood)+ - List objects with the storage class they were put with (Nick Craig-Wood)+ - Add `--directory-markers` to list every directory as a marker (Nick Craig-Wood)+ - Remove a directory with its last key after its marker is deleted (Nick Craig-Wood)+ - Make `--no-cleanup` a deprecated alias of `--directory-markers` (Nick Craig-Wood)+ - Document which S3 features are supported and the design goals (Nick Craig-Wood)+ - serve webdav+ - Sort directory listings in the browser (Nick Craig-Wood)+ - Redesign the directory listing page (Nick Craig-Wood)+ - size: Include source in JSON output (Jiacheng Xu)+ - sync: Add pattern-based transfer ordering (Shurong Cao)+ - touch: Parse documented timestamp formats (Nicholas Velten)+- Bug Fixes+ - accounting: Fix totalBytes counting server-side copies and moves twice (ilya lavrenov)+ - bisync+ - Fix conflict winner missing from listings when Path2 wins (nielash)+ - Fix `--conflict-resolve` comparing a zero size or modtime (nielash)+ - Fix endless retry loop for skipped files (nielash)+ - Fix listings ignoring a modtime less than 1s older (nielash)+ - Fix rc ignoring global `--dry-run` and `--max-delete` settings (nielash)+ - Fix rc ignoring legacy backupdir1/2 and invalid values (nielash)+ - Fix internal error when a conflict rename reuses deleted names (TastyHeadphones)+ - fs: Fix durations like 1d12h being rejected (Rohit Behera)+ - lib/encoder: Fix decoding a trailing character after an escaped quote (kevin9327)+ - list: Fix ext sort error sometimes being ignored at the end of a listing (Nick Craig-Wood)+ - ncdu: Update to tcell v3 which fixes duplicated keys on Windows (Nick Craig-Wood)+ - operations+ - Fix making a directory with modtime asking twice with `--interactive` (Nick Craig-Wood)+ - Fix case- or Unicode-only renames deleting the file (nielash)+ - Fix moving a file onto its own name deleting it (nielash)+ - Fix cat `--offset` before the start of a file printing its end (b84330808)+ - Fix oversized getfile tail requests (唐僧洗发用飘柔)+ - rc: Fix LogLevel in `_config` being ignored (Nick Craig-Wood)+ - rest: Fix retries reading the request body after the request has finished (Nick Craig-Wood)+ - serve dlna: Fix crash on Browse with a negative StartingIndex or RequestedCount (phatlc)+ - serve restic: Fix temporary errors confusing restic into leaving locks behind (Azhar)+ - serve s3+ - Update to gofakes3 v0.0.8 which fixes 500 errors with go1.27 (Nick Craig-Wood)+ - Upgrade gofakes3 to v0.0.11 to fix IsTruncated and Docker Distribution Registry (DEVN)+ - Fix buckets being undeletable after an aborted multipart upload (Nick Craig-Wood)+ - Fix CopyObject metadata including the source's and an mtime (Nick Craig-Wood)+ - serve sftp: Fix rmdir deleting a file instead of returning an error (phatlc)+ - sync+ - Fix directory modtimes being set twice (Nick Craig-Wood)+ - Fix `--delete-empty-src-dirs` disabled by a prior error in a reused stats group (ghwany)+ - walk: Fix duplicate directories with `--fast-list` when directories are listed after their contents (Nick Craig-Wood)+- Mount+ - Fix VFS instance leaks on mount failures and unmounts (Hakan İSMAİL)+- VFS+ - Preserve shrinking truncate on uncached files (47th)+ - Fix crash renaming a directory while the dir cache expires (Shreyansh Jain)+ - Fix virtual entries leak in removed directory (Biagio Peccerillo)+ - Fix renamed items listed twice after a case or Unicode-only match (nielash)+ - Fix uploads never starting while a dirty file keeps being read (Neil Cawse)+- Local+ - Speed up default checksummed copies by writing in larger blocks (Dominik Sander)+ - Make out of space errors fatal during multi-thread transfers (ferrumclaudepilgrim)+ - Stop `--copy-links` following symlink loops (phatlc)+ - Limit the size of symlink targets read from .rclonelink files (Nick Craig-Wood)+ - Fix server-side copy on macOS when the destination exists (Ayush Sharma)+ - Set only the ownership a non-root user may set (Dirk Petersen)+- Crypt+ - Fix nil pointer panic in ObjectInfo.Hash (fixfon)+- Archive+ - Read symlinks stored in zip and squashfs archives (Lev Devaev)+ - Fix reading an archive in the root of a remote (Oshino Penguin)+- Azure Blob+ - Enable on Solaris (acoeur)+ - Fix directories on storage accounts with hierarchical namespace (Nick Craig-Wood)+ - Fix NewObject on a directory returning "is not a regular file" (Nick Craig-Wood)+ - Fix listing a file as a directory on hierarchical namespace accounts (Nick Craig-Wood)+ - Fix sync at the root failing on soft deleted containers (Ludovic Bondon)+- Box+ - Reuse multipart upload part buffers via the global pool (Nick Craig-Wood)+ - Fix ~ and environment variables not being expanded in `box_config_file` (Nick Craig-Wood)+- Chunker+ - Clean stale chunks after server-side overwrite (hahahahahayesyeseys)+- Compress+ - Buffer small uploads in the global memory pool (Nick Craig-Wood)+ - Stop buffering the compressibility check output in memory (Nick Craig-Wood)+- Drime+ - Fix endless listing of directories with more than 20,000 entries (Maksym Tymoshyk)+- Drive+ - Fetch metadata permissions concurrently (Loi Nguyen)+ - Reuse resumable-upload chunk buffers via the global memory pool (SillyZir)+ - Stop a fast-list error from hanging the sync (Sash)+- Dropbox+ - Reuse upload chunk buffers via the global memory pool (Nick Craig-Wood)+- Filelu+ - Reuse multipart upload buffers via the global pool and retry failed parts (Nick Craig-Wood)+ - Encode multipart init paths (wjc)+- FTP+ - Discard connections that fail with non-FTP errors (Daniel Bae)+ - Fix ranged reads failing when the server answers 425 to an early close (Robert Kubiś)+- Hasher+ - Keep cached hash when only the modtime changes (Aniket)+- Hidrive+ - Stop copying every upload chunk a second time (Nick Craig-Wood)+ - Buffer the start of each upload from the global memory pool (Nick Craig-Wood)+ - Reuse chunked upload buffers via the global memory pool (Nick Craig-Wood)+ - Buffer small file updates from the global memory pool (Nick Craig-Wood)+ - Fix truncated file left behind when an upload hits `--max-transfer` (Nick Craig-Wood)+- HTTP+ - Add metadata defaults for `--http-no-head` (Giridhar)+- Huaweidrive+ - Build small uploads in pooled memory instead of a growing buffer (Nick Craig-Wood)+ - Reuse resumable-upload chunk buffers via the global pool (Nick Craig-Wood)+ - Fix creating, copying and renaming files with non-ASCII names (Nick Craig-Wood)+- Iclouddrive+ - Mark the backend as case insensitive (berce)+- Imagekit+ - Stop multipart writers after failed uploads (Florian Josef Altendorfer)+- Jottacloud+ - Buffer MD5 pre-reads in the global memory pool (Nick Craig-Wood)+ - Fix server-side copy onto a trashed file failing as corrupted (Nick Craig-Wood)+ - Fix server-side copy onto a trashed file keeping the old name case (Nick Craig-Wood)+ - Fix authentication with whitelabel services elkjop and elko (albertony)+ - Fix traditional authentication failing with invalid_client (albertony)+- Linkbox+ - Buffer the hashed 10 MiB file prefix in the global memory pool (Nick Craig-Wood)+- Mailru+ - Buffer speedup hashing in the global memory pool and fix upload retries (Nick Craig-Wood)+- Onedrive+ - OneDrive for Business and SharePoint users must upgrade to v1.76+ before April 1, 2027 (Nick Craig-Wood)+ - Authenticate SharePoint downloads, uploads and copies after tempauth retirement (Nick Craig-Wood)+ - Return an error if a ranged download doesn't return the requested range (Nick Craig-Wood)+ - Reuse upload chunk buffers via the global memory pool (Nick Craig-Wood)+ - Allow `--onedrive-upload-cutoff` up to the documented limit of 250 MiB (Can Arslan)+ - Add configurable SharePoint tenant API version (Mattias Michaux)+ - Make `--onedrive-delta` quicker when not listing from the drive root (Nick Craig-Wood)+ - Keep the delta token when a ChangeNotify poll fails (xiaozou-wine)+ - Show the size of directories in listings (GingerAdonis)+ - Poll async copy status with the configured HTTP client (Soujas K)+- Opendrive+ - Reuse upload chunk buffers via the global memory pool (Nick Craig-Wood)+- Oracleobjectstorage+ - Add server-side object moves (tomaszni)+ - Encode carriage returns and line feeds (tomaszni)+ - Purge objects in batches (tomaszni)+ - Retry 409 ConcurrentObjectUpdate errors (berce)+- Pcloud+ - Stop multipart writers after failed uploads (Florian Josef Altendorfer)+- Pikpak+ - Share upload chunk buffers with the global memory pool (Nick Craig-Wood)+ - Use pooled memory for the gcid hash buffer and spool unknown-size uploads to disk (Nick Craig-Wood)+ - Fix downloads that return an HTTP error page as the file (Cestercian)+ - Stop multipart writers after failed uploads (Florian Josef Altendorfer)+ - Fix uploads failing when the server is slow to complete them (Nick Craig-Wood)+- Putio+ - Reuse upload chunk buffers via the global memory pool (Nick Craig-Wood)+- S3+ - Avoid buffering Object Lock uploads when the source MD5 is known (Nick Craig-Wood)+ - Add backend link command with signed response header overrides (eliotee)+ - Add Fastly EU Central 1 region (Leon Brocard)+ - Add the MD5 of multipart chunks to the debug log (interested.tortoise)+ - Report the real storage class when an object needs restoring (Dirk Petersen)+ - Speed up multi-thread uploads by checksumming chunks in 1 MiB reads (Roman Butor)+ - Add IT-MIL region to Scaleway (Alessio Margelli)+ - Add missing Huawei OBS regions (Enbiya Göral)+ - Fix rmdir of the root directory with `--s3-directory-markers` (Nick Craig-Wood)+ - Fix presigned URLs for GCS, Linode and Ceph (Leon Brocard)+- Seafile+ - Fix operations on files whose names are not in Unicode NFC form (Fred)+ - Stop multipart writers after failed uploads (Florian Josef Altendorfer)+ - Fix double slash in the URL when getting a 2FA token (Nick Craig-Wood)+- SFTP+ - Implement multi-thread uploads (Splainte)+ - Fix data race when using `--sftp-ssh` (Dirk Petersen)+ - Fix hash check for names starting with "-" with relative `--sftp-path-override` (Dirk Einecke)+ - Fix sync deleting files with control characters in their names (nielash)+ - Fix hashing files with a newline in their name on rsync.net (Nick Craig-Wood)+- Shade+ - Stop copying each upload chunk into a second heap buffer (Nick Craig-Wood)+- Smb+ - Add workstation field to smb backend (Kalin Stoyanov)+- Swift+ - Add Infomaniak Swiss Backup provider (Louis)+- WebDAV+ - Reject ranged responses which ignore the range (Morax)+ - Retry Nextcloud chunk uploads individually (Stochastic)+ - Fix Nextcloud public share access by sending X-Requested-With (drakeo338)+ - Fix memory leak uploading to Infinite Scale (jxj)+- Yandex+ - Add `app_folder` option to support `cloud_api:disk.app_folder` OAuth scope (Sanjay Kanth A)+- Zoho+ - Stop multipart writers after failed uploads (Florian Josef Altendorfer)+## v1.75.2 - 2026-10-09[See commits](https://github.com/rclone/rclone/compare/v1.75.1...v1.75.2)diff --git a/docs/content/commands/rclone.md b/docs/content/commands/rclone.mdindex 33446e0ae..2ac7491da 100644--- a/docs/content/commands/rclone.md+++ b/docs/content/commands/rclone.md@@ -24,6 +24,7 @@ rclone [flags]```--alias-description string Description of the remote--alias-remote string Remote or path to alias+ --allow-https-downgrade Follow redirects from HTTPS to HTTP (insecure)--archive-description string Description of the remote--archive-remote string Remote to wrap to read archives from--ask-password Allow prompt for password for encrypted configuration (default true)@@ -50,8 +51,10 @@ rclone [flags]--azureblob-encoding Encoding The encoding for the backend (default Slash,BackSlash,Del,Ctl,RightPeriod,InvalidUtf8)--azureblob-endpoint string Endpoint for the service--azureblob-env-auth Read credentials from runtime (environment variables, CLI or MSI)+ --azureblob-hns Tristate Set if the storage account has a hierarchical namespace (default unset)--azureblob-key string Storage Account Shared Key--azureblob-list-chunk int Size of blob list (default 5000)+ --azureblob-list-parallelism int Number of parallel shards to list a directory with--azureblob-msi-client-id string Object ID of the user-assigned MSI to use, if any--azureblob-msi-mi-res-id string Azure resource ID of the user-assigned MSI to use, if any--azureblob-msi-object-id string Object ID of the user-assigned MSI to use, if any@@ -64,6 +67,7 @@ rclone [flags]--azureblob-tenant string ID of the service principal's tenant. Also called its directory ID--azureblob-upload-concurrency int Concurrency for multipart uploads (default 16)--azureblob-upload-cutoff string Cutoff for switching to chunked upload (<= 256 MiB) (deprecated)+ --azureblob-use-arrow-list Use the Apache Arrow listing format--azureblob-use-az Use Azure CLI tool az for authentication--azureblob-use-copy-blob Whether to use the Copy Blob API when copying to the same storage account (default true)--azureblob-use-emulator Uses local storage emulator if provided as 'true'@@ -224,6 +228,12 @@ rclone [flags]--doi-doi string The DOI or the doi.org URL--doi-doi-resolver-api-url string The URL of the DOI resolver API to use--doi-provider string DOI provider+ --dosya-api-key string Your API Key, get it from https://dosya.dev/profile under "API keys"+ --dosya-api-url string API base URL (default "https://api.dosya.dev")+ --dosya-description string Description of the remote+ --dosya-encoding Encoding The encoding for the backend (default Slash,LtGt,DoubleQuote,SingleQuote,BackSlash,Del,Ctl,LeftSpace,RightSpace,InvalidUtf8,Dot)+ --dosya-upload-concurrency int Concurrency for multipart uploads (default 4)+ --dosya-workspace-id string Your workspace ID--drime-access-token string API Access token--drime-chunk-size SizeSuffix Chunk size to use for uploading (default 5Mi)--drime-description string Description of the remote@@ -470,6 +480,8 @@ rclone [flags]--http-headers CommaSepList Set HTTP headers for all transactions--http-no-escape Do not escape URL metacharacters in path names--http-no-head Don't use HEAD requests+ --http-no-head-size SizeSuffix Set the reported file size for --http-no-head (default off)+ --http-no-head-time Time Set a default modification time for --http-no-head (default off)--http-no-slash Set this if the site doesn't end directories with /--http-proxy string HTTP proxy URL--http-url string URL of HTTP host to connect to@@ -543,6 +555,17 @@ rclone [flags]--jottacloud-token-url string Token server url--jottacloud-trashed-only Only show files that are in the trash--jottacloud-upload-resume-limit SizeSuffix Files bigger than this can be resumed if the upload fail's (default 10Mi)+ --kdrive-auth-url string Auth server URL+ --kdrive-chunk-size SizeSuffix Files above this size are uploaded in chunks of this size (default 20Mi)+ --kdrive-client-id string OAuth Client Id+ --kdrive-description string Description of the remote+ --kdrive-drive-id string ID of the kDrive to use+ --kdrive-encoding Encoding The encoding for the backend (default Slash,Del,Ctl,LeftSpace,RightSpace,InvalidUtf8,Dot)+ --kdrive-endpoint string The API endpoint to use+ --kdrive-root-folder-id string The folder to use as root+ --kdrive-token string OAuth Access Token as a JSON blob+ --kdrive-token-url string Token server url+ --kdrive-upload-concurrency int Concurrency for chunked uploads (default 4)--koofr-description string Description of the remote--koofr-encoding Encoding The encoding for the backend (default Slash,BackSlash,Del,Ctl,InvalidUtf8,Dot)--koofr-endpoint string The Koofr API endpoint to use@@ -695,6 +718,7 @@ rclone [flags]--onedrive-root-folder-id string ID of the root folder--onedrive-server-side-across-configs Deprecated: use --server-side-across-configs instead--onedrive-tenant string ID of the service principal's tenant. Also called its directory ID+ --onedrive-tenant-api-version string The SharePoint API version to use with tenant_url (default "v2.0")--onedrive-tenant-url string The tenant URL for non-admin OneDrive access--onedrive-token string OAuth Access Token as a JSON blob--onedrive-token-url string Token server url@@ -709,7 +733,7 @@ rclone [flags]--oos-decompress If set this will decompress gzip encoded objects--oos-description string Description of the remote--oos-disable-checksum Don't store MD5 checksum with object metadata- --oos-encoding Encoding The encoding for the backend (default Slash,InvalidUtf8,Dot)+ --oos-encoding Encoding The encoding for the backend (default Slash,CrLf,InvalidUtf8,Dot)--oos-endpoint string Endpoint for Object storage API--oos-leave-parts-on-error If true avoid calling abort upload on a failure, leaving all successfully uploaded parts for manual recovery--oos-max-upload-parts int Maximum number of parts in a multipart upload (default 10000)@@ -840,11 +864,6 @@ rclone [flags]--rc-template string User-specified template--rc-user string User name for authentication--rc-user-from-header string User name from a defined HTTP header- --rc-web-fetch-url string URL to fetch the releases for webgui (default "https://api.github.com/repos/rclone/rclone-webui-react/releases/latest")- --rc-web-gui Launch WebGUI on localhost- --rc-web-gui-force-update Force update to latest version of web gui- --rc-web-gui-no-open-browser Don't open the browser automatically- --rc-web-gui-update Check and update to latest version of web gui--refresh-times Refresh the modtime of remote files--retries int Retry operations this many times if they fail (default 3)--retries-sleep Duration Interval between retrying operations if they fail, e.g. 500ms, 60s, 5m (0 to disable) (default 0s)@@ -945,7 +964,7 @@ rclone [flags]--sftp-disable-concurrent-reads If set don't use concurrent reads--sftp-disable-concurrent-writes If set don't use concurrent writes--sftp-disable-hashcheck Disable the execution of SSH commands to determine if remote file hashing is available- --sftp-encoding Encoding The encoding for the backend (default Slash,Del,Ctl,Dot)+ --sftp-encoding Encoding The encoding for the backend (default Slash,Dot)--sftp-hashes CommaSepList Comma separated list of supported checksum types--sftp-host string SSH host to connect to--sftp-host-key-algorithms SpaceSepList Space separated list of host key algorithms, ordered by preference@@ -960,6 +979,7 @@ rclone [flags]--sftp-known-hosts-file string Optional path to known_hosts file--sftp-macs SpaceSepList Space separated list of MACs (message authentication code) algorithms, ordered by preference--sftp-md5sum-command string The command used to read MD5 hashes+ --sftp-multithread-upload Set this to use multi-thread (multi-connection) uploads--sftp-pass string SSH password, leave blank to use ssh-agent (obscured)--sftp-path-override string Override path used by SSH shell commands--sftp-pin-host-key Pin the server host key on first connection (Trust On First Use)@@ -1023,7 +1043,8 @@ rclone [flags]--smb-port int SMB port number (default 445)--smb-spn string Service principal name--smb-use-kerberos Use Kerberos authentication- --smb-user string SMB username (default "$USER")+ --smb-user string SMB username+ --smb-workstation string Workstation name sent in NTLM authentication--stats Duration Interval between printing stats, e.g. 500ms, 60s, 5m (0 to disable) (default 1m0s)--stats-file-name-length int Max file name length in stats (0 for no limit) (default 45)--stats-log-level LogLevel Log level to show --stats output DEBUG|INFO|NOTICE|ERROR (default INFO)@@ -1107,7 +1128,7 @@ rclone [flags]--use-json-log Use json log format--use-mmap Use mmap allocator (see docs)--use-server-modtime Use server modified time instead of object metadata- --user-agent string Set the user-agent to a specified string (default "rclone/v1.75.0")+ --user-agent string Set the user-agent to a specified string (default "rclone/v1.76.0")-v, --verbose count Print lots more stuff (repeat for more)-V, --version Print the version number--webdav-auth-redirect Preserve authentication on redirect@@ -1125,6 +1146,7 @@ rclone [flags]--webdav-url string URL of http host to connect to--webdav-user string User name--webdav-vendor string Name of the WebDAV site/service/software you are using+ --yandex-app-folder Use the application folder as the root--yandex-auth-url string Auth server URL--yandex-client-credentials Use client credentials OAuth flow--yandex-client-id string OAuth Client Id@@ -1183,6 +1205,7 @@ rclone [flags]* [rclone gitannex](/commands/rclone_gitannex/) - Speaks with git-annex over stdin/stdout.* [rclone gui](/commands/rclone_gui/) - Open the web based GUI.* [rclone hashsum](/commands/rclone_hashsum/) - Produces a hashsum file for all the objects in the path.+* [rclone index](/commands/rclone_index/) - Write static directory listings into a remote.* [rclone link](/commands/rclone_link/) - Generate public link to file/folder.* [rclone listremotes](/commands/rclone_listremotes/) - List all the remotes in the config file and defined in environment variables.* [rclone ls](/commands/rclone_ls/) - List the objects in the path with size and path.diff --git a/docs/content/commands/rclone_bisync.md b/docs/content/commands/rclone_bisync.mdindex c74c2a5e4..4d1ce16e3 100644--- a/docs/content/commands/rclone_bisync.md+++ b/docs/content/commands/rclone_bisync.md@@ -52,6 +52,7 @@ rclone bisync remote1:path1 remote2:path2 [flags]--force Bypass --max-delete safety check and run the sync. Consider using with --verbose-h, --help help for bisync--ignore-listing-checksum Do not use checksums for listings (add --ignore-checksum to additionally skip post-copy checksum checks)+ --max-delete-renames-aware Exclude tracked renames from the --max-delete safety check (requires --track-renames)--max-lock Duration Consider lock files older than this to be expired (default: 0 (never expire)) (minimum: 2m) (default 0s)--no-cleanup Retain working files (useful for troubleshooting and testing).--no-slow-hash Ignore listing checksums only on backends where they are slowdiff --git a/docs/content/commands/rclone_completion.md b/docs/content/commands/rclone_completion.mdindex 2f43ab108..a06fc10fc 100644--- a/docs/content/commands/rclone_completion.md+++ b/docs/content/commands/rclone_completion.md@@ -14,6 +14,8 @@ Output completion script for a given shell.Generates a shell completion script for rclone.Run with `--help` to list the supported shells.+Enum-valued flags such as `--log-level` and `--vfs-cache-mode`+also complete their supported values.## Optionsdiff --git a/docs/content/commands/rclone_convmv.md b/docs/content/commands/rclone_convmv.mdindex 379c8e17a..00f7e13c5 100644--- a/docs/content/commands/rclone_convmv.md+++ b/docs/content/commands/rclone_convmv.md@@ -231,12 +231,12 @@ rclone convmv "stories/The Quick Brown Fox!.txt" --name-transform "all,command=e```consolerclone convmv "stories/The Quick Brown Fox!" --name-transform "date=-{YYYYMMDD}"-// Output: stories/The Quick Brown Fox!-20260731+// Output: stories/The Quick Brown Fox!-20261106``````consolerclone convmv "stories/The Quick Brown Fox!" --name-transform "date=-{macfriendlytime}"-// Output: stories/The Quick Brown Fox!-2026-07-31 0340PM+// Output: stories/The Quick Brown Fox!-2026-11-06 0220PM``````consolediff --git a/docs/content/commands/rclone_deletefile.md b/docs/content/commands/rclone_deletefile.mdindex 19143176c..87c9a4d8d 100644--- a/docs/content/commands/rclone_deletefile.md+++ b/docs/content/commands/rclone_deletefile.md@@ -10,9 +10,11 @@ Remove a single file from remote.## Synopsis-Remove a single file from remote. Unlike `delete` it cannot be used to+Remove a single file from remote. Unlike `delete` it cannot be used toremove a directory and it doesn't obey include/exclude filters - if the-specified file exists, it will always be removed.+specified file exists, it will always be removed from its original location.++The `--backup-dir` and `--suffix` flags are honored when set.```rclone deletefile remote:path [flags]@@ -27,6 +29,26 @@ rclone deletefile remote:path [flags]Options shared with other commands are described next.See the [global flags page](/flags/) for global options not listed here.+### Sync Options++Flags used for sync commands++```text+ --backup-dir string Make backups into hierarchy based in DIR+ --delete-after When synchronizing, delete files on destination after transferring (default)+ --delete-before When synchronizing, delete files on destination before transferring+ --delete-during When synchronizing, delete files during transfer+ --fix-case Force rename of case insensitive dest to match source+ --ignore-errors Delete even if there are I/O errors+ --list-cutoff int To save memory, sort directory listings on disk above this threshold (default 1000000)+ --max-delete int When synchronizing, limit the number of deletes (default -1)+ --max-delete-size SizeSuffix When synchronizing, limit the total size of deletes (default off)+ --suffix string Suffix to add to changed files+ --suffix-keep-extension Preserve the extension when using --suffix+ --track-renames When synchronizing, track file renames and do a server-side move if possible+ --track-renames-strategy string Strategies to use when synchronizing using track-renames hash|modtime|leaf (default "hash")+```+### Important OptionsImportant flags useful for most commandsdiff --git a/docs/content/commands/rclone_gui.md b/docs/content/commands/rclone_gui.mdindex 77c4e54ec..98ea25a76 100644--- a/docs/content/commands/rclone_gui.md+++ b/docs/content/commands/rclone_gui.md@@ -101,11 +101,6 @@ Flags to control the Remote Control API--rc-template string User-specified template--rc-user string User name for authentication--rc-user-from-header string User name from a defined HTTP header- --rc-web-fetch-url string URL to fetch the releases for webgui (default "https://api.github.com/repos/rclone/rclone-webui-react/releases/latest")- --rc-web-gui Launch WebGUI on localhost- --rc-web-gui-force-update Force update to latest version of web gui- --rc-web-gui-no-open-browser Don't open the browser automatically- --rc-web-gui-update Check and update to latest version of web gui```## See Alsodiff --git a/docs/content/commands/rclone_hashsum.md b/docs/content/commands/rclone_hashsum.mdindex 4d8fa3796..873b8ea7c 100644--- a/docs/content/commands/rclone_hashsum.md+++ b/docs/content/commands/rclone_hashsum.md@@ -41,6 +41,7 @@ Supported hashes are:- blake3- xxh3- xxh128+- crc32c```Thendiff --git a/docs/content/commands/rclone_index.md b/docs/content/commands/rclone_index.mdnew file mode 100644index 000000000..eed11006d--- /dev/null+++ b/docs/content/commands/rclone_index.md@@ -0,0 +1,331 @@+---+title: "rclone index"+description: "Write static directory listings into a remote."+versionIntroduced: v1.76+# autogenerated - DO NOT EDIT, instead edit the source code in cmd/index/ and as part of making a release run "make commanddocs"+---+# rclone index++Write static directory listings into a remote.++## Synopsis++Write a directory listing, `index.html` by default, into every+directory of `remote:path` so that it can be browsed when it is served+as a static website, for example from an S3 website endpoint, a bucket+behind a CDN, or a plain web server.++This works like `rclone sync`: each run makes the minimum changes needed+to bring the listings into line with the files. Listings which haven't+changed aren't rewritten, so a run with no changes makes no uploads. The+listings never show the listing files themselves.++**Note** that `rclone index` treats every file with an output name+(`index.html` by default) in the directories it indexes as its own, and+will overwrite it or delete it without checking where it came from. To+protect a hand-written index, exclude its directory with+`--index-exclude`.++## Outputs++`--output NAME=FORMAT` says what to write in each directory and can be+repeated. NAME is the file name, and FORMAT is one of the built-in+formats below or the path of a Go template file.++| Format | MIME type | Contents |+|:-------|:----------|:---------|+| `html` | `text/html; charset=utf-8` | The HTML listing from the `rclone serve http` template |+| `json` | `application/json` | JSON in the format of `rclone lsjson` for the directory |+| `caddy` | `application/json` | JSON in the format of Caddy's `file_server browse` |++The default is `--output index.html=html`. The built-in formats are+templates, so they can be printed with `--print-template FORMAT`, copied+and changed. Outputs whose name ends in `.html` or `.htm` are rendered+with Go's `html/template` and everything else with `text/template`, which+has a `json` function for safe encoding. The template data is that of+`rclone serve http` with `.Static` set (see+[rclone serve http](/commands/rclone_serve_http/#template)). Outputs are+uploaded with the MIME type of the built-in format, or that of the+output name's extension for template outputs, since static hosts serve+the stored type.++## Filters++The normal filters (`--include`, `--exclude`, `--filter-from`, `--max-age`, ...)+control what appears in the listings. An excluded file isn't listed, and+an excluded directory isn't listed or indexed.++The `--index-include`, `--index-exclude`, `--index-filter` flags and their+`-from` variants control which directories get listings. They take the same+patterns as the normal filters, applied to directory paths, and+`--index-include` implies excluding all other directories. A directory+excluded here still appears in its parent's listing, but nothing is written+or deleted inside it, so it can have its own hand-made index.+`--index-max-depth` limits how deep listings are written.++## Modification times++Listings show the modification time of each file. When using the s3,+oracleobjectstorage and swift backends rclone stores the modification time as+metadata which isn't returned by the bucket listing, so **every object needs+a HEAD request**, which can mean thousands of transactions per run. Avoid+this with one of:++- `--use-server-modtime` uses the time the object was uploaded instead,+ which is what the web server sends as `Last-Modified`. Recommended for s3+ unless the original modification times matter.+- `--no-modtime` shows no times at all, which is the cheapest option.++`--dir-time` sets the time shown for directories:++- `newest` (default): the newest modification time of anything below the+ directory. Note that this means one new file changes the listing of+ every directory above it.+- `dir`: the directory's own modification time. On most file systems+ writing a listing into a directory changes its time, so this can take+ more than one run to settle.+- `none`: no time.++The listing files are given the newest modification time below their+directory, so the `Last-Modified` header of a listing is meaningful.++## Links++Directory links are `dir/` by default, which works on hosts that serve+index documents: S3 website endpoints, GCS and Azure static websites,+nginx and Apache, and R2 custom domains with a URL rewrite rule. For+hosts that don't (plain S3 REST URLs, B2 friendly URLs) use+`--link-index` to make directory links `dir/index.html` instead.++## Partial runs++After a sync which changed a known set of files there is no need to+walk the whole remote. Tell `rclone index` what changed and it re-indexes+only the directories that could be affected:++ rclone index r2:bucket --changed v1.76.0/ --changed version.txt+ rclone index r2:bucket --changed-from changes.txt+ rclone sync ./public r2:bucket --exclude index.html --combined - | rclone index r2:bucket --changed-combined -++- `--changed PATH` names a changed file or directory relative to+ `remote:path` and can be repeated. A directory, given with a trailing+ slash or found to be one, is re-indexed completely.+- `--changed-from FILE` reads one path per line exactly as written.+- `--changed-combined FILE` reads the `--combined` report of+ `rclone sync`: lines starting with `+`, `-`, `*` and `!` are changes and+ `=` lines are ignored.+- Both files accept `-` for standard input and all three flags can be+ combined.++A changed file re-indexes its directory and every directory above it,+since their times and their listings can change too, and on bucket+based storage a directory can cease to exist. Each of those is one+directory listing. With `--dir-time newest` the directories above also+need the times of their unchanged subdirectories, which are read from+the modification times of those subdirectories' listings at the cost+of one small request each. With `--use-server-modtime` that is the time+the listing was written rather than the time of the newest file, so+partial runs can drift by the indexing delay until the next full run.+`--dir-time dir` and `none` need no such lookups.++Directories which are neither named nor above a named path are left+alone, as are directories deeper than `--max-depth`. An empty change+list does nothing, and naming the root does a full run. There is no+automatic cutover to a full run since that depends on the size of the+whole remote: a partial run costs one listing per affected directory,+a full run one request per thousand objects on S3-like storage or one+per directory elsewhere. The number+of directories a partial run will list is logged with `-v`, and+`--changed-max-dirs` makes it fall back to a full run above that+number. A good pattern is a partial run after each upload and a+scheduled full run to catch anything missed.++## Site icon++The listings carry no icon or branding of their own. To give a site an+icon, upload a `favicon.ico` to the root of the remote, which browsers+fetch from there for every page. Exclude it from the listings with+`--exclude /favicon.ico` so it stays on the site but isn't shown.++## Deleting listings++On backends which can't have empty directories (S3, B2 and other+bucket-based storage) a directory only exists while it has files in it.+Once the last file in a directory has been removed its listing is+deleted, since the directory only existed because of it. Backends which+can have empty directories keep a listing for each empty directory.++## Using with sync++A sync from a local build deletes files on the destination which aren't+in the source, which includes the listings. Either index the local build+before syncing:++ rclone index ./public+ rclone sync ./public r2:bucket++or exclude the listings from the sync so it neither uploads nor deletes+them, then index the remote:++ rclone sync ./public r2:bucket --exclude index.html+ rclone index r2:bucket++## Examples++An S3 website bucket, using the upload times so no object needs a HEAD+request:++ rclone index s3:my-site --use-server-modtime++HTML and rclone JSON listings, setting `Cache-Control` on them:++ rclone index s3:bucket --output index.html=html --output index.json=json --header-upload "Cache-Control: max-age=300"++Only list release files, and don't write listings under `/private`:++ rclone index r2:bucket --include "*.{zip,deb,rpm}" --index-exclude "/private/**"++A B2 bucket served from its friendly URLs:++ rclone index b2:bucket --link-index++See what would change:++ rclone index r2:bucket --dry-run -v++Listings which haven't changed aren't normally rewritten, so changing+the headers or metadata set on them needs `--index-rewrite`, which+writes every listing a run would render whether it changed or not:++ rclone index r2:bucket --index-rewrite --header-upload "Cache-Control: max-age=3600"++```+rclone index remote:path [flags]+```++## Options++```+ --changed stringArray Only re-index directories affected by this changed path (directories end in /)+ --changed-combined stringArray Read changed paths from a sync --combined report (use - to read from stdin)+ --changed-from stringArray Read changed paths from file, one per line (use - to read from stdin)+ --changed-max-dirs int Do a full index if a partial one would list more than this many directories (0 for no limit)+ --dir-time newest|dir|none How to work out the time shown for a directory (default newest)+ -h, --help help for index+ --index-exclude stringArray Don't write listings in directories matching pattern+ --index-exclude-from stringArray Read directory exclude patterns from file (use - to read from stdin)+ --index-filter stringArray Add a rule for which directories get listings+ --index-filter-from stringArray Read directory rules from a file (use - to read from stdin)+ --index-include stringArray Only write listings in directories matching pattern+ --index-include-from stringArray Read directory include patterns from file (use - to read from stdin)+ --index-max-depth int Only write listings this many directories deep (-1 for no limit) (default -1)+ --index-rewrite Write every listing even if it is unchanged+ --link-index Link to dir/NAME rather than dir/ for hosts without index documents+ --no-modtime Don't show modification times in listings+ --output stringArray Listing to write in each directory as NAME=FORMAT (default [index.html=html])+ --print-template string Print the built-in template for FORMAT and exit+```++Options shared with other commands are described next.+See the [global flags page](/flags/) for global options not listed here.++### Copy Options++Flags for anything which can copy a file++```text+ --check-first Do all the checks before starting transfers+ -c, --checksum Check for changes with size & checksum (if available, or fallback to size only)+ --compare-dest stringArray Include additional server-side paths during comparison+ --copy-dest stringArray Implies --compare-dest but also copies files from paths into destination+ --cutoff-mode HARD|SOFT|CAUTIOUS Mode to stop transfers when reaching the max transfer limit HARD|SOFT|CAUTIOUS (default HARD)+ --ignore-case-sync Ignore case when synchronizing+ --ignore-checksum Skip post copy check of checksums+ --ignore-existing Skip all files that exist on destination+ --ignore-size Ignore size when skipping use modtime or checksum+ -I, --ignore-times Don't skip items that match size and time - transfer all unconditionally+ --immutable Do not modify files, fail if existing files have been modified+ --inplace Download directly to destination file instead of atomic download to temp/rename+ -l, --links Translate symlinks to/from regular files with a '.rclonelink' extension+ --max-backlog int Maximum number of objects in sync or check backlog (default 10000)+ --max-duration Duration Maximum duration rclone will transfer data for (default 0s)+ --max-transfer SizeSuffix Maximum size of data to transfer (default off)+ -M, --metadata If set, preserve metadata when copying objects+ --modify-window Duration Max time diff to be considered the same (default 1ns)+ --multi-thread-chunk-size SizeSuffix Chunk size for multi-thread downloads / uploads, if not set by filesystem (default 64Mi)+ --multi-thread-cutoff SizeSuffix Use multi-thread downloads for files above this size (default 256Mi)+ --multi-thread-streams int Number of streams to use for multi-thread downloads (default 4)+ --multi-thread-write-buffer-size SizeSuffix In memory buffer size for writing when in multi-thread mode (default 128Ki)+ --name-transform stringArray Transform paths during the copy process+ --no-check-dest Don't check the destination, copy regardless+ --no-traverse Don't traverse destination file system on copy+ --no-update-dir-modtime Don't update directory modification times+ --no-update-modtime Don't update destination modtime if files identical+ --order-by string Instructions on how to order the transfers, e.g. 'size,descending'+ --partial-suffix string Add partial-suffix to temporary file name when --inplace is not used (default ".partial")+ --refresh-times Refresh the modtime of remote files+ --server-side-across-configs Allow server-side operations (e.g. copy) to work across different configs+ --size-only Skip based on size only, not modtime or checksum+ --streaming-upload-cutoff SizeSuffix Cutoff for switching to chunked upload if file size is unknown, upload starts after reaching cutoff or when file ends (default 100Ki)+ -u, --update Skip files that are newer on the destination+```++### Important Options++Important flags useful for most commands++```text+ -n, --dry-run Do a trial run with no permanent changes+ -i, --interactive Enable interactive mode+ -v, --verbose count Print lots more stuff (repeat for more)+```++### Filter Options++Flags for filtering directory listings++```text+ --delete-excluded Delete files on dest excluded from sync+ --exclude stringArray Exclude files matching pattern+ --exclude-from stringArray Read file exclude patterns from file (use - to read from stdin)+ --exclude-if-present stringArray Exclude directories if filename is present+ --files-from stringArray Read list of source-file names from file (use - to read from stdin)+ --files-from-raw stringArray Read list of source-file names from file without any processing of lines (use - to read from stdin)+ --files-from0 stringArray Read list of source-file names from file using NUL as separator (use - to read from stdin)+ -f, --filter stringArray Add a file filtering rule+ --filter-from stringArray Read file filtering patterns from a file (use - to read from stdin)+ --hash-filter string Partition filenames by hash k/n or randomly @/n+ --ignore-case Ignore case in filters (case insensitive)+ --include stringArray Include files matching pattern+ --include-from stringArray Read file include patterns from file (use - to read from stdin)+ --max-age Duration Only transfer files younger than this in s or suffix ms|s|m|h|d|w|M|y (default off)+ --max-depth int If set limits the recursion depth to this (default -1)+ --max-size SizeSuffix Only transfer files smaller than this in KiB or suffix B|K|M|G|T|P (default off)+ --metadata-exclude stringArray Exclude metadatas matching pattern+ --metadata-exclude-from stringArray Read metadata exclude patterns from file (use - to read from stdin)+ --metadata-filter stringArray Add a metadata filtering rule+ --metadata-filter-from stringArray Read metadata filtering patterns from a file (use - to read from stdin)+ --metadata-include stringArray Include metadatas matching pattern+ --metadata-include-from stringArray Read metadata include patterns from file (use - to read from stdin)+ --min-age Duration Only transfer files older than this in s or suffix ms|s|m|h|d|w|M|y (default off)+ --min-size SizeSuffix Only transfer files bigger than this in KiB or suffix B|K|M|G|T|P (default off)+```++### Listing Options++Flags for listing directories++```text+ --default-time Time Time to show if modtime is unknown for files and directories (default 2000-01-01T00:00:00Z)+ --fast-list Use recursive list if available; uses more memory but fewer transactions+```++## See Also++<!-- markdownlint-capture -->+<!-- markdownlint-disable ul-style line-length -->++* [rclone](/commands/rclone/) - Show help for rclone commands, flags and backends.+++<!-- markdownlint-restore -->diff --git a/docs/content/commands/rclone_ls.md b/docs/content/commands/rclone_ls.mdindex e463876b6..cdb42ac0e 100644--- a/docs/content/commands/rclone_ls.md+++ b/docs/content/commands/rclone_ls.md@@ -41,6 +41,10 @@ Note that `ls` and `lsl` recurse by default - use `--max-depth 1` to stop the reThe other list commands `lsd`,`lsf`,`lsjson` do not recurse by default -use `-R` to make them recurse.+On the bucket-based remotes (e.g. s3, gcs, b2) a directory is a key prefix+rather than an object, so a listing without `-R` shows each prefix as one+directory entry and not the objects stored under it.+List commands prefer a recursive method that uses more memory but fewertransactions by default. Use `--disable ListR` to suppress the behavior.See [`--fast-list`](/docs/#fast-list) for more details.diff --git a/docs/content/commands/rclone_lsd.md b/docs/content/commands/rclone_lsd.mdindex b736c08c3..e82dd621e 100644--- a/docs/content/commands/rclone_lsd.md+++ b/docs/content/commands/rclone_lsd.md@@ -13,7 +13,7 @@ Lists the directories in the source path to standard output. Does notrecurse by default. Use the `-R` flag to recurse.This command lists the total size of the directory (if known, -1 if-not), the modification time (if known, the current time if not), the+not), the modification time (if known, the `--default-time` if not), thenumber of objects in the directory (if known, -1 if not) and the nameof the directory, E.g.@@ -53,6 +53,10 @@ Note that `ls` and `lsl` recurse by default - use `--max-depth 1` to stop the reThe other list commands `lsd`,`lsf`,`lsjson` do not recurse by default -use `-R` to make them recurse.+On the bucket-based remotes (e.g. s3, gcs, b2) a directory is a key prefix+rather than an object, so a listing without `-R` shows each prefix as one+directory entry and not the objects stored under it.+List commands prefer a recursive method that uses more memory but fewertransactions by default. Use `--disable ListR` to suppress the behavior.See [`--fast-list`](/docs/#fast-list) for more details.diff --git a/docs/content/commands/rclone_lsf.md b/docs/content/commands/rclone_lsf.mdindex 6c53e66bd..cfd3aab88 100644--- a/docs/content/commands/rclone_lsf.md+++ b/docs/content/commands/rclone_lsf.md@@ -164,6 +164,10 @@ Note that `ls` and `lsl` recurse by default - use `--max-depth 1` to stop the reThe other list commands `lsd`,`lsf`,`lsjson` do not recurse by default -use `-R` to make them recurse.+On the bucket-based remotes (e.g. s3, gcs, b2) a directory is a key prefix+rather than an object, so a listing without `-R` shows each prefix as one+directory entry and not the objects stored under it.+List commands prefer a recursive method that uses more memory but fewertransactions by default. Use `--disable ListR` to suppress the behavior.See [`--fast-list`](/docs/#fast-list) for more details.diff --git a/docs/content/commands/rclone_lsjson.md b/docs/content/commands/rclone_lsjson.mdindex dd50cbb48..017c77f76 100644--- a/docs/content/commands/rclone_lsjson.md+++ b/docs/content/commands/rclone_lsjson.md@@ -111,6 +111,10 @@ Note that `ls` and `lsl` recurse by default - use `--max-depth 1` to stop the reThe other list commands `lsd`,`lsf`,`lsjson` do not recurse by default -use `-R` to make them recurse.+On the bucket-based remotes (e.g. s3, gcs, b2) a directory is a key prefix+rather than an object, so a listing without `-R` shows each prefix as one+directory entry and not the objects stored under it.+List commands prefer a recursive method that uses more memory but fewertransactions by default. Use `--disable ListR` to suppress the behavior.See [`--fast-list`](/docs/#fast-list) for more details.diff --git a/docs/content/commands/rclone_lsl.md b/docs/content/commands/rclone_lsl.mdindex 9e3b6acc3..5fdf194bc 100644--- a/docs/content/commands/rclone_lsl.md+++ b/docs/content/commands/rclone_lsl.md@@ -42,6 +42,10 @@ Note that `ls` and `lsl` recurse by default - use `--max-depth 1` to stop the reThe other list commands `lsd`,`lsf`,`lsjson` do not recurse by default -use `-R` to make them recurse.+On the bucket-based remotes (e.g. s3, gcs, b2) a directory is a key prefix+rather than an object, so a listing without `-R` shows each prefix as one+directory entry and not the objects stored under it.+List commands prefer a recursive method that uses more memory but fewertransactions by default. Use `--disable ListR` to suppress the behavior.See [`--fast-list`](/docs/#fast-list) for more details.diff --git a/docs/content/commands/rclone_mount.md b/docs/content/commands/rclone_mount.mdindex 89fff842a..812f8a72a 100644--- a/docs/content/commands/rclone_mount.md+++ b/docs/content/commands/rclone_mount.md@@ -292,6 +292,11 @@ for all `mount` and `serve` commands on macOS. For details, see [vfs-case-sensit### NFS mount+For macOS (and other platforms where this path is supported), prefer the dedicated+[rclone nfsmount](/commands/rclone_nfsmount/) command. It starts the NFS server and+performs the mount for you. `rclone mount` itself still uses FUSE (macFUSE/FUSE-T)+and does not switch to NFS via a flag.+This method spins up an NFS server using [serve nfs](/commands/rclone_serve_nfs/)command and mounts it to the specified mountpoint. If you run this in backgroundmode using |--daemon|, you will need to send SIGTERM signal to the rclone process@@ -629,7 +634,7 @@ at all times. The buffered data is bound to one open file and won't beshared.This flag is a upper limit for the used memory per open file. The-buffer will only use memory for data that is downloaded but not not+buffer will only use memory for data that is downloaded but notyet read. If the buffer is empty, only a small amount of memory willbe used.@@ -684,7 +689,8 @@ longest. This cache flushing strategy is efficient and more relevantfiles are likely to remain cached.The `--vfs-cache-max-age` will evict files from the cache-after the set time since last access has passed. The default value of+after the set time since last access has passed; it is based on access time,+not on when the file was first added to the cache. The default value of1 hour will start evicting files from cache that haven't been accessedfor 1 hour. When a cached file is accessed the 1 hour timer is reset to 0and will wait for 1 more hour before evicting. Specify the time with@@ -838,7 +844,7 @@ Rclone reads `--vfs-read-chunk-streams` chunks of size`--vfs-read-chunk-size` concurrently. The size for each read will stayconstant.-This improves performance performance massively on high latency links+This improves performance massively on high latency linksor very high bandwidth links to high performance object stores.Some experimentation will be needed to find the optimum values ofdiff --git a/docs/content/commands/rclone_nfsmount.md b/docs/content/commands/rclone_nfsmount.mdindex f298ed210..925c298cd 100644--- a/docs/content/commands/rclone_nfsmount.md+++ b/docs/content/commands/rclone_nfsmount.md@@ -293,6 +293,11 @@ for all `mount` and `serve` commands on macOS. For details, see [vfs-case-sensit### NFS mount+For macOS (and other platforms where this path is supported), prefer the dedicated+[rclone nfsmount](/commands/rclone_nfsmount/) command. It starts the NFS server and+performs the mount for you. `rclone mount` itself still uses FUSE (macFUSE/FUSE-T)+and does not switch to NFS via a flag.+This method spins up an NFS server using [serve nfs](/commands/rclone_serve_nfs/)command and mounts it to the specified mountpoint. If you run this in backgroundmode using |--daemon|, you will need to send SIGTERM signal to the rclone process@@ -630,7 +635,7 @@ at all times. The buffered data is bound to one open file and won't beshared.This flag is a upper limit for the used memory per open file. The-buffer will only use memory for data that is downloaded but not not+buffer will only use memory for data that is downloaded but notyet read. If the buffer is empty, only a small amount of memory willbe used.@@ -685,7 +690,8 @@ longest. This cache flushing strategy is efficient and more relevantfiles are likely to remain cached.The `--vfs-cache-max-age` will evict files from the cache-after the set time since last access has passed. The default value of+after the set time since last access has passed; it is based on access time,+not on when the file was first added to the cache. The default value of1 hour will start evicting files from cache that haven't been accessedfor 1 hour. When a cached file is accessed the 1 hour timer is reset to 0and will wait for 1 more hour before evicting. Specify the time with@@ -839,7 +845,7 @@ Rclone reads `--vfs-read-chunk-streams` chunks of size`--vfs-read-chunk-size` concurrently. The size for each read will stayconstant.-This improves performance performance massively on high latency links+This improves performance massively on high latency linksor very high bandwidth links to high performance object stores.Some experimentation will be needed to find the optimum values ofdiff --git a/docs/content/commands/rclone_rcd.md b/docs/content/commands/rclone_rcd.mdindex 2a87513f8..3aa7538c4 100644--- a/docs/content/commands/rclone_rcd.md+++ b/docs/content/commands/rclone_rcd.md@@ -47,6 +47,13 @@ accept in the HTTP header.will overriding existing values. The flag may be repeated to add multipleheaders. Use the format `Header-Name: value`.+`--rc-allow-origin` enables cross-origin (CORS) requests from browsers+by setting the `Access-Control-Allow-Origin` header on every response to the+origin given, e.g. `--rc-allow-origin https://app.example.com`. Browser+CORS preflight OPTIONS requests are answered by rclone without needing+authentication as browsers never send credentials with them. Setting this+to `*` allows any origin which is a security risk on an authenticated server.+`--rc-baseurl` controls the URL prefix that rclone serves from. By defaultrclone will serve from the root. If you used `--rc-baseurl "/rclone"` thenrclone would serve from a URL starting with "/rclone/". This is@@ -116,6 +123,13 @@ to be used within the template to server pages:| .Sort | | The current sort used. This is changeable via '?sort=' parameter. Possible values: namedirfirst, name, size, time (default namedirfirst). || .Order | | The current ordering used. This is changeable via '?order=' parameter. Possible values: asc, desc (default asc). || .Query | | Currently unused. |+| .Path | | The path of the directory from the root with leading and trailing slashes, e.g. '/dir/'. |+| .IsRoot | | Boolean for if this is the root directory. |+| .UpLink | | The link to the parent directory. |+| .Static | | Boolean for if the listing is being written as a static page by rclone index. |+| .NumDirs | | The number of directories in the listing. |+| .NumFiles | | The number of files in the listing. |+| .TotalSize | | The total size in bytes of the files in the listing. || .Breadcrumb | | Allows for creating a relative navigation. || | .Link | The link of the Text relative to the root. || | .Text | The Name of the directory. |@@ -125,6 +139,7 @@ to be used within the template to server pages:| | .IsDir | Boolean for if an entry is a directory or not. || | .Size | Size in bytes of the entry. || | .ModTime | The UTC timestamp of an entry. |+| | .MimeType | The MIME type of an entry guessed from its name, or 'inode/directory'. |The server also makes the following functions available so that they can be usedwithin the template. These functions help extend the options for dynamic@@ -165,8 +180,20 @@ htpasswd -B htpasswd userhtpasswd -B htpasswd anotherUser```+Use the same hash format and cost for every entry in the file. If the+entries differ, the time taken to reject a wrong password can reveal+which usernames exist.+The password file can be updated while rclone is running.+When authentication is enabled every request must be authenticated,+including HTTP OPTIONS requests, as the answer to an OPTIONS request can+reveal whether a path exists. Health checks or monitoring which use+unauthenticated OPTIONS requests will need to send credentials. The+exception is browser CORS preflight requests when `--rc-allow-origin`+is set, which are answered without authentication as browsers never send+credentials with them.+Use `--rc-realm` to set the authentication realm.Use `--rc-salt` to change the password hashing salt from the default.@@ -215,11 +242,6 @@ Flags to control the Remote Control API--rc-template string User-specified template--rc-user string User name for authentication--rc-user-from-header string User name from a defined HTTP header- --rc-web-fetch-url string URL to fetch the releases for webgui (default "https://api.github.com/repos/rclone/rclone-webui-react/releases/latest")- --rc-web-gui Launch WebGUI on localhost- --rc-web-gui-force-update Force update to latest version of web gui- --rc-web-gui-no-open-browser Don't open the browser automatically- --rc-web-gui-update Check and update to latest version of web gui```## See Alsodiff --git a/docs/content/commands/rclone_serve_dlna.md b/docs/content/commands/rclone_serve_dlna.mdindex c93394468..7dc17799a 100644--- a/docs/content/commands/rclone_serve_dlna.md+++ b/docs/content/commands/rclone_serve_dlna.md@@ -100,7 +100,7 @@ at all times. The buffered data is bound to one open file and won't beshared.This flag is a upper limit for the used memory per open file. The-buffer will only use memory for data that is downloaded but not not+buffer will only use memory for data that is downloaded but notyet read. If the buffer is empty, only a small amount of memory willbe used.@@ -155,7 +155,8 @@ longest. This cache flushing strategy is efficient and more relevantfiles are likely to remain cached.The `--vfs-cache-max-age` will evict files from the cache-after the set time since last access has passed. The default value of+after the set time since last access has passed; it is based on access time,+not on when the file was first added to the cache. The default value of1 hour will start evicting files from cache that haven't been accessedfor 1 hour. When a cached file is accessed the 1 hour timer is reset to 0and will wait for 1 more hour before evicting. Specify the time with@@ -309,7 +310,7 @@ Rclone reads `--vfs-read-chunk-streams` chunks of size`--vfs-read-chunk-size` concurrently. The size for each read will stayconstant.-This improves performance performance massively on high latency links+This improves performance massively on high latency linksor very high bandwidth links to high performance object stores.Some experimentation will be needed to find the optimum values ofdiff --git a/docs/content/commands/rclone_serve_docker.md b/docs/content/commands/rclone_serve_docker.mdindex c0674d67e..8464081c7 100644--- a/docs/content/commands/rclone_serve_docker.md+++ b/docs/content/commands/rclone_serve_docker.md@@ -162,7 +162,7 @@ at all times. The buffered data is bound to one open file and won't beshared.This flag is a upper limit for the used memory per open file. The-buffer will only use memory for data that is downloaded but not not+buffer will only use memory for data that is downloaded but notyet read. If the buffer is empty, only a small amount of memory willbe used.@@ -217,7 +217,8 @@ longest. This cache flushing strategy is efficient and more relevantfiles are likely to remain cached.The `--vfs-cache-max-age` will evict files from the cache-after the set time since last access has passed. The default value of+after the set time since last access has passed; it is based on access time,+not on when the file was first added to the cache. The default value of1 hour will start evicting files from cache that haven't been accessedfor 1 hour. When a cached file is accessed the 1 hour timer is reset to 0and will wait for 1 more hour before evicting. Specify the time with@@ -371,7 +372,7 @@ Rclone reads `--vfs-read-chunk-streams` chunks of size`--vfs-read-chunk-size` concurrently. The size for each read will stayconstant.-This improves performance performance massively on high latency links+This improves performance massively on high latency linksor very high bandwidth links to high performance object stores.Some experimentation will be needed to find the optimum values ofdiff --git a/docs/content/commands/rclone_serve_ftp.md b/docs/content/commands/rclone_serve_ftp.mdindex c8bb7c7b5..b8e2ebd36 100644--- a/docs/content/commands/rclone_serve_ftp.md+++ b/docs/content/commands/rclone_serve_ftp.md@@ -93,7 +93,7 @@ at all times. The buffered data is bound to one open file and won't beshared.This flag is a upper limit for the used memory per open file. The-buffer will only use memory for data that is downloaded but not not+buffer will only use memory for data that is downloaded but notyet read. If the buffer is empty, only a small amount of memory willbe used.@@ -148,7 +148,8 @@ longest. This cache flushing strategy is efficient and more relevantfiles are likely to remain cached.The `--vfs-cache-max-age` will evict files from the cache-after the set time since last access has passed. The default value of+after the set time since last access has passed; it is based on access time,+not on when the file was first added to the cache. The default value of1 hour will start evicting files from cache that haven't been accessedfor 1 hour. When a cached file is accessed the 1 hour timer is reset to 0and will wait for 1 more hour before evicting. Specify the time with@@ -302,7 +303,7 @@ Rclone reads `--vfs-read-chunk-streams` chunks of size`--vfs-read-chunk-size` concurrently. The size for each read will stayconstant.-This improves performance performance massively on high latency links+This improves performance massively on high latency linksor very high bandwidth links to high performance object stores.Some experimentation will be needed to find the optimum values of@@ -545,9 +546,10 @@ This config generated must have this extra parameter- `_root` - root to use for the backend-And it may have this parameter+And it may have these parameters- `_obscure` - comma separated strings for parameters to obscure+- `_secret_access_key` - the secret for S3 access key auth (see below)If password authentication was used by the client, input to the proxyprocess (on STDIN) would look similar to this:@@ -555,7 +557,8 @@ process (on STDIN) would look similar to this:```json{"user": "me",- "pass": "mypassword"+ "pass": "mypassword",+ "client_ip": "192.168.1.1"}```@@ -565,10 +568,44 @@ proxy process (on STDIN) would look similar to this:```json{"user": "me",- "public_key": "AAAAB3NzaC1yc2EAAAADAQABAAABAQDuwESFdAe14hVS6omeyX7edc...JQdf"+ "public_key": "AAAAB3NzaC1yc2EAAAADAQABAAABAQDuwESFdAe14hVS6omeyX7edc...JQdf",+ "client_ip": "192.168.1.1"+}+```++If the client authenticated with an S3 access key (`rclone serve s3`),+the client never sends its secret, only a signature made with it, so+the input contains just the access key ID as the `user` with no `pass`+or `public_key`:++```json+{+ "user": "AKIAIOSFODNN7EXAMPLE",+ "client_ip": "192.168.1.1"}```+In this case the program must look up the secret access key for that+access key ID and return it in the `_secret_access_key` field of the+output. Rclone then uses that secret to verify the signature on the+request, refusing the request if it does not match. This means the+proxy program is the source of truth for both the credentials and the+backend they map to. If the program does not return+`_secret_access_key` or returns it empty the request is refused.++The program's answer for an access key ID is cached (see below) but+is checked with the program again after 5 minutes even if the access+key ID is in constant use, so revoking an access key ID in the+program takes effect within 5 minutes. A rotated secret takes effect+on the first request signed with it.++The `client_ip` key holds the IP address the client connected from,+without a port number. It can be used to restrict logins to certain+networks, or to log authentication attempts centrally. It is omitted if+the client has no IP address, for example when connecting over a unix+socket. Note that if rclone is behind a reverse proxy this will be the+address of the reverse proxy and not the original client.+And as an example return this on STDOUT```json@@ -594,11 +631,20 @@ to make proxy to many different sftp backends, you could make thein the output and the user to `user`. For security you'd probably wantto restrict the `host` to a limited list.-An internal cache of backends is keyed on the `user` and a hash of the-`pass` or `public_key`. This means that if a user's password or-public-key changes, or the proxy returns different config parameters-(eg a rotated `api_key`), a fresh backend will be created on the next-request rather than the cached one being reused.+An internal cache of backends is keyed on the `user`, the kind of+credential presented, a hash of the `pass` or `public_key`, and the+`client_ip`. This means that if a user's password or public-key+changes, the client connects from a new IP address, or the proxy+returns different config parameters (eg a rotated `api_key`), a fresh+backend will be created on the next request rather than the cached one+being reused. A backend created for a public key login is never+reused for a password login, so a public key cannot be sent as a+password to skip the program.++A cached answer is checked with the program again after 5 minutes even+if the user is in constant use, so revoking a user in the program+takes effect within 5 minutes. For `serve sftp` this applies to new+logins - an existing connection is not closed.This can be used to build general purpose proxies to any kind ofbackend that rclone supports.diff --git a/docs/content/commands/rclone_serve_http.md b/docs/content/commands/rclone_serve_http.mdindex 1ea78cae8..4837b37a3 100644--- a/docs/content/commands/rclone_serve_http.md+++ b/docs/content/commands/rclone_serve_http.md@@ -49,6 +49,13 @@ accept in the HTTP header.will overriding existing values. The flag may be repeated to add multipleheaders. Use the format `Header-Name: value`.+`--allow-origin` enables cross-origin (CORS) requests from browsers+by setting the `Access-Control-Allow-Origin` header on every response to the+origin given, e.g. `--allow-origin https://app.example.com`. Browser+CORS preflight OPTIONS requests are answered by rclone without needing+authentication as browsers never send credentials with them. Setting this+to `*` allows any origin which is a security risk on an authenticated server.+`--baseurl` controls the URL prefix that rclone serves from. By defaultrclone will serve from the root. If you used `--baseurl "/rclone"` thenrclone would serve from a URL starting with "/rclone/". This is@@ -118,6 +125,13 @@ to be used within the template to server pages:| .Sort | | The current sort used. This is changeable via '?sort=' parameter. Possible values: namedirfirst, name, size, time (default namedirfirst). || .Order | | The current ordering used. This is changeable via '?order=' parameter. Possible values: asc, desc (default asc). || .Query | | Currently unused. |+| .Path | | The path of the directory from the root with leading and trailing slashes, e.g. '/dir/'. |+| .IsRoot | | Boolean for if this is the root directory. |+| .UpLink | | The link to the parent directory. |+| .Static | | Boolean for if the listing is being written as a static page by rclone index. |+| .NumDirs | | The number of directories in the listing. |+| .NumFiles | | The number of files in the listing. |+| .TotalSize | | The total size in bytes of the files in the listing. || .Breadcrumb | | Allows for creating a relative navigation. || | .Link | The link of the Text relative to the root. || | .Text | The Name of the directory. |@@ -127,6 +141,7 @@ to be used within the template to server pages:| | .IsDir | Boolean for if an entry is a directory or not. || | .Size | Size in bytes of the entry. || | .ModTime | The UTC timestamp of an entry. |+| | .MimeType | The MIME type of an entry guessed from its name, or 'inode/directory'. |The server also makes the following functions available so that they can be usedwithin the template. These functions help extend the options for dynamic@@ -167,8 +182,20 @@ htpasswd -B htpasswd userhtpasswd -B htpasswd anotherUser```+Use the same hash format and cost for every entry in the file. If the+entries differ, the time taken to reject a wrong password can reveal+which usernames exist.+The password file can be updated while rclone is running.+When authentication is enabled every request must be authenticated,+including HTTP OPTIONS requests, as the answer to an OPTIONS request can+reveal whether a path exists. Health checks or monitoring which use+unauthenticated OPTIONS requests will need to send credentials. The+exception is browser CORS preflight requests when `--allow-origin`+is set, which are answered without authentication as browsers never send+credentials with them.+Use `--realm` to set the authentication realm.Use `--salt` to change the password hashing salt from the default.@@ -236,7 +263,7 @@ at all times. The buffered data is bound to one open file and won't beshared.This flag is a upper limit for the used memory per open file. The-buffer will only use memory for data that is downloaded but not not+buffer will only use memory for data that is downloaded but notyet read. If the buffer is empty, only a small amount of memory willbe used.@@ -291,7 +318,8 @@ longest. This cache flushing strategy is efficient and more relevantfiles are likely to remain cached.The `--vfs-cache-max-age` will evict files from the cache-after the set time since last access has passed. The default value of+after the set time since last access has passed; it is based on access time,+not on when the file was first added to the cache. The default value of1 hour will start evicting files from cache that haven't been accessedfor 1 hour. When a cached file is accessed the 1 hour timer is reset to 0and will wait for 1 more hour before evicting. Specify the time with@@ -445,7 +473,7 @@ Rclone reads `--vfs-read-chunk-streams` chunks of size`--vfs-read-chunk-size` concurrently. The size for each read will stayconstant.-This improves performance performance massively on high latency links+This improves performance massively on high latency linksor very high bandwidth links to high performance object stores.Some experimentation will be needed to find the optimum values of@@ -688,9 +716,10 @@ This config generated must have this extra parameter- `_root` - root to use for the backend-And it may have this parameter+And it may have these parameters- `_obscure` - comma separated strings for parameters to obscure+- `_secret_access_key` - the secret for S3 access key auth (see below)If password authentication was used by the client, input to the proxyprocess (on STDIN) would look similar to this:@@ -698,7 +727,8 @@ process (on STDIN) would look similar to this:```json{"user": "me",- "pass": "mypassword"+ "pass": "mypassword",+ "client_ip": "192.168.1.1"}```@@ -708,10 +738,44 @@ proxy process (on STDIN) would look similar to this:```json{"user": "me",- "public_key": "AAAAB3NzaC1yc2EAAAADAQABAAABAQDuwESFdAe14hVS6omeyX7edc...JQdf"+ "public_key": "AAAAB3NzaC1yc2EAAAADAQABAAABAQDuwESFdAe14hVS6omeyX7edc...JQdf",+ "client_ip": "192.168.1.1"}```+If the client authenticated with an S3 access key (`rclone serve s3`),+the client never sends its secret, only a signature made with it, so+the input contains just the access key ID as the `user` with no `pass`+or `public_key`:++```json+{+ "user": "AKIAIOSFODNN7EXAMPLE",+ "client_ip": "192.168.1.1"+}+```++In this case the program must look up the secret access key for that+access key ID and return it in the `_secret_access_key` field of the+output. Rclone then uses that secret to verify the signature on the+request, refusing the request if it does not match. This means the+proxy program is the source of truth for both the credentials and the+backend they map to. If the program does not return+`_secret_access_key` or returns it empty the request is refused.++The program's answer for an access key ID is cached (see below) but+is checked with the program again after 5 minutes even if the access+key ID is in constant use, so revoking an access key ID in the+program takes effect within 5 minutes. A rotated secret takes effect+on the first request signed with it.++The `client_ip` key holds the IP address the client connected from,+without a port number. It can be used to restrict logins to certain+networks, or to log authentication attempts centrally. It is omitted if+the client has no IP address, for example when connecting over a unix+socket. Note that if rclone is behind a reverse proxy this will be the+address of the reverse proxy and not the original client.+And as an example return this on STDOUT```json@@ -737,11 +801,20 @@ to make proxy to many different sftp backends, you could make thein the output and the user to `user`. For security you'd probably wantto restrict the `host` to a limited list.-An internal cache of backends is keyed on the `user` and a hash of the-`pass` or `public_key`. This means that if a user's password or-public-key changes, or the proxy returns different config parameters-(eg a rotated `api_key`), a fresh backend will be created on the next-request rather than the cached one being reused.+An internal cache of backends is keyed on the `user`, the kind of+credential presented, a hash of the `pass` or `public_key`, and the+`client_ip`. This means that if a user's password or public-key+changes, the client connects from a new IP address, or the proxy+returns different config parameters (eg a rotated `api_key`), a fresh+backend will be created on the next request rather than the cached one+being reused. A backend created for a public key login is never+reused for a password login, so a public key cannot be sent as a+password to skip the program.++A cached answer is checked with the program again after 5 minutes even+if the user is in constant use, so revoking a user in the program+takes effect within 5 minutes. For `serve sftp` this applies to new+logins - an existing connection is not closed.This can be used to build general purpose proxies to any kind ofbackend that rclone supports.diff --git a/docs/content/commands/rclone_serve_nfs.md b/docs/content/commands/rclone_serve_nfs.mdindex 52a94b3e2..88ca29966 100644--- a/docs/content/commands/rclone_serve_nfs.md+++ b/docs/content/commands/rclone_serve_nfs.md@@ -167,7 +167,7 @@ at all times. The buffered data is bound to one open file and won't beshared.This flag is a upper limit for the used memory per open file. The-buffer will only use memory for data that is downloaded but not not+buffer will only use memory for data that is downloaded but notyet read. If the buffer is empty, only a small amount of memory willbe used.@@ -222,7 +222,8 @@ longest. This cache flushing strategy is efficient and more relevantfiles are likely to remain cached.The `--vfs-cache-max-age` will evict files from the cache-after the set time since last access has passed. The default value of+after the set time since last access has passed; it is based on access time,+not on when the file was first added to the cache. The default value of1 hour will start evicting files from cache that haven't been accessedfor 1 hour. When a cached file is accessed the 1 hour timer is reset to 0and will wait for 1 more hour before evicting. Specify the time with@@ -376,7 +377,7 @@ Rclone reads `--vfs-read-chunk-streams` chunks of size`--vfs-read-chunk-size` concurrently. The size for each read will stayconstant.-This improves performance performance massively on high latency links+This improves performance massively on high latency linksor very high bandwidth links to high performance object stores.Some experimentation will be needed to find the optimum values ofdiff --git a/docs/content/commands/rclone_serve_restic.md b/docs/content/commands/rclone_serve_restic.mdindex 5a4a9e6d2..7202146d8 100644--- a/docs/content/commands/rclone_serve_restic.md+++ b/docs/content/commands/rclone_serve_restic.md@@ -126,6 +126,13 @@ accept in the HTTP header.will overriding existing values. The flag may be repeated to add multipleheaders. Use the format `Header-Name: value`.+`--allow-origin` enables cross-origin (CORS) requests from browsers+by setting the `Access-Control-Allow-Origin` header on every response to the+origin given, e.g. `--allow-origin https://app.example.com`. Browser+CORS preflight OPTIONS requests are answered by rclone without needing+authentication as browsers never send credentials with them. Setting this+to `*` allows any origin which is a security risk on an authenticated server.+`--baseurl` controls the URL prefix that rclone serves from. By defaultrclone will serve from the root. If you used `--baseurl "/rclone"` thenrclone would serve from a URL starting with "/rclone/". This is@@ -210,8 +217,20 @@ htpasswd -B htpasswd userhtpasswd -B htpasswd anotherUser```+Use the same hash format and cost for every entry in the file. If the+entries differ, the time taken to reject a wrong password can reveal+which usernames exist.+The password file can be updated while rclone is running.+When authentication is enabled every request must be authenticated,+including HTTP OPTIONS requests, as the answer to an OPTIONS request can+reveal whether a path exists. Health checks or monitoring which use+unauthenticated OPTIONS requests will need to send credentials. The+exception is browser CORS preflight requests when `--allow-origin`+is set, which are answered without authentication as browsers never send+credentials with them.+Use `--realm` to set the authentication realm.Use `--salt` to change the password hashing salt from the default.diff --git a/docs/content/commands/rclone_serve_s3.md b/docs/content/commands/rclone_serve_s3.mdindex c831576eb..d59bf8b63 100644--- a/docs/content/commands/rclone_serve_s3.md+++ b/docs/content/commands/rclone_serve_s3.md@@ -26,6 +26,21 @@ docs](https://docs.aws.amazon.com/general/latest/gr/signature-version-4.html)).`--auth-key` is not provided then `serve s3` will allow anonymousaccess.+Alternatively `--auth-proxy` can be used to look up the secret for each+access key ID and choose the backend it maps to (see [Auth+Proxy](#auth-proxy) below). When an auth proxy is in use `--auth-key`+is ignored and every request must be signed with the secret the proxy+returns for its access key ID. Each access key ID is a separate user,+whose multipart uploads and object metadata are private to it, even if+the proxy maps several access key IDs to the same backend.++As on AWS, every `x-amz-*` header in an authenticated request must be+in its signed headers list; a request with an unsigned `x-amz-*`+header is refused with `AccessDenied`. S3 clients sign these headers+as a matter of course, but a header added after signing (for example+by a proxy in front of the server, or `x-amz-meta-*` headers sent+with a presigned URL which only signs the host) is refused.+Like all rclone flags `--auth-key` can be set via environmentvariables, in this case `RCLONE_AUTH_KEY`. Since this flag can berepeated, the input to `RCLONE_AUTH_KEY` is CSV encoded. Because the@@ -60,6 +75,14 @@ part of the hostname (such as mybucket.local)Use `--etag-hash` if you want to change the hash uses for the `ETag`.Note that using anything other than `MD5` (the default) is likely tocause problems for S3 clients which rely on the Etag being the MD5.+Use `--etag-hash auto` to use the best hash the backend supports - with+`--auth-proxy` that of each user's backend.++The `ETag` of an object made by a multipart upload is its hash, like+any other object, rather than the `-N` suffixed `ETag` AWS S3 gives+these. If the remote doesn't have the MD5 of objects, `HeadObject`+and listings only return the MD5 of objects uploaded through `serve+s3` while the object's metadata is kept (see below).## Quickstart@@ -101,29 +124,78 @@ access_key_id = ACCESS_KEY_IDsecret_access_key = SECRET_ACCESS_KEY```-## Multipart uploads+## Object uploads (PUT)-By default `serve s3` **streams** each multipart upload, in part-number-order, into a single `PutStream` upload to the underlying remote, so the-whole file is never buffered in memory - memory use stays bounded by the-parts in flight. The remote then performs its own internal upload (for-example its own multipart upload, still with bounded memory). This works-for any remote that supports `PutStream`, which is nearly all of them,-including through `crypt`.+A `PutObject` upload only ever changes the object at its key atomically, on+success, a failed or interrupted PUT neither removes nor overwrites the+object already stored at the key, and never leaves a partial object visible+at it.++Remotes that upload atomically (e.g. object stores such as `s3`) are streamed+straight to the destination. On remotes where a partial upload would+otherwise be visible (e.g. `local`), and whenever `--vfs-cache-mode` is+`writes` or above, the upload is written to a temporary object that is+renamed into place on success; these remotes need to support a server-side+move or copy for this (nearly all do - without move or copy the upload is+written directly and a failed PUT may leave a partial object at the key). If+`serve s3` is killed part-way through an upload the temporary object (named+with a leading `.rclone_temp_put_`) may be left behind; it is hidden from+S3 listings but must be removed manually.++## Multipart uploads-The upload is atomic so the destination object only ever changes on a+Multipart uploads are written, in part-number order, to a temporary+object which is renamed into place, server-side, on completion, so the+upload is atomic. The object at the key only ever changes on asuccessful completion. A failed or aborted upload never affects any-object already stored under that name. Remotes that upload atomically-already (object stores such as `s3`) are streamed straight to the-destination. On remotes where a partial upload would otherwise be visible-(such as `local`), the parts are streamed to a temporary object that is-moved into place, server-side, on completion; these remotes therefore-also need to support a server-side move or copy.+object already stored under that name and a partly-uploaded object+never becomes visible under it.++With the default `--vfs-cache-mode off` `serve s3` **streams** each+multipart upload, in part-number order, into a single streaming upload+to the underlying remote, so the whole file is never buffered in+memory: only parts which arrive out of order are buffered, within the+limits described below. The remote then performs its own internal upload (for example its own multipart+upload, still with bounded memory). Remotes that don't support+streaming uploads (those that must know the file size before the+upload starts, such as `onedrive`, `pcloud`, `jottacloud`, `mailru`,+`opendrive`, `putio`, `protondrive` and `zoho`) have the parts spooled+to a temporary file on **local disk** instead, and uploaded with the+size then known on completion, so they need local disk space for the+largest objects in flight rather than memory.++With `--vfs-cache-mode writes` (or `full`) the parts are written to a+temporary file in the VFS cache and uploaded by the VFS write-back -+see [Multipart uploads and the VFS+cache](#multipart-uploads-and-the-vfs-cache) below.++The rename into place needs the remote to support a server-side move+or copy, which nearly all do. It is a cheap rename on most remotes,+but on object stores without a real rename (such as `s3` itself) the+move is performed as a server-side copy and delete of the whole+object, which can take time and API calls for large objects.+Concurrent multipart uploads of the same key (which S3 permits) are+safe. Each writes its own temporary object and the last to complete+wins.++On the few remotes that support neither server side move nor copy, the+parts are written straight to the destination object instead and never+buffered in memory. This is at some cost in atomicity - the incomplete+object is visible under its final name while the upload is in flight,+as it also is for a plain object PUT on such remotes, and concurrent+multipart uploads of the same key write to the same object and can+interleave. A failed or aborted upload still leaves any pre-existing+object untouched provided the remote uploads atomically and the VFS+cache is off; on a remote where partial uploads are visible it may+leave partial data at the key (like a plain PUT there), and with+`--vfs-cache-mode writes` (or `full`) a write to the cache cannot be+abandoned, so an aborted upload's partial data is written back to the+remote as if it had completed.**Features**-- The whole object is never buffered in memory; memory use is bounded by- the parts in flight, not the upload size.+- The whole object is never buffered in memory; only parts arriving out+ of order are, within fixed limits, whatever the upload size.- Parts can be any size. Clients that don't produce uniform-sized partswork fine - for example PostgreSQL backup tools such as **pgBarman**and **pgBackRest**, which flush an upload buffer once it grows past@@ -133,69 +205,275 @@ also need to support a server-side move or copy.as one continuous stream.- The destination object only ever changes atomically, on completion: anaborted or failed upload leaves any pre-existing object of the same- name untouched, and a partly-uploaded object never becomes visible.-- Backend-agnostic - it only needs the remote to support `PutStream`- (plus a server-side move or copy on remotes that don't upload- atomically).+ name untouched, and a partly-uploaded object never becomes visible+ (except on the few remotes with no server-side move or copy, as+ above).+- Multipart uploads go through the VFS like any other upload, so they+ show in rclone's transfer stats and obey `--bwlimit`.+- Backend-agnostic - it only needs the remote to support a server-side+ move or copy for the rename into place, which nearly all do; a remote+ without streaming upload support spools to local disk as above.**Limitations**-- Parts must arrive in ascending, contiguous part-number order- (1, 2, 3, ...). Parts the client uploads concurrently or out of order- are buffered until their turn. The memory used for this buffering is- capped, per upload, by `--multipart-streaming-buffer-limit` (default- `256M`, `0` for no limit): a part that would take the buffer over the- limit is stalled until the stream drains, so a client that uploads- faster than the remote can accept sees backpressure rather than- unbounded server memory use. Since a stalled part holds its HTTP- request open, clients whose upload concurrency times chunk size- exceeds the limit may need a longer read timeout when the remote is- slow. Non-contiguous part numbers are rejected on completion.- Configure the client to upload in part order, ideally with low- concurrency, for the lowest memory use.+- Parts must arrive in ascending, contiguous part-number order (1, 2, 3,+ ...). The part the stream needs next is written straight through as it+ arrives, so it is never buffered, whatever its size. Parts the client+ uploads concurrently or out of order are buffered until their turn.+ The memory used for this buffering is capped, per upload, by+ `--multipart-streaming-buffer-limit` (default `256M`, `0` for no+ limit), and across all the uploads of each user (each access key) by+ `--multipart-streaming-buffer-total` (default `1G`, `0` for no+ limit), counting each buffered part in the+ whole 1 MiB memory pages it occupies: a part that would take the+ buffer over either limit is stalled until the stream drains (or, if it+ is bigger than the limits, until it is the next part), so a client+ that uploads faster than the remote can accept sees backpressure+ rather than unbounded server memory use. A stalled part holds its HTTP+ request open, and one stalled for more than a minute is failed with a+ `SlowDown` error, which S3 clients retry. Clients whose upload+ concurrency times part size exceeds the limits will see these stalls+ and retries when the remote is slower than the client. Non-contiguous+ part numbers are rejected on completion. Configure the client to+ upload in part order, ideally with low concurrency, for the lowest+ memory use.- A part uploaded again before completion - typically a client retryingafter a timeout - is accepted: if the earlier copy is still bufferedit is replaced, and if it has already been streamed an identical- re-upload is a no-op. What isn't possible is replacing a part that has- already been streamed with *different* content - that is rejected. A- failure in the stream to the remote itself still aborts the whole- upload and the client must start it again. (The remote's own upload+ re-upload is a no-op (a re-upload which arrives while the part is+ being streamed waits for it to finish first). If a part fails part way+ through being streamed+ (for example the client's connection drops) the client's retry of it+ carries on where it stopped, provided it has the same content. What+ isn't possible is replacing a part that has already been streamed,+ in whole or in part, with *different* content - that is rejected. A+ part which arrives in full but is then rejected, for example because+ it doesn't match the `Content-MD5` the client declared, has already+ been streamed, so the whole upload is failed and the client must+ start it again. So is a failure in the stream to the remote itself. (The remote's own uploadstill retries its internal chunks.)+- A `CompleteMultipartUpload` which fails because the parts it lists+ don't match those uploaded, or there is a gap in the part numbers,+ leaves the upload open, as on S3, so the client can upload any+ missing parts and complete it again. If writing the completed object+ to the remote fails, the upload is failed and must be started again.- Parts are serialised into one stream, so ingest from the client is- effectively single-threaded, although the remote's own upload still- runs concurrently.-- On remotes that don't upload atomically (such as `local`), the- completed object is moved into place with a server-side operation.- This is a cheap rename on most such remotes. On these remotes, if- `serve s3` is killed part-way through an upload the temporary object- (named with a leading `.rclone_multipart_upload_`) may be left behind;- it is hidden from S3 listings but must be removed manually.+ effectively single-threaded. When streaming, the remote's own upload+ runs concurrently with the parts arriving; with the local disk spool+ or the VFS cache the upload to the remote only starts on completion.+- If `serve s3` is killed part-way through an upload the temporary+ object (named with a leading `.rclone_temp_multipart_`) may be left+ behind; it is hidden from S3 listings but must be removed manually.++### Multipart uploads and the VFS cache++With `--vfs-cache-mode writes` (or `full`) multipart uploads do not+stream to the remote at all. The parts are written, in part-number+order, to a temporary file in the VFS cache. On completion the file is+renamed into place and uploaded by the VFS write-back, exactly like a+plain object PUT. This needs no streaming upload support from the+remote. The rename normally happens in the cache before the upload has+started, but the VFS requires the remote to support a server-side move+or copy to rename files at all (and uses one if the temporary file has+already been written back, e.g. with `--vfs-write-back 0`). On remotes+without either, the parts are written to the cache directly under the+final key instead: the upload still never touches memory, but it loses+its atomicity - the in-flight upload is visible at the key, and an+aborted upload cannot be abandoned once in the cache, so its partial+data is written back to the remote as if it were a completed object.++Remotes that benefit from `--vfs-cache-mode writes`:++- **Remotes over slow or unreliable links.** A failure in a streamed+ upload aborts the whole multipart upload and the client must start+ again from the first part; a failed write-back upload is retried by+ the VFS (see `--vfs-cache-max-age` and friends) without the client+ being involved. Ingest from the client also runs at local disk speed+ rather than being throttled to the remote's pace.+- **Workloads that read back or overwrite what they just wrote.** The+ completed object stays in the cache, so subsequent `GET`/`HEAD`+ requests are served locally, and plain PUTs and multipart uploads to+ the same key go through the same cache entry so the last write wins+ regardless of upload style.++The trade-offs of the VFS cache:++- The whole object lands on local disk, so the cache (`--cache-dir`)+ needs space for the largest objects in flight; `--vfs-cache-max-size`+ cannot evict files which are still being uploaded.+- The `200 OK` for `CompleteMultipartUpload` means the data is safely+ in the **local cache**, not yet on the remote - the same durability+ the cache gives plain PUTs. If an acknowledgement must mean the data+ has reached the remote (for example WAL archiving), use the default+ `--vfs-cache-mode off`.+- The upload to the remote only starts on completion, rather than+ overlapping with the parts arriving, so the data reaches the remote+ later than with streaming.+- If `serve s3` is killed part-way through an upload, the temporary+ file survives in the cache and the VFS cache recovery uploads it to+ the remote on restart as a temporary object (named with a leading+ `.rclone_temp_multipart_`); as with the streaming path, it is+ hidden from S3 listings but must be removed manually.++### Cleaning up temporary objects++If `serve s3` is killed part-way through an upload it can leave a+temporary object behind, named with a leading `.rclone_temp_`. This+whole prefix is reserved: any object whose name (the last+`/`-separated segment of its key) starts with `.rclone_temp_` is+hidden from S3 listings, so don't give real objects such names - an+existing object with such a name disappears from listings (though it+stays accessible directly by its key: only listings hide reserved+names, `GET`, `HEAD` and `DELETE` of the exact key still work). A+temporary object never holds acknowledged data - uploads whose+temporary object survived were never confirmed to the client - so old+ones are safe to delete:++ rclone delete --min-age 24h --include ".rclone_temp_*" remote:path++The `--min-age` protects uploads which are still in progress: make sure+it is longer than your longest upload, especially if several `serve s3`+instances share the same remote.++rclone v1.75 named its temporary multipart objects+`.rclone_multipart_upload_*`; leftovers from an older server are also+hidden from listings and can be cleaned up the same way.++### Abandoned uploads++A client which starts a multipart upload and vanishes without either+completing or aborting it would otherwise hold on to its resources+forever.++An incomplete multipart upload which has had no activity for+`--multipart-expiry` (default `24h`) is therefore aborted and cleaned+up, exactly as if the client had called `AbortMultipartUpload`, and a+`NOTICE` is logged.++An upload with a part still being received is never expired, however+slowly the part is arriving, and each completed part restarts the+clock, so the expiry only needs to outlast the client's pauses+*between* parts, not the whole upload.++Late operations on an expired upload fail with `NoSuchUpload`, as they+do on real S3 when a lifecycle rule has aborted the upload. Set+`--multipart-expiry 0` to keep incomplete uploads forever - but note+that abandoned uploads are then never cleaned up and go on counting+towards `--multipart-max-uploads` below, so enough of them will stop+the user starting any new multipart upload until the server is+restarted.++Each multipart upload in progress holds server resources - for example+the buffers the remote's own upload uses once the first part has+arrived - so each user (each access key) can have no more than+`--multipart-max-uploads` (default `1000`, `0` for no limit) in+progress at once. Starting another fails with a `SlowDown` error,+which S3 clients retry, and a one-off `NOTICE` is logged.### Disabling streaming-If you pass `--disable-multipart-streaming`, or the remote doesn't-support `PutStream` (or doesn't upload atomically and can't move or copy-server-side), multipart uploads are instead **buffered in memory**-by the underlying S3 library: every part is held in memory and the whole-object is written out in one go when the upload completes (the previous-behaviour). This removes the in-order/contiguous-part restriction above,-so parts can be uploaded in any order, but **memory use grows with the-size of the upload**, so it is only suitable for small objects. A one-off-`NOTICE` is logged the first time this happens.--Alternatively, if the client is an rclone `s3` remote (like the-`[serves3]` example above), you can set `use_multipart_uploads = false`-on it so it uploads each object as a single stream and skips multipart-uploads altogether.+If you pass `--disable-multipart-streaming`, multipart uploads are+instead **buffered in memory** by the underlying S3 library: every+part is held in memory and the whole object is written out in one go+when the upload completes. This removes the in-order/contiguous-part+restriction above, so parts can be uploaded in any order, but **memory+use grows with the size of the upload**, so it is only suitable for+small objects. A one-off `NOTICE` is logged the first time this+happens. This flag is the only thing that makes multipart uploads+buffer in memory - it is never done because of missing remote+capabilities. The limits on multipart uploads above+(`--multipart-streaming-buffer-limit`,+`--multipart-streaming-buffer-total` and `--multipart-max-uploads`) do+not apply to uploads buffered in memory, so don't use this flag if the+server is open to clients you don't trust. Consider+`--vfs-cache-mode writes` instead, which buffers the upload in the VFS+cache on disk and takes precedence over+`--disable-multipart-streaming`.++## Memory use++`serve s3` never buffers whole objects in memory (except with+`--disable-multipart-streaming`, above), and limits the memory each+kind of request can use, so that a client, even a hostile one, can't+make it use unbounded memory.++- **Object data.** `PutObject` bodies and multipart parts arriving in+ order are streamed straight through to the remote. Parts arriving out+ of order are buffered, up to `--multipart-streaming-buffer-limit`+ per upload and `--multipart-streaming-buffer-total` per user.+- **Uploads in progress.** Each user can have at most+ `--multipart-max-uploads` multipart uploads in progress. An upload+ holds little memory until its+ first part arrives. After that, with `--vfs-cache-mode off`, it holds+ a `--streaming-upload-cutoff` sized buffer plus whatever the remote's+ own upload buffers, which for many remotes is up to their chunk size+ times their upload concurrency (for example `--s3-chunk-size` times+ `--s3-upload-concurrency`). Plain `PutObject` uploads use the same+ while they are in progress.+- **Metadata.** The metadata of at most `--metadata-max-objects`+ objects of each user is kept, each limited to 2 KB.+- **Request bodies.** The XML bodies of requests such as+ `CompleteMultipartUpload` and `DeleteObjects` are limited to 10 MiB,+ and `DeleteObjects` to 1000 keys. A browser form (`POST`) upload+ holds at most 1 MiB in memory, spooling the rest to a temporary file,+ and is limited to 5 GB.+- **Listings.** Each page of a `ListObjects` listing reads only the+ directories needed to fill it, and returns at most 1000 keys.+ Directory listings are cached by the [VFS directory+ cache](#vfs-directory-cache).++The largest of these is usually the remote's upload buffers, which can+be up to `--multipart-max-uploads` times the buffers of one upload for+each user if clients start many uploads and send part of each. The+limits apply to each user separately, so that one user can't stop+others uploading, which means the memory use grows with the number of+users. To limit it, lower+`--multipart-max-uploads`, the remote's chunk size or upload+concurrency, or set `--max-buffer-memory` to limit the memory used by+the buffers of all uploads to remotes which use rclone's buffer pool+(including `s3`, `azureblob` and `b2`). Buffered multipart parts come+from the same pool, so if you set `--max-buffer-memory` make it+comfortably bigger than `--multipart-streaming-buffer-total` times the+number of users uploading at once, otherwise+buffered parts can use up all the memory the remote needs to upload the+parts which would free them.++The number of connections is not limited, so if the server is open to+clients you don't trust, use `--auth-key` or `--auth-proxy` and+consider putting it behind a reverse proxy which limits connections.++## Design goals++These goals explain many of the limitations below.++- **Objects are plain files on the remote.** A bucket is a directory+ in the root and a key is a path in it, so the remote stays usable by+ other tools, such as `rclone mount` or `rclone sync`. So keys must+ be paths (see [Object names](#object-names)).+- **No database.** `serve s3` stores nothing but the objects. What the+ remote can't hold, such as user metadata, tags, the MD5 `ETag` of an+ object on a remote without MD5 and the directory markers put, is+ kept in memory as a best effort, and lost on restart.+- **Works with any rclone remote.** What `serve s3` can do depends on+ what the remote supports, such as the hashes used for `ETag`s and+ modification times, rather than needing a particular backend.+- **Bounded memory.** Uploads, including multipart uploads, are+ streamed to the remote and the memory each request can use is+ limited (see [Memory use](#memory-use)). Nothing is written to local+ disk unless the VFS cache is used or the remote can't stream uploads+ (see [Multipart uploads](#multipart-uploads)).+- **Refuse rather than ignore.** Unsupported features are refused, or+ answered as S3 answers when the feature isn't configured, so a+ client is never told something was done which wasn't (see+ [Compatibility](#compatibility)).+- **A storage gateway, not a full S3.** `serve s3` implements the parts+ of the S3 API clients use to store and fetch objects. Versioning,+ ACLs, bucket policies, lifecycle rules and the like are out of scope.+- **Compatible with rclone's s3 backend.** This includes its directory+ markers, which the tests check.## Bugs-Multipart server side copies do not work (see-[#7454](https://github.com/rclone/rclone/issues/7454)). These take a-very long time and eventually fail. The default threshold for-multipart server side copies is 5G which is the maximum it can be, so-files above this side will fail to be server side copied.-For a current list of `serve s3` bugs see the [serves3](https://github.com/rclone/rclone/labels/serve%20s3) bug categoryon GitHub.@@ -204,57 +482,124 @@ on GitHub.`serve s3` will treat all directories in the root as buckets andignore all files in the root. You can use `CreateBucket` to create-folders under the root, but you can't create empty folders under other-folders not in the root.+folders under the root, and put a directory marker to create an empty+folder below them (see [Object names](#object-names)).When using `PutObject` or `DeleteObject`, rclone will automaticallycreate or clean up empty folders. If you don't want to clean up empty-folders automatically, use `--no-cleanup`.--When using `ListObjects`, rclone will use `/` when the delimiter is-empty. This reduces backend requests with no effect on most-operations, but if the delimiter is something other than `/` and-empty, rclone will do a full recursive search of the backend, which-can take some time.--Versioning is not currently supported.--Metadata will only be saved in memory other than the rclone `mtime`-metadata which will be set as the modification time of the file.+folders automatically, use `--directory-markers` (see [Object+names](#object-names)). The deprecated `--no-cleanup` does the same.++When using `ListObjects` with the delimiter `/`, rclone lists the+directory the prefix is in, showing its subdirectories as common+prefixes, including empty ones. Without a delimiter the listing is+recursive: rclone walks the directory tree below the prefix, reading+only as many directories as it needs to fill each page, and lists+empty directories as their directory markers (see below). Any other+delimiter is supported too, but needs the whole directory tree below+the prefix walking, unless the delimiter is in the directory names.++Versioning is not supported (see [Design goals](#design-goals)).++Conditional writes, `PutObject` and `CompleteMultipartUpload` with+`If-None-Match` or `If-Match`, are supported. The conditions are+checked before the object is written, and again as it is moved into+place when it is written to a temporary object first. This happens for+multipart uploads and, for `PutObject`, with remotes which can upload+partial files (such as local disk) or with `--vfs-cache-mode minimal`+or above. The second check is atomic against other writes through+`serve s3`, but neither is atomic against changes made to the remote+some other way. If the remote doesn't support the `--etag-hash`,+objects have no `ETag` so `If-Match` with an `ETag` always fails.++Metadata will only be saved in memory (see [Design+goals](#design-goals)) other than the rclone `mtime` metadata which+will be set as the modification time of the file. The+metadata of at most `--metadata-max-objects` objects (default `100000`,+`0` for no limit) of each user (each access key) is kept; after that+the metadata of the user's least recently used object is forgotten. The metadata of an object is+ignored once the object has been changed some other way, for example+by another `--auth-proxy` user or directly on the backend. Object tags+are kept with the metadata in the same way.## Object names`serve s3` stores objects as files in the backend, so object keys aremapped to file paths rather than treated as the opaque strings AWS S3allows. Keys must be in canonical path form: keys that contain `..` or-`.` path segments, repeated slashes (`//`), or a leading or trailing-slash are rejected with a `400 Bad Request` (`InvalidArgument`)-instead of being normalised, since normalising them could alias two-distinct keys to the same file or resolve a key outside its bucket.-This matches the behaviour of other S3 servers such as MinIO.--## Supported operations--`serve s3` currently supports the following operations.--- Bucket- - `ListBuckets`- - `CreateBucket`- - `DeleteBucket`-- Object- - `HeadObject`- - `ListObjects`- - `GetObject`- - `PutObject`- - `DeleteObject`- - `DeleteObjects`- - `CreateMultipartUpload`- - `CompleteMultipartUpload`- - `AbortMultipartUpload`- - `CopyObject`- - `UploadPart`--Other operations will return error `Unimplemented`.+`.` path segments, repeated slashes (`//`), or a leading slash are+rejected with a `400 Bad Request` (`InvalidArgument`) instead of being+normalised, since normalising them could alias two distinct keys to+the same file or resolve a key outside its bucket. This matches the+behaviour of other S3 servers such as MinIO.++A key ending in `/`, such as `dir/`, is a directory marker, standing+for the directory `dir`. Putting an empty directory marker creates the+directory (one with data is rejected with `400 Bad Request`), and+deleting it removes the directory if it is empty. Directory markers+are found by `HeadObject`, `GetObject` and `CopyObject` for any+directory, but only the markers of empty directories, and of the+directories a marker was put for, are listed as objects, as other+directories may not have been made by putting a marker. The markers+put are remembered in memory with the metadata (see+[Limitations](#limitations)), so after a restart, or once forgotten,+the marker of a directory holding other keys is no longer listed.+While a marker is remembered its directory isn't removed when the+last key below it is deleted.++With `--directory-markers` every directory is listed as a directory+marker, so the markers are listed after a restart too, as needed by+clients which rely on them, such as rclone's s3 backend with+`--s3-directory-markers`. As every directory has a marker, a directory+isn't removed when the last key below it is deleted. Deleting the+marker of a directory which holds other keys stops it being listed,+and the directory is then removed with the last key below it, but this+is only remembered in memory, like the markers put without+`--directory-markers`.++As keys are file paths, a key can't be both an object and a directory+holding other objects, for example `a/b` and `a/b/c`. Writing an+object which would need this fails with `400 Bad Request`+(`InvalidRequest`).++## Compatibility++`serve s3` implements the parts of the S3 API that S3 clients commonly+use to store and fetch objects. Features it doesn't support are+refused with `501 NotImplemented`, or answered as AWS S3 answers for a+bucket without that feature configured, rather than being silently+ignored.++| Feature | Supported | Notes |+| ------- | --------- | ----- |+| `ListBuckets`, `CreateBucket`, `DeleteBucket`, `HeadBucket`, `GetBucketLocation` | Yes | Buckets are the directories in the root of the remote. |+| `ListObjects`, `ListObjectsV2` | Yes | Prefixes, delimiters, markers, `start-after`, continuation tokens, `fetch-owner` and `encoding-type=url` all work. Delimiters other than `/` need the whole directory tree walking. |+| `HeadObject`, `GetObject` | Yes | Including `Range`, `partNumber`, the `response-*` header overrides and the `If-Match`, `If-None-Match`, `If-Modified-Since` and `If-Unmodified-Since` conditions. |+| `PutObject` | Yes | Including `Content-MD5`, `aws-chunked` uploads and uploads with no `Content-Length`. |+| `CopyObject` | Yes | Including the metadata and tagging directives and the `x-amz-copy-source-if-*` conditions. |+| `DeleteObject`, `DeleteObjects` | Yes | |+| Multipart uploads | Yes | `CreateMultipartUpload`, `UploadPart`, `UploadPartCopy`, `CompleteMultipartUpload`, `AbortMultipartUpload`, `ListMultipartUploads` and `ListParts`. The `ETag` of the object is its hash, not the `-N` suffixed one AWS gives. The 5 MiB minimum part size isn't enforced. |+| Conditional writes | Partial | `If-Match` and `If-None-Match` on `PutObject` and `CompleteMultipartUpload` are checked, but not atomically against changes made to the remote some other way (see [Limitations](#limitations)). |+| Authentication | Partial | AWS Signature Version 4, in headers, presigned URLs and browser form (`POST`) uploads. Signature Version 2 isn't supported. |+| Anonymous access | Partial | All requests are allowed without `--auth-key` or `--auth-proxy`, and none are with them, as buckets and objects can't be made public. |+| Metadata | Partial | `x-amz-meta-*`, the storage class and the standard headers such as `Content-Type` are kept in memory only (see [Limitations](#limitations)), except the modification time. The storage class is only recorded, so doesn't change how the object is stored. |+| Directory markers | Partial | Keys ending in `/` are stored as directories, so can't hold data. A marker put for a directory holding other keys is only listed while it is remembered in memory, unless `--directory-markers` is set (see [Object names](#object-names)). |+| Checksums | Partial | The `x-amz-checksum-*` of a `PutObject` (CRC32, CRC32C, CRC64NVME, SHA1 or SHA256) is checked and returned with the metadata. The `Content-MD5` and `x-amz-checksum-*` of the bodies of other requests, such as `DeleteObjects`, are checked. The checksums of multipart uploads aren't supported. |+| `GetObjectAttributes` | Partial | Everything but the checksums of multipart uploads. |+| Object tagging | Partial | `GetObjectTagging`, `PutObjectTagging` and `DeleteObjectTagging` work, but tags are kept in memory with the metadata. |+| Bucket owner | No | `serve s3` has no AWS account IDs, so `x-amz-expected-bucket-owner` and `x-amz-source-expected-bucket-owner` are ignored. |+| ACLs | Partial | `GetBucketAcl` and `GetObjectAcl` give the owner full control. Only the `private` ACL can be set. |+| CORS | Partial | Preflight (`OPTIONS`) requests are answered, allowing any origin. CORS configuration (`PutBucketCors`) isn't supported. |+| Server side encryption | No | SSE-S3 (`AES256`) is accepted but the data isn't encrypted. SSE-KMS and SSE-C are refused. |+| Versioning | No | Buckets are reported as unversioned and `ListObjectVersions` lists each object as version `null`. Enabling versioning is refused. |+| Object lock, retention and legal hold | No | |+| Bucket configuration | No | Bucket policies, tagging, lifecycle, logging, encryption, website, replication, notifications and the like. Getting them fails as for a bucket without them configured. |+| `SelectObjectContent`, `RestoreObject`, torrents | No | |++The [ceph s3-tests](https://github.com/ceph/s3-tests) can be run+against `serve s3` to check its compatibility. See+[cmd/serve/s3/s3tests](https://github.com/rclone/rclone/tree/master/cmd/serve/s3/s3tests)+for how.### Authentication@@ -284,8 +629,20 @@ htpasswd -B htpasswd userhtpasswd -B htpasswd anotherUser```+Use the same hash format and cost for every entry in the file. If the+entries differ, the time taken to reject a wrong password can reveal+which usernames exist.+The password file can be updated while rclone is running.+When authentication is enabled every request must be authenticated,+including HTTP OPTIONS requests, as the answer to an OPTIONS request can+reveal whether a path exists. Health checks or monitoring which use+unauthenticated OPTIONS requests will need to send credentials. The+exception is browser CORS preflight requests when `--allow-origin`+is set, which are answered without authentication as browsers never send+credentials with them.+Use `--realm` to set the authentication realm.Use `--salt` to change the password hashing salt from the default.@@ -317,6 +674,13 @@ accept in the HTTP header.will overriding existing values. The flag may be repeated to add multipleheaders. Use the format `Header-Name: value`.+`--allow-origin` enables cross-origin (CORS) requests from browsers+by setting the `Access-Control-Allow-Origin` header on every response to the+origin given, e.g. `--allow-origin https://app.example.com`. Browser+CORS preflight OPTIONS requests are answered by rclone without needing+authentication as browsers never send credentials with them. Setting this+to `*` allows any origin which is a security risk on an authenticated server.+`--baseurl` controls the URL prefix that rclone serves from. By defaultrclone will serve from the root. If you used `--baseurl "/rclone"` thenrclone would serve from a URL starting with "/rclone/". This is@@ -436,7 +800,7 @@ at all times. The buffered data is bound to one open file and won't beshared.This flag is a upper limit for the used memory per open file. The-buffer will only use memory for data that is downloaded but not not+buffer will only use memory for data that is downloaded but notyet read. If the buffer is empty, only a small amount of memory willbe used.@@ -491,7 +855,8 @@ longest. This cache flushing strategy is efficient and more relevantfiles are likely to remain cached.The `--vfs-cache-max-age` will evict files from the cache-after the set time since last access has passed. The default value of+after the set time since last access has passed; it is based on access time,+not on when the file was first added to the cache. The default value of1 hour will start evicting files from cache that haven't been accessedfor 1 hour. When a cached file is accessed the 1 hour timer is reset to 0and will wait for 1 more hour before evicting. Specify the time with@@ -645,7 +1010,7 @@ Rclone reads `--vfs-read-chunk-streams` chunks of size`--vfs-read-chunk-size` concurrently. The size for each read will stayconstant.-This improves performance performance massively on high latency links+This improves performance massively on high latency linksor very high bandwidth links to high performance object stores.Some experimentation will be needed to find the optimum values of@@ -862,6 +1227,135 @@ If the file has no metadata it will be returned as `{}` and if thereis an error reading the metadata the error will be returned as`{"error":"error string"}`.+## Auth Proxy++If you supply the parameter `--auth-proxy /path/to/program` then+rclone will use that program to generate backends on the fly which+then are used to authenticate incoming requests. This uses a simple+JSON based protocol with input on STDIN and output on STDOUT.++**PLEASE NOTE:** `--auth-proxy` and `--authorized-keys` cannot be used+together, if `--auth-proxy` is set the authorized keys option will be+ignored.++There is an example program+[bin/test_proxy.py](https://github.com/rclone/rclone/blob/master/bin/test_proxy.py)+in the rclone source code.++The program's job is to take a `user` and `pass` on the input and turn+those into the config for a backend on STDOUT in JSON format. This+config will have any default parameters for the backend added, but it+won't use configuration from environment variables or command line+options - it is the job of the proxy program to make a complete+config.++This config generated must have this extra parameter++- `_root` - root to use for the backend++And it may have these parameters++- `_obscure` - comma separated strings for parameters to obscure+- `_secret_access_key` - the secret for S3 access key auth (see below)++If password authentication was used by the client, input to the proxy+process (on STDIN) would look similar to this:++```json+{+ "user": "me",+ "pass": "mypassword",+ "client_ip": "192.168.1.1"+}+```++If public-key authentication was used by the client, input to the+proxy process (on STDIN) would look similar to this:++```json+{+ "user": "me",+ "public_key": "AAAAB3NzaC1yc2EAAAADAQABAAABAQDuwESFdAe14hVS6omeyX7edc...JQdf",+ "client_ip": "192.168.1.1"+}+```++If the client authenticated with an S3 access key (`rclone serve s3`),+the client never sends its secret, only a signature made with it, so+the input contains just the access key ID as the `user` with no `pass`+or `public_key`:++```json+{+ "user": "AKIAIOSFODNN7EXAMPLE",+ "client_ip": "192.168.1.1"+}+```++In this case the program must look up the secret access key for that+access key ID and return it in the `_secret_access_key` field of the+output. Rclone then uses that secret to verify the signature on the+request, refusing the request if it does not match. This means the+proxy program is the source of truth for both the credentials and the+backend they map to. If the program does not return+`_secret_access_key` or returns it empty the request is refused.++The program's answer for an access key ID is cached (see below) but+is checked with the program again after 5 minutes even if the access+key ID is in constant use, so revoking an access key ID in the+program takes effect within 5 minutes. A rotated secret takes effect+on the first request signed with it.++The `client_ip` key holds the IP address the client connected from,+without a port number. It can be used to restrict logins to certain+networks, or to log authentication attempts centrally. It is omitted if+the client has no IP address, for example when connecting over a unix+socket. Note that if rclone is behind a reverse proxy this will be the+address of the reverse proxy and not the original client.++And as an example return this on STDOUT++```json+{+ "type": "sftp",+ "_root": "",+ "_obscure": "pass",+ "user": "me",+ "pass": "mypassword",+ "host": "sftp.example.com"+}+```++This would mean that an SFTP backend would be created on the fly for+the `user` and `pass`/`public_key` returned in the output to the host given. Note+that since `_obscure` is set to `pass`, rclone will obscure the `pass`+parameter before creating the backend (which is required for sftp+backends).++The program can manipulate the supplied `user` in any way, for example+to make proxy to many different sftp backends, you could make the+`user` be `[email protected]` and then set the `host` to `example.com`+in the output and the user to `user`. For security you'd probably want+to restrict the `host` to a limited list.++An internal cache of backends is keyed on the `user`, the kind of+credential presented, a hash of the `pass` or `public_key`, and the+`client_ip`. This means that if a user's password or public-key+changes, the client connects from a new IP address, or the proxy+returns different config parameters (eg a rotated `api_key`), a fresh+backend will be created on the next request rather than the cached one+being reused. A backend created for a public key login is never+reused for a password login, so a public key cannot be sent as a+password to skip the program.++A cached answer is checked with the program again after 5 minutes even+if the user is in constant use, so revoking a user in the program+takes effect within 5 minutes. For `serve sftp` this applies to new+logins - an existing connection is not closed.++This can be used to build general purpose proxies to any kind of+backend that rclone supports.+```rclone serve s3 remote:path [flags]```@@ -878,7 +1372,8 @@ rclone serve s3 remote:path [flags]--client-ca string Client certificate authority to verify clients with--dir-cache-time Duration Time to cache directory entries for (default 5m0s)--dir-perms FileMode Directory permissions (default 777)- --disable-multipart-streaming Buffer multipart uploads in memory instead of streaming them to the backend (see the Multipart uploads docs section)+ --directory-markers List every directory as a directory marker object+ --disable-multipart-streaming Buffer multipart uploads in memory instead of streaming them to the backend--etag-hash string Which hash to use for the ETag, or auto or blank for off (default "MD5")--file-perms FileMode File permissions (default 666)--force-path-style If true use path style access if false use virtual hosted style (default true)@@ -888,10 +1383,14 @@ rclone serve s3 remote:path [flags]--key string TLS PEM Private key--link-perms FileMode Link permissions (default 666)--max-header-bytes int Maximum size of request header (default 4096)+ --metadata-max-objects int Maximum number of objects of a user whose metadata is kept in memory, 0 for unlimited (default 100000)--min-tls-version string Minimum TLS version that is acceptable (default "tls1.0")- --multipart-streaming-buffer-limit SizeSuffix Maximum memory buffered per streamed multipart upload for parts arriving out of order, 0 for unlimited (see the Multipart uploads docs section) (default 256Mi)+ --multipart-expiry Duration Abort incomplete multipart uploads idle for longer than this, 0 to keep forever (default 1d)+ --multipart-max-uploads int Maximum number of streamed multipart uploads a user can have in progress at once, 0 for unlimited (default 1000)+ --multipart-streaming-buffer-limit SizeSuffix Maximum memory buffered per streamed multipart upload for parts arriving out of order, 0 for unlimited (default 256Mi)+ --multipart-streaming-buffer-total SizeSuffix Maximum memory buffered by all the streamed multipart uploads of a user for parts arriving out of order, 0 for unlimited (default 1Gi)--no-checksum Don't compare checksums on up/download- --no-cleanup Not to cleanup empty folder after object is deleted+ --no-cleanup Deprecated: use --directory-markers instead--no-modtime Don't read/write the modification time (can speed things up)--no-seek Don't allow seeking in files--pass string Password for authenticationdiff --git a/docs/content/commands/rclone_serve_sftp.md b/docs/content/commands/rclone_serve_sftp.mdindex d74f23bea..a9c3958c1 100644--- a/docs/content/commands/rclone_serve_sftp.md+++ b/docs/content/commands/rclone_serve_sftp.md@@ -140,7 +140,7 @@ at all times. The buffered data is bound to one open file and won't beshared.This flag is a upper limit for the used memory per open file. The-buffer will only use memory for data that is downloaded but not not+buffer will only use memory for data that is downloaded but notyet read. If the buffer is empty, only a small amount of memory willbe used.@@ -195,7 +195,8 @@ longest. This cache flushing strategy is efficient and more relevantfiles are likely to remain cached.The `--vfs-cache-max-age` will evict files from the cache-after the set time since last access has passed. The default value of+after the set time since last access has passed; it is based on access time,+not on when the file was first added to the cache. The default value of1 hour will start evicting files from cache that haven't been accessedfor 1 hour. When a cached file is accessed the 1 hour timer is reset to 0and will wait for 1 more hour before evicting. Specify the time with@@ -349,7 +350,7 @@ Rclone reads `--vfs-read-chunk-streams` chunks of size`--vfs-read-chunk-size` concurrently. The size for each read will stayconstant.-This improves performance performance massively on high latency links+This improves performance massively on high latency linksor very high bandwidth links to high performance object stores.Some experimentation will be needed to find the optimum values of@@ -592,9 +593,10 @@ This config generated must have this extra parameter- `_root` - root to use for the backend-And it may have this parameter+And it may have these parameters- `_obscure` - comma separated strings for parameters to obscure+- `_secret_access_key` - the secret for S3 access key auth (see below)If password authentication was used by the client, input to the proxyprocess (on STDIN) would look similar to this:@@ -602,7 +604,8 @@ process (on STDIN) would look similar to this:```json{"user": "me",- "pass": "mypassword"+ "pass": "mypassword",+ "client_ip": "192.168.1.1"}```@@ -612,10 +615,44 @@ proxy process (on STDIN) would look similar to this:```json{"user": "me",- "public_key": "AAAAB3NzaC1yc2EAAAADAQABAAABAQDuwESFdAe14hVS6omeyX7edc...JQdf"+ "public_key": "AAAAB3NzaC1yc2EAAAADAQABAAABAQDuwESFdAe14hVS6omeyX7edc...JQdf",+ "client_ip": "192.168.1.1"+}+```++If the client authenticated with an S3 access key (`rclone serve s3`),+the client never sends its secret, only a signature made with it, so+the input contains just the access key ID as the `user` with no `pass`+or `public_key`:++```json+{+ "user": "AKIAIOSFODNN7EXAMPLE",+ "client_ip": "192.168.1.1"}```+In this case the program must look up the secret access key for that+access key ID and return it in the `_secret_access_key` field of the+output. Rclone then uses that secret to verify the signature on the+request, refusing the request if it does not match. This means the+proxy program is the source of truth for both the credentials and the+backend they map to. If the program does not return+`_secret_access_key` or returns it empty the request is refused.++The program's answer for an access key ID is cached (see below) but+is checked with the program again after 5 minutes even if the access+key ID is in constant use, so revoking an access key ID in the+program takes effect within 5 minutes. A rotated secret takes effect+on the first request signed with it.++The `client_ip` key holds the IP address the client connected from,+without a port number. It can be used to restrict logins to certain+networks, or to log authentication attempts centrally. It is omitted if+the client has no IP address, for example when connecting over a unix+socket. Note that if rclone is behind a reverse proxy this will be the+address of the reverse proxy and not the original client.+And as an example return this on STDOUT```json@@ -641,11 +678,20 @@ to make proxy to many different sftp backends, you could make thein the output and the user to `user`. For security you'd probably wantto restrict the `host` to a limited list.-An internal cache of backends is keyed on the `user` and a hash of the-`pass` or `public_key`. This means that if a user's password or-public-key changes, or the proxy returns different config parameters-(eg a rotated `api_key`), a fresh backend will be created on the next-request rather than the cached one being reused.+An internal cache of backends is keyed on the `user`, the kind of+credential presented, a hash of the `pass` or `public_key`, and the+`client_ip`. This means that if a user's password or public-key+changes, the client connects from a new IP address, or the proxy+returns different config parameters (eg a rotated `api_key`), a fresh+backend will be created on the next request rather than the cached one+being reused. A backend created for a public key login is never+reused for a password login, so a public key cannot be sent as a+password to skip the program.++A cached answer is checked with the program again after 5 minutes even+if the user is in constant use, so revoking a user in the program+takes effect within 5 minutes. For `serve sftp` this applies to new+logins - an existing connection is not closed.This can be used to build general purpose proxies to any kind ofbackend that rclone supports.diff --git a/docs/content/commands/rclone_serve_webdav.md b/docs/content/commands/rclone_serve_webdav.mdindex 3196a03ab..b7d4ac1be 100644--- a/docs/content/commands/rclone_serve_webdav.md+++ b/docs/content/commands/rclone_serve_webdav.md@@ -126,6 +126,13 @@ accept in the HTTP header.will overriding existing values. The flag may be repeated to add multipleheaders. Use the format `Header-Name: value`.+`--allow-origin` enables cross-origin (CORS) requests from browsers+by setting the `Access-Control-Allow-Origin` header on every response to the+origin given, e.g. `--allow-origin https://app.example.com`. Browser+CORS preflight OPTIONS requests are answered by rclone without needing+authentication as browsers never send credentials with them. Setting this+to `*` allows any origin which is a security risk on an authenticated server.+`--baseurl` controls the URL prefix that rclone serves from. By defaultrclone will serve from the root. If you used `--baseurl "/rclone"` thenrclone would serve from a URL starting with "/rclone/". This is@@ -195,6 +202,13 @@ to be used within the template to server pages:| .Sort | | The current sort used. This is changeable via '?sort=' parameter. Possible values: namedirfirst, name, size, time (default namedirfirst). || .Order | | The current ordering used. This is changeable via '?order=' parameter. Possible values: asc, desc (default asc). || .Query | | Currently unused. |+| .Path | | The path of the directory from the root with leading and trailing slashes, e.g. '/dir/'. |+| .IsRoot | | Boolean for if this is the root directory. |+| .UpLink | | The link to the parent directory. |+| .Static | | Boolean for if the listing is being written as a static page by rclone index. |+| .NumDirs | | The number of directories in the listing. |+| .NumFiles | | The number of files in the listing. |+| .TotalSize | | The total size in bytes of the files in the listing. || .Breadcrumb | | Allows for creating a relative navigation. || | .Link | The link of the Text relative to the root. || | .Text | The Name of the directory. |@@ -204,6 +218,7 @@ to be used within the template to server pages:| | .IsDir | Boolean for if an entry is a directory or not. || | .Size | Size in bytes of the entry. || | .ModTime | The UTC timestamp of an entry. |+| | .MimeType | The MIME type of an entry guessed from its name, or 'inode/directory'. |The server also makes the following functions available so that they can be usedwithin the template. These functions help extend the options for dynamic@@ -244,8 +259,20 @@ htpasswd -B htpasswd userhtpasswd -B htpasswd anotherUser```+Use the same hash format and cost for every entry in the file. If the+entries differ, the time taken to reject a wrong password can reveal+which usernames exist.+The password file can be updated while rclone is running.+When authentication is enabled every request must be authenticated,+including HTTP OPTIONS requests, as the answer to an OPTIONS request can+reveal whether a path exists. Health checks or monitoring which use+unauthenticated OPTIONS requests will need to send credentials. The+exception is browser CORS preflight requests when `--allow-origin`+is set, which are answered without authentication as browsers never send+credentials with them.+Use `--realm` to set the authentication realm.Use `--salt` to change the password hashing salt from the default.@@ -313,7 +340,7 @@ at all times. The buffered data is bound to one open file and won't beshared.This flag is a upper limit for the used memory per open file. The-buffer will only use memory for data that is downloaded but not not+buffer will only use memory for data that is downloaded but notyet read. If the buffer is empty, only a small amount of memory willbe used.@@ -368,7 +395,8 @@ longest. This cache flushing strategy is efficient and more relevantfiles are likely to remain cached.The `--vfs-cache-max-age` will evict files from the cache-after the set time since last access has passed. The default value of+after the set time since last access has passed; it is based on access time,+not on when the file was first added to the cache. The default value of1 hour will start evicting files from cache that haven't been accessedfor 1 hour. When a cached file is accessed the 1 hour timer is reset to 0and will wait for 1 more hour before evicting. Specify the time with@@ -522,7 +550,7 @@ Rclone reads `--vfs-read-chunk-streams` chunks of size`--vfs-read-chunk-size` concurrently. The size for each read will stayconstant.-This improves performance performance massively on high latency links+This improves performance massively on high latency linksor very high bandwidth links to high performance object stores.Some experimentation will be needed to find the optimum values of@@ -765,9 +793,10 @@ This config generated must have this extra parameter- `_root` - root to use for the backend-And it may have this parameter+And it may have these parameters- `_obscure` - comma separated strings for parameters to obscure+- `_secret_access_key` - the secret for S3 access key auth (see below)If password authentication was used by the client, input to the proxyprocess (on STDIN) would look similar to this:@@ -775,7 +804,8 @@ process (on STDIN) would look similar to this:```json{"user": "me",- "pass": "mypassword"+ "pass": "mypassword",+ "client_ip": "192.168.1.1"}```@@ -785,10 +815,44 @@ proxy process (on STDIN) would look similar to this:```json{"user": "me",- "public_key": "AAAAB3NzaC1yc2EAAAADAQABAAABAQDuwESFdAe14hVS6omeyX7edc...JQdf"+ "public_key": "AAAAB3NzaC1yc2EAAAADAQABAAABAQDuwESFdAe14hVS6omeyX7edc...JQdf",+ "client_ip": "192.168.1.1"}```+If the client authenticated with an S3 access key (`rclone serve s3`),+the client never sends its secret, only a signature made with it, so+the input contains just the access key ID as the `user` with no `pass`+or `public_key`:++```json+{+ "user": "AKIAIOSFODNN7EXAMPLE",+ "client_ip": "192.168.1.1"+}+```++In this case the program must look up the secret access key for that+access key ID and return it in the `_secret_access_key` field of the+output. Rclone then uses that secret to verify the signature on the+request, refusing the request if it does not match. This means the+proxy program is the source of truth for both the credentials and the+backend they map to. If the program does not return+`_secret_access_key` or returns it empty the request is refused.++The program's answer for an access key ID is cached (see below) but+is checked with the program again after 5 minutes even if the access+key ID is in constant use, so revoking an access key ID in the+program takes effect within 5 minutes. A rotated secret takes effect+on the first request signed with it.++The `client_ip` key holds the IP address the client connected from,+without a port number. It can be used to restrict logins to certain+networks, or to log authentication attempts centrally. It is omitted if+the client has no IP address, for example when connecting over a unix+socket. Note that if rclone is behind a reverse proxy this will be the+address of the reverse proxy and not the original client.+And as an example return this on STDOUT```json@@ -814,11 +878,20 @@ to make proxy to many different sftp backends, you could make thein the output and the user to `user`. For security you'd probably wantto restrict the `host` to a limited list.-An internal cache of backends is keyed on the `user` and a hash of the-`pass` or `public_key`. This means that if a user's password or-public-key changes, or the proxy returns different config parameters-(eg a rotated `api_key`), a fresh backend will be created on the next-request rather than the cached one being reused.+An internal cache of backends is keyed on the `user`, the kind of+credential presented, a hash of the `pass` or `public_key`, and the+`client_ip`. This means that if a user's password or public-key+changes, the client connects from a new IP address, or the proxy+returns different config parameters (eg a rotated `api_key`), a fresh+backend will be created on the next request rather than the cached one+being reused. A backend created for a public key login is never+reused for a password login, so a public key cannot be sent as a+password to skip the program.++A cached answer is checked with the program again after 5 minutes even+if the user is in constant use, so revoking a user in the program+takes effect within 5 minutes. For `serve sftp` this applies to new+logins - an existing connection is not closed.This can be used to build general purpose proxies to any kind ofbackend that rclone supports.diff --git a/docs/content/commands/rclone_test_speed.md b/docs/content/commands/rclone_test_speed.mdindex 8a5cb9697..32f4d452d 100644--- a/docs/content/commands/rclone_test_speed.md+++ b/docs/content/commands/rclone_test_speed.md@@ -16,7 +16,7 @@ This command runs a series of uploads and downloads to the remote, measuringand printing the speed of each test using varying file sizes and numbers offiles.-Test time can be innaccurate with small file caps and large files. As it+Test time can be inaccurate with small file caps and large files. As ituses the results of an initial test to determine how many files to use ineach subsequent test.diff --git a/docs/content/drive.md b/docs/content/drive.mdindex 4c9f07111..04f0f007f 100644--- a/docs/content/drive.md+++ b/docs/content/drive.md@@ -1131,7 +1131,7 @@ recommended usage is using the flag form --drive-size-as-quota whendoing rclone ls/lsl/lsf/lsjson/etc only.If you do use this flag for syncing (not recommended) then you will-need to use --ignore size also.+need to use --ignore-size also.Properties:@@ -1838,7 +1838,7 @@ The result is a JSON array of matches, for example:"webViewLink": "https://drive.google.com/file/d/0AxBe_CDEF4zkGHI4d0FjYko2QkD/view?usp=drivesdk\u0026resourcekey=0-ABCDEFGHIXJQpIGqBJq3MC"}]-```+```console### rescuediff --git a/docs/content/flags.md b/docs/content/flags.mdindex 335608f0a..7e933490b 100644--- a/docs/content/flags.md+++ b/docs/content/flags.md@@ -98,6 +98,7 @@ Flags used for check commands.Flags for general networking and HTTP stuff.```+ --allow-https-downgrade Follow redirects from HTTPS to HTTP (insecure)--bind string Local address to bind to for outgoing connections, IPv4, IPv6 or name--bwlimit BwTimetable Bandwidth limit in KiB/s, or use suffix B|K|M|G|T|P or a full timetable--bwlimit-file BwTimetable Bandwidth limit per file in KiB/s, or use suffix B|K|M|G|T|P or a full timetable@@ -121,7 +122,7 @@ Flags for general networking and HTTP stuff.--tpslimit float Limit HTTP transactions per second to this--tpslimit-burst int Max burst of transactions for --tpslimit (default 1)--use-cookies Enable session cookiejar- --user-agent string Set the user-agent to a specified string (default "rclone/v1.75.0")+ --user-agent string Set the user-agent to a specified string (default "rclone/v1.76.0")```@@ -302,11 +303,6 @@ Flags to control the Remote Control API.--rc-template string User-specified template--rc-user string User name for authentication--rc-user-from-header string User name from a defined HTTP header- --rc-web-fetch-url string URL to fetch the releases for webgui (default "https://api.github.com/repos/rclone/rclone-webui-react/releases/latest")- --rc-web-gui Launch WebGUI on localhost- --rc-web-gui-force-update Force update to latest version of web gui- --rc-web-gui-no-open-browser Don't open the browser automatically- --rc-web-gui-update Check and update to latest version of web gui```@@ -368,8 +364,10 @@ Backend-only flags (these can be set in the config file also).--azureblob-encoding Encoding The encoding for the backend (default Slash,BackSlash,Del,Ctl,RightPeriod,InvalidUtf8)--azureblob-endpoint string Endpoint for the service--azureblob-env-auth Read credentials from runtime (environment variables, CLI or MSI)+ --azureblob-hns Tristate Set if the storage account has a hierarchical namespace (default unset)--azureblob-key string Storage Account Shared Key--azureblob-list-chunk int Size of blob list (default 5000)+ --azureblob-list-parallelism int Number of parallel shards to list a directory with--azureblob-msi-client-id string Object ID of the user-assigned MSI to use, if any--azureblob-msi-mi-res-id string Azure resource ID of the user-assigned MSI to use, if any--azureblob-msi-object-id string Object ID of the user-assigned MSI to use, if any@@ -382,6 +380,7 @@ Backend-only flags (these can be set in the config file also).--azureblob-tenant string ID of the service principal's tenant. Also called its directory ID--azureblob-upload-concurrency int Concurrency for multipart uploads (default 16)--azureblob-upload-cutoff string Cutoff for switching to chunked upload (<= 256 MiB) (deprecated)+ --azureblob-use-arrow-list Use the Apache Arrow listing format--azureblob-use-az Use Azure CLI tool az for authentication--azureblob-use-copy-blob Whether to use the Copy Blob API when copying to the same storage account (default true)--azureblob-use-emulator Uses local storage emulator if provided as 'true'@@ -514,6 +513,12 @@ Backend-only flags (these can be set in the config file also).--doi-doi string The DOI or the doi.org URL--doi-doi-resolver-api-url string The URL of the DOI resolver API to use--doi-provider string DOI provider+ --dosya-api-key string Your API Key, get it from https://dosya.dev/profile under "API keys"+ --dosya-api-url string API base URL (default "https://api.dosya.dev")+ --dosya-description string Description of the remote+ --dosya-encoding Encoding The encoding for the backend (default Slash,LtGt,DoubleQuote,SingleQuote,BackSlash,Del,Ctl,LeftSpace,RightSpace,InvalidUtf8,Dot)+ --dosya-upload-concurrency int Concurrency for multipart uploads (default 4)+ --dosya-workspace-id string Your workspace ID--drime-access-token string API Access token--drime-chunk-size SizeSuffix Chunk size to use for uploading (default 5Mi)--drime-description string Description of the remote@@ -736,6 +741,8 @@ Backend-only flags (these can be set in the config file also).--http-headers CommaSepList Set HTTP headers for all transactions--http-no-escape Do not escape URL metacharacters in path names--http-no-head Don't use HEAD requests+ --http-no-head-size SizeSuffix Set the reported file size for --http-no-head (default off)+ --http-no-head-time Time Set a default modification time for --http-no-head (default off)--http-no-slash Set this if the site doesn't end directories with /--http-url string URL of HTTP host to connect to--huaweidrive-auth-url string Auth server URL@@ -795,6 +802,17 @@ Backend-only flags (these can be set in the config file also).--jottacloud-token-url string Token server url--jottacloud-trashed-only Only show files that are in the trash--jottacloud-upload-resume-limit SizeSuffix Files bigger than this can be resumed if the upload fail's (default 10Mi)+ --kdrive-auth-url string Auth server URL+ --kdrive-chunk-size SizeSuffix Files above this size are uploaded in chunks of this size (default 20Mi)+ --kdrive-client-id string OAuth Client Id+ --kdrive-description string Description of the remote+ --kdrive-drive-id string ID of the kDrive to use+ --kdrive-encoding Encoding The encoding for the backend (default Slash,Del,Ctl,LeftSpace,RightSpace,InvalidUtf8,Dot)+ --kdrive-endpoint string The API endpoint to use+ --kdrive-root-folder-id string The folder to use as root+ --kdrive-token string OAuth Access Token as a JSON blob+ --kdrive-token-url string Token server url+ --kdrive-upload-concurrency int Concurrency for chunked uploads (default 4)--koofr-description string Description of the remote--koofr-encoding Encoding The encoding for the backend (default Slash,BackSlash,Del,Ctl,InvalidUtf8,Dot)--koofr-endpoint string The Koofr API endpoint to use@@ -880,6 +898,7 @@ Backend-only flags (these can be set in the config file also).--onedrive-root-folder-id string ID of the root folder--onedrive-server-side-across-configs Deprecated: use --server-side-across-configs instead--onedrive-tenant string ID of the service principal's tenant. Also called its directory ID+ --onedrive-tenant-api-version string The SharePoint API version to use with tenant_url (default "v2.0")--onedrive-tenant-url string The tenant URL for non-admin OneDrive access--onedrive-token string OAuth Access Token as a JSON blob--onedrive-token-url string Token server url@@ -894,7 +913,7 @@ Backend-only flags (these can be set in the config file also).--oos-decompress If set this will decompress gzip encoded objects--oos-description string Description of the remote--oos-disable-checksum Don't store MD5 checksum with object metadata- --oos-encoding Encoding The encoding for the backend (default Slash,InvalidUtf8,Dot)+ --oos-encoding Encoding The encoding for the backend (default Slash,CrLf,InvalidUtf8,Dot)--oos-endpoint string Endpoint for Object storage API--oos-leave-parts-on-error If true avoid calling abort upload on a failure, leaving all successfully uploaded parts for manual recovery--oos-max-upload-parts int Maximum number of parts in a multipart upload (default 10000)@@ -1089,7 +1108,7 @@ Backend-only flags (these can be set in the config file also).--sftp-disable-concurrent-reads If set don't use concurrent reads--sftp-disable-concurrent-writes If set don't use concurrent writes--sftp-disable-hashcheck Disable the execution of SSH commands to determine if remote file hashing is available- --sftp-encoding Encoding The encoding for the backend (default Slash,Del,Ctl,Dot)+ --sftp-encoding Encoding The encoding for the backend (default Slash,Dot)--sftp-hashes CommaSepList Comma separated list of supported checksum types--sftp-host string SSH host to connect to--sftp-host-key-algorithms SpaceSepList Space separated list of host key algorithms, ordered by preference@@ -1104,6 +1123,7 @@ Backend-only flags (these can be set in the config file also).--sftp-known-hosts-file string Optional path to known_hosts file--sftp-macs SpaceSepList Space separated list of MACs (message authentication code) algorithms, ordered by preference--sftp-md5sum-command string The command used to read MD5 hashes+ --sftp-multithread-upload Set this to use multi-thread (multi-connection) uploads--sftp-pass string SSH password, leave blank to use ssh-agent (obscured)--sftp-path-override string Override path used by SSH shell commands--sftp-pin-host-key Pin the server host key on first connection (Trust On First Use)@@ -1166,7 +1186,8 @@ Backend-only flags (these can be set in the config file also).--smb-port int SMB port number (default 445)--smb-spn string Service principal name--smb-use-kerberos Use Kerberos authentication- --smb-user string SMB username (default "$USER")+ --smb-user string SMB username+ --smb-workstation string Workstation name sent in NTLM authentication--storj-access-grant string Access grant--storj-api-key string API key--storj-description string Description of the remote@@ -1241,6 +1262,7 @@ Backend-only flags (these can be set in the config file also).--webdav-url string URL of http host to connect to--webdav-user string User name--webdav-vendor string Name of the WebDAV site/service/software you are using+ --yandex-app-folder Use the application folder as the root--yandex-auth-url string Auth server URL--yandex-client-credentials Use client credentials OAuth flow--yandex-client-id string OAuth Client Iddiff --git a/docs/content/http.md b/docs/content/http.mdindex c3c8beccd..9f4e36d8e 100644--- a/docs/content/http.md+++ b/docs/content/http.md@@ -212,6 +212,13 @@ For example, to set a Cookie use 'Cookie,name=value', or '"Cookie","name=value"'You can set multiple headers, e.g. '"Cookie","name=value","Authorization","xxx"'.+The headers are only sent to the host in the configured URL. If the+server redirects to another host (including a subdomain or a different+port) the headers are not sent to it, or to any further hop in that+redirect chain. When headers are set, a redirect from https to http is+refused as it would send them in cleartext, even with+--allow-https-downgrade.+Properties:- Config: headers@@ -265,6 +272,42 @@ Properties:- Type: bool- Default: false+#### --http-no-head-size++Set the reported file size for --http-no-head.++When HEAD requests are disabled, rclone normally reports an unknown file size.+Set this option to a non-negative value to report that size for every file,+including 0. Leave it unset (the default, shown as "off") to keep reporting+an unknown size.++The configured size is authoritative. If it does not match the actual file+size, transfers may fail their size check and retry unless --ignore-size is+used. It can also make multi-thread transfers request incorrect ranges and+make mounted files unreadable past the reported size.++Properties:++- Config: no_head_size+- Env Var: RCLONE_HTTP_NO_HEAD_SIZE+- Type: SizeSuffix+- Default: off++#### --http-no-head-time++Set a default modification time for --http-no-head.++When HEAD requests are disabled, rclone normally reports an unset modification+time. Set this option to report the specified time for every file instead.+Changing the configured time later makes every file appear modified to a sync.++Properties:++- Config: no_head_time+- Env Var: RCLONE_HTTP_NO_HEAD_TIME+- Type: Time+- Default: off+#### --http-descriptionDescription of the remote.diff --git a/docs/content/local.md b/docs/content/local.mdindex 433a48774..68ee79948 100644--- a/docs/content/local.md+++ b/docs/content/local.md@@ -583,6 +583,13 @@ deduplicated by default (i.e. having two identical files does not consume morestorage than having just one.) However, for use cases where data redundancy ispreferable, --local-no-clone can be used to disable cloning and force "deep" copies.+Cloning also preserves macOS-specific metadata such as extended attributes,+which a normal copy does not, so this metadata is lost when cloning is disabled.++If the destination file already exists, cloning replaces it with a new file,+even with --inplace, so hard links to it are broken. Use --local-no-clone to+update existing files in place.+Currently, cloning is only supported when using APFS on macOS (support for otherplatforms may be added in the future.)@@ -672,10 +679,10 @@ Properties:Make out-of-space errors fatal during transfers.-When enabled, an ENOSPC error during a write returns a fatal error so-that rclone aborts rather than retrying the operation. Useful for-backup scripts that should halt loudly on a full disk rather than spin-retrying.+When enabled, an out of space error while writing, creating a directory,+or creating a file returns a fatal error so that rclone aborts rather+than retrying the operation. Useful for backup scripts that should halt+loudly on a full disk rather than spin retrying.Properties:@@ -773,8 +780,10 @@ Metadata is supported on files and directories.When restoring metadata with `--metadata` rclone applies the"mode", "uid" and "gid" from the source. These come from the sourceremote which may not be trusted, so restoring metadata as root from an-untrusted source can change file ownership and is not recommended. The-setuid, setgid and sticky bits are not restored by default - see the+untrusted source can change file ownership and is not recommended. A+user other than root keeps its own uid and sets the gid only to a group+it is a member of, as rsync does. The setuid, setgid and sticky bits are+not restored by default - see the`--local-metadata-restore-special-bits` flag.Here are the possible system metadata items for the local backend.diff --git a/docs/content/onedrive.md b/docs/content/onedrive.mdindex 88884003c..e3df69c9d 100644--- a/docs/content/onedrive.md+++ b/docs/content/onedrive.md@@ -528,7 +528,7 @@ Properties:The tenant URL for non-admin OneDrive access.-Set this to your SharePoint tenant URL to use the SharePoint v2.0 API+Set this to your SharePoint tenant URL to use the SharePoint APIendpoint instead of the standard Microsoft Graph API. This allowsaccessing business OneDrive without admin consent.@@ -545,6 +545,22 @@ Properties:- Type: string- Required: false+#### --onedrive-tenant-api-version++The SharePoint API version to use with tenant_url.++Set this to the SharePoint API version matching the browser-extracted+access token. For example, use v2.1 with a driveAccessTokenV21 token.++This only applies when tenant_url is set.++Properties:++- Config: tenant_api_version+- Env Var: RCLONE_ONEDRIVE_TENANT_API_VERSION+- Type: string+- Default: "v2.0"+#### --onedrive-chunk-sizeChunk size to upload files with - must be multiple of 320k (327,680 bytes).@@ -671,7 +687,7 @@ Properties:Remove all versions on modifying operations.-Onedrive for business creates versions when rclone uploads new files+Onedrive creates versions when rclone uploads new filesoverwriting an existing one and when it sets the modification time.These versions take up space out of the quota.@@ -679,9 +695,6 @@ These versions take up space out of the quota.This flag checks for versions after file upload and settingmodification time and removes all but the last version.-**NB** Onedrive personal can't currently delete versions so don't use-this flag there.-Properties:@@ -697,8 +710,7 @@ Permanently delete files on removal.Normally files will get sent to the recycle bin on deletion. Settingthis flag causes them to be permanently deleted. Use with care.-OneDrive personal accounts do not support the permanentDelete API,-it only applies to OneDrive for Business and SharePoint document libraries.+This works with OneDrive for Business, SharePoint document libraries, and OneDrive personal accounts, including free accounts.Properties:@@ -744,12 +756,17 @@ Properties:- Creates a read-write link to the item.- "embed"- Creates an embeddable link to the item.+ - Only available in OneDrive personal.#### --onedrive-link-passwordSet the password for links created by the link command.-At the time of writing this only works with OneDrive personal paid accounts.+At the time of writing this works with OneDrive for Business and+OneDrive personal paid accounts.++OneDrive personal free accounts can't set a password or an expiry time+(with --expire) on links.Properties:@@ -848,21 +865,12 @@ Setting this flag speeds up these things greatly:rclone size onedrive:rclone rc vfs/refresh recursive=true-**However** the delta listing API **only** works at the root of the-drive. If you use it not at the root then it recurses from the root-and discards all the data that is not under the directory you asked-for. So it will be correct but may not be very efficient.--This is why this flag is not set as the default.--As a rule of thumb if nearly all of your data is under rclone's root-directory (the `root/directory` in `onedrive:root/directory`) then-using this flag will be a big performance win. If your data is-mostly not under the root then using this flag will be a big-performance loss.--It is recommended if you are mounting your onedrive at the root-(or near the root when using crypt) and using rclone `rc vfs/refresh`.+Rclone asks for the delta listing of the directory being listed. If+the drive only supports delta listings at the root of the drive (as+some older OneDrive for Business and SharePoint drives do) then+rclone lists from the root and discards all the data that is not+under the directory you asked for. So it will be correct but may not+be very efficient.Properties:@@ -1060,7 +1068,7 @@ Here are the possible system metadata items for the onedrive backend.| Name | Help | Type | Example | Read Only ||------|------|------|---------|-----------|-| btime | Time of file birth (creation) with S accuracy (mS for OneDrive Personal). | RFC 3339 | 2006-01-02T15:04:05Z | N |+| btime | Time of file birth (creation) with S accuracy. | RFC 3339 | 2006-01-02T15:04:05Z | N || content-type | The MIME type of the file. | string | text/plain | **Y** || created-by-display-name | Display name of the user that created the item. | string | John Doe | **Y** || created-by-id | ID of the user that created the item. | string | 48d31887-5fad-4d73-a9f5-3c356e68a038 | **Y** |@@ -1069,14 +1077,14 @@ Here are the possible system metadata items for the onedrive backend.| last-modified-by-display-name | Display name of the user that last modified the item. | string | John Doe | **Y** || last-modified-by-id | ID of the user that last modified the item. | string | 48d31887-5fad-4d73-a9f5-3c356e68a038 | **Y** || malware-detected | Whether OneDrive has detected that the item contains malware. | boolean | true | **Y** |-| mtime | Time of last modification with S accuracy (mS for OneDrive Personal). | RFC 3339 | 2006-01-02T15:04:05Z | N |+| mtime | Time of last modification with S accuracy. | RFC 3339 | 2006-01-02T15:04:05Z | N || package-type | If present, indicates that this item is a package instead of a folder or file. Packages are treated like files in some contexts and folders in others. | string | oneNote | **Y** || permissions | Permissions in a JSON dump of OneDrive format. Enable with --onedrive-metadata-permissions. Properties: id, grantedTo, grantedToIdentities, invitation, inheritedFrom, link, roles, shareId | JSON | {} | N || shared-by-id | ID of the user that shared the item (if shared). | string | 48d31887-5fad-4d73-a9f5-3c356e68a038 | **Y** || shared-owner-id | ID of the owner of the shared item (if shared). | string | 48d31887-5fad-4d73-a9f5-3c356e68a038 | **Y** || shared-scope | If shared, indicates the scope of how the item is shared: anonymous, organization, or users. | string | users | **Y** |-| shared-time | Time when the item was shared, with S accuracy (mS for OneDrive Personal). | RFC 3339 | 2006-01-02T15:04:05Z | **Y** |-| utime | Time of upload with S accuracy (mS for OneDrive Personal). | RFC 3339 | 2006-01-02T15:04:05Z | **Y** |+| shared-time | Time when the item was shared, with S accuracy. | RFC 3339 | 2006-01-02T15:04:05Z | **Y** |+| utime | Time of upload with S accuracy. | RFC 3339 | 2006-01-02T15:04:05Z | **Y** |See the [metadata](/docs/#metadata) docs for more info.diff --git a/docs/content/oracleobjectstorage/_index.md b/docs/content/oracleobjectstorage/_index.mdindex 5398e1e9a..09b05ccd5 100644--- a/docs/content/oracleobjectstorage/_index.md+++ b/docs/content/oracleobjectstorage/_index.md@@ -610,7 +610,7 @@ Properties:- Config: encoding- Env Var: RCLONE_OOS_ENCODING- Type: Encoding-- Default: Slash,InvalidUtf8,Dot+- Default: Slash,CrLf,InvalidUtf8,Dot#### --oos-leave-parts-on-errordiff --git a/docs/content/pikpak.md b/docs/content/pikpak.mdindex bb374c9d8..d446fd21e 100644--- a/docs/content/pikpak.md+++ b/docs/content/pikpak.md@@ -196,6 +196,8 @@ Properties:Files bigger than this will be cached on disk to calculate hash if required.+Files of unknown size are always cached on disk.+Properties:- Config: hash_memory_limitdiff --git a/docs/content/rc.md b/docs/content/rc.mdindex 309437921..33aa23777 100644--- a/docs/content/rc.md+++ b/docs/content/rc.md@@ -1225,6 +1225,7 @@ Returns the following values:"totalTransfers": total number of transfers in the group,"transferTime" : total time spent on running jobs,"transfers": number of transferred files,+ "updatedDirs": number of directories updated (modtime or metadata set),"transferring": an array of currently active file transfers:[{@@ -1490,8 +1491,6 @@ Results:- runningIds - array of integer job ids that are running- finishedIds - array of integer job ids that are finished-**Authentication is not required for this call.**-### job/status: Reads the status of the job ID {#job-status}Parameters:@@ -1512,8 +1511,6 @@ Results:- output - output of the job as would have been returned if called synchronously- progress - output of the progress related to the underlying job-**Authentication is not required for this call.**-### job/stop: Stop the running job {#job-stop}Parameters:@@ -1843,6 +1840,25 @@ This command does not have a command line equivalent so use this instead:rclone rc --loopback operations/fsinfo fs=remote:+### operations/getfile: Get a single file into a JSON response. {#operations-getfile}++This takes the following parameters:++- fs - a remote name string e.g. "drive:path/to/file" or "drive:path/to/dir"+- remote - (optional) a path within that remote e.g. "file.txt"+- offset - (optional) start reading at offset N (or from end if negative) (default 0)+- count - (optional) only read N bytes (default -1)+- head - (optional) only read the first N bytes (default 0)+- tail - (optional) only read the last N bytes (default 0)+- maxSize - (optional) maximum size limit in bytes (can only lower the built-in maximum) (default 2097152 / 2MiB)+- base64 - (optional) if true, returns the file contents base64-encoded (default false)++Returns:++- result - contents of the file as string or base64-encoded string++For streaming file content over HTTP with Range support, use the --rc-serve flag instead.+### operations/hashsum: Produces a hashsum file for all the objects in the path. {#operations-hashsum}Produces a hash file for all the objects in the path using the hash@@ -1917,6 +1933,26 @@ Example:See the [hashsum](/commands/rclone_hashsum/) command for more information on the above.+### operations/index: Write static directory listings into a remote {#operations-index}++This takes the following parameters:++- fs - a remote name string e.g. "drive:"+- opt - a dictionary of options to control the indexing (optional)+ - output - list of listings to write in each directory as "NAME=FORMAT" (default ["index.html=html"])+ - rules - which directories get listings, e.g. {"ExcludeRule": ["/private/**"]}+ - maxDepth - only make listings this many directories deep (default -1 for no limit)+ - linkIndex - link to "dir/NAME" rather than "dir/"+ - dirTime - how to work out directory times: "newest", "dir" or "none"+ - noModTime - don't show modification times+ - rewrite - write every listing even if it is unchanged+ - changed - list of changed paths for a partial run, directories ending in "/"+ - changedFrom - list of files of changed paths, one per line+ - changedCombined - list of files in the sync --combined report format+ - changedMaxDirs - do a full run if a partial one would list more directories than this++See the [index](/commands/rclone_index/) command for more information on the above.+### operations/list: List the given remote and path in JSON format {#operations-list}This takes the following parameters:@@ -2136,85 +2172,6 @@ And this sets NOTICE level logs (normal without -v)rclone rc options/set --json '{"main": {"LogLevel": "NOTICE"}}'-### pluginsctl/addPlugin: Add a plugin using url {#pluginsctl-addPlugin}--Used for adding a plugin to the webgui.--This takes the following parameters:--- url - http url of the github repo where the plugin is hosted (http://github.com/rclone/rclone-webui-react).--Example:-- rclone rc pluginsctl/addPlugin--### pluginsctl/getPluginsForType: Get plugins with type criteria {#pluginsctl-getPluginsForType}--This shows all possible plugins by a mime type.--This takes the following parameters:--- type - supported mime type by a loaded plugin e.g. (video/mp4, audio/mp3).-- pluginType - filter plugins based on their type e.g. (DASHBOARD, FILE_HANDLER, TERMINAL).--Returns:--- loadedPlugins - list of current production plugins.-- testPlugins - list of temporarily loaded development plugins, usually running on a different server.--Example:-- rclone rc pluginsctl/getPluginsForType type=video/mp4--### pluginsctl/listPlugins: Get the list of currently loaded plugins {#pluginsctl-listPlugins}--This allows you to get the currently enabled plugins and their details.--This takes no parameters and returns:--- loadedPlugins - list of current production plugins.-- testPlugins - list of temporarily loaded development plugins, usually running on a different server.--E.g.-- rclone rc pluginsctl/listPlugins--### pluginsctl/listTestPlugins: Show currently loaded test plugins {#pluginsctl-listTestPlugins}--Allows listing of test plugins with the rclone.test set to true in package.json of the plugin.--This takes no parameters and returns:--- loadedTestPlugins - list of currently available test plugins.--E.g.-- rclone rc pluginsctl/listTestPlugins--### pluginsctl/removePlugin: Remove a loaded plugin {#pluginsctl-removePlugin}--This allows you to remove a plugin using it's name.--This takes parameters:--- name - name of the plugin in the format `author`/`plugin_name`.--E.g.-- rclone rc pluginsctl/removePlugin name=rclone/video-plugin--### pluginsctl/removeTestPlugin: Remove a test plugin {#pluginsctl-removeTestPlugin}--This allows you to remove a plugin using it's name.--This takes the following parameters:--- name - name of the plugin in the format `author`/`plugin_name`.--Example:-- rclone rc pluginsctl/removeTestPlugin name=rclone/rclone-webui-react-### rc/error: This returns an error {#rc-error}This returns an error with the input as part of its error string.@@ -2389,6 +2346,8 @@ This takes the following parameters:- path1 (required) - (string) a remote directory string e.g. `drive:path1`- path2 (required) - (string) a remote directory string e.g. `drive:path2`- dryRun - (bool) dry-run mode+- maxDelete - (int) Safety check on maximum percentage of deleted files+allowed. If exceeded, the bisync run will abort. (default: 50%)- backupDir1 - (string) --backup-dir for Path1. Must be a non-overlapping path onthe same remote.- backupDir2 - (string) --backup-dir for Path2. Must be a non-overlapping path on@@ -2418,6 +2377,8 @@ unavailable. (warning: may be slow and use lots of data!)using with --verbose- ignoreListingChecksum - (bool) Do not use checksums for listings (add --ignore-checksum to additionally skip post-copy checksum checks)+- maxDeleteRenamesAware - (bool) Exclude tracked renames from the --max-delete+safety check (requires --track-renames)- maxLock - (Duration) Consider lock files older than this to be expired(default: 0 (never expire)) (minimum: 2m)- noCleanup - (bool) Retain working files (useful for troubleshooting and@@ -2451,7 +2412,27 @@ This takes the following parameters:- srcFs - a remote name string e.g. "drive:src" for the source- dstFs - a remote name string e.g. "drive:dst" for the destination- createEmptySrcDirs - create empty src directories on destination if set+- combined - make a combined report of changes (default false)+- missingOnSrc - report all files missing from the source (default false)+- missingOnDst - report all files missing from the destination (default false)+- match - report all matching files (default false)+- differ - report all non-matching files (default false)+- error - report all files with errors (hashing or reading) (default false)+- destAfter - report all files that exist on the destination post-copy (default false)++Returns:+- combined - array of strings of combined report of changes+- missingOnSrc - array of strings of all files missing from the source+- missingOnDst - array of strings of all files missing from the destination+- match - array of strings of all matching files+- differ - array of strings of all non-matching files+- error - array of strings of all files with errors (hashing or reading)+- destAfter - array of strings of all files that exist on the destination post-copy++Each report is only returned if its parameter is set to true. If the+operation fails the reports are only available if it was run with+`_async`, as part of the job output.See the [copy](/commands/rclone_copy/) command for more information on the above.@@ -2463,7 +2444,27 @@ This takes the following parameters:- dstFs - a remote name string e.g. "drive:dst" for the destination- createEmptySrcDirs - create empty src directories on destination if set- deleteEmptySrcDirs - delete empty src directories if set+- combined - make a combined report of changes (default false)+- missingOnSrc - report all files missing from the source (default false)+- missingOnDst - report all files missing from the destination (default false)+- match - report all matching files (default false)+- differ - report all non-matching files (default false)+- error - report all files with errors (hashing or reading) (default false)+- destAfter - report all files that exist on the destination post-move (default false)++Returns:++- combined - array of strings of combined report of changes+- missingOnSrc - array of strings of all files missing from the source+- missingOnDst - array of strings of all files missing from the destination+- match - array of strings of all matching files+- differ - array of strings of all non-matching files+- error - array of strings of all files with errors (hashing or reading)+- destAfter - array of strings of all files that exist on the destination post-move+Each report is only returned if its parameter is set to true. If the+operation fails the reports are only available if it was run with+`_async`, as part of the job output.See the [move](/commands/rclone_move/) command for more information on the above.@@ -2474,7 +2475,27 @@ This takes the following parameters:- srcFs - a remote name string e.g. "drive:src" for the source- dstFs - a remote name string e.g. "drive:dst" for the destination- createEmptySrcDirs - create empty src directories on destination if set+- combined - make a combined report of changes (default false)+- missingOnSrc - report all files missing from the source (default false)+- missingOnDst - report all files missing from the destination (default false)+- match - report all matching files (default false)+- differ - report all non-matching files (default false)+- error - report all files with errors (hashing or reading) (default false)+- destAfter - report all files that exist on the destination post-sync (default false)++Returns:++- combined - array of strings of combined report of changes+- missingOnSrc - array of strings of all files missing from the source+- missingOnDst - array of strings of all files missing from the destination+- match - array of strings of all matching files+- differ - array of strings of all non-matching files+- error - array of strings of all files with errors (hashing or reading)+- destAfter - array of strings of all files that exist on the destination post-sync+Each report is only returned if its parameter is set to true. If the+operation fails the reports are only available if it was run with+`_async`, as part of the job output.See the [sync](/commands/rclone_sync/) command for more information on the above.diff --git a/docs/content/s3.md b/docs/content/s3.mdindex e26320760..aa3b7c7b3 100644--- a/docs/content/s3.md+++ b/docs/content/s3.md@@ -975,7 +975,7 @@ mode from COMPLIANCE to GOVERNANCE while preserving the original retention date:<!-- autogenerated options start - DO NOT EDIT - instead edit fs.RegInfo in backend/s3/s3.go and run make backenddocs to verify --> <!-- markdownlint-disable-line line-length -->### Standard options-Here are the Standard options specific to s3 (Amazon S3 Compliant Storage Providers including AWS, Alibaba, ArvanCloud, BizflyCloud, Ceph, ChinaMobile, Cloudflare, Cubbit, DigitalOcean, Dreamhost, Exaba, Fastly, FileLu, FlashBlade, GCS, HCP, Hetzner, HuaweiOBS, IBMCOS, IDrive, ImpossibleCloud, Intercolo, IONOS, Leviia, Liara, Linode, LyveCloud, Magalu, Mega, Minio, Netease, Outscale, OVHcloud, Petabox, Qiniu, Rabata, RackCorp, Rclone, Scaleway, Scality, SeaweedFS, Selectel, Servercore, SpectraLogic, Storj, Synology, TencentCOS, US3, Wasabi, Zadara, Zata, ZeroServices, Other).+Here are the Standard options specific to s3 (Amazon S3 Compliant Storage Providers including AWS, Alibaba, ArvanCloud, Autonomys, BizflyCloud, Cellar, Ceph, ChinaMobile, Cloudflare, Cubbit, DigitalOcean, Dreamhost, Exaba, Fastly, FileLu, FlashBlade, GCS, HCP, Hetzner, HuaweiOBS, IBMCOS, IDrive, ImpossibleCloud, Infomaniak, Intercolo, IONOS, Leviia, Liara, Linode, LyveCloud, Magalu, Mega, Minio, Neon, Netease, Outscale, OVHcloud, Petabox, Qiniu, Rabata, RackCorp, Rclone, RelAix, Scaleway, Scality, SeaweedFS, Selectel, Servercore, SpectraLogic, Storj, Synology, TencentCOS, US3, Wasabi, Zadara, Zata, ZeroServices, Other).#### --s3-provider@@ -994,8 +994,12 @@ Properties:- Alibaba Cloud Object Storage System (OSS) formerly Aliyun- "ArvanCloud"- Arvan Cloud Object Storage (AOS)+ - "Autonomys"+ - Autonomys Auto Drive- "BizflyCloud"- Bizfly Cloud Simple Storage+ - "Cellar"+ - Clever Cloud Cellar Object Storage- "Ceph"- Ceph Object Storage- "ChinaMobile"@@ -1030,6 +1034,8 @@ Properties:- IDrive e2- "ImpossibleCloud"- Impossible Cloud Object Storage+ - "Infomaniak"+ - Infomaniak Swiss Backup- "Intercolo"- Intercolo Object Storage- "IONOS"@@ -1048,6 +1054,8 @@ Properties:- MEGA S4 Object Storage- "Minio"- Minio Object Storage+ - "Neon"+ - Neon Object Storage- "Netease"- Netease Object Storage (NOS)- "Outscale"@@ -1064,6 +1072,8 @@ Properties:- RackCorp Object Storage- "Rclone"- Rclone S3 Server+ - "RelAix"+ - RelAix.Storage- "Scaleway"- Scaleway Object Storage- "Scality"@@ -1149,7 +1159,7 @@ Properties:- Config: region- Env Var: RCLONE_S3_REGION-- Provider: AWS,BizflyCloud,Ceph,Cloudflare,Cubbit,DigitalOcean,Dreamhost,Exaba,Fastly,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IBMCOS,ImpossibleCloud,Intercolo,IONOS,Leviia,LyveCloud,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,Rabata,RackCorp,Scaleway,Scality,SeaweedFS,Selectel,Servercore,Synology,Wasabi,Zadara,Zata,ZeroServices,Other+- Provider: AWS,BizflyCloud,Ceph,Cloudflare,Cubbit,DigitalOcean,Dreamhost,Exaba,Fastly,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IBMCOS,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,LyveCloud,Minio,Neon,Netease,Outscale,OVHcloud,Petabox,Qiniu,Rabata,RackCorp,Scaleway,Scality,SeaweedFS,Selectel,Servercore,Synology,Wasabi,Zadara,Zata,ZeroServices,Other- Type: string- Required: false- Examples:@@ -1284,6 +1294,9 @@ Properties:- "eu-central"- EU Central- Provider: Fastly+ - "eu-central-1"+ - EU Central 1+ - Provider: Fastly- "eu-south-1"- EU South 1- Provider: Fastly@@ -1335,50 +1348,104 @@ Properties:- "nbg1"- Nuremberg- Provider: Hetzner+ - "af-north-1"+ - AF-Cairo+ - Provider: HuaweiOBS- "af-south-1"- AF-Johannesburg- Provider: HuaweiOBS+ - "ap-southeast-1"+ - CN-Hong Kong+ - Provider: HuaweiOBS- "ap-southeast-2"- AP-Bangkok- Provider: HuaweiOBS- "ap-southeast-3"- AP-Singapore- Provider: HuaweiOBS- - "cn-east-3"- - CN East-Shanghai1+ - "ap-southeast-4"+ - AP-Jakarta+ - Provider: HuaweiOBS+ - "ap-southeast-5"+ - AP-Manila- Provider: HuaweiOBS- "cn-east-2"- CN East-Shanghai2- Provider: HuaweiOBS+ - "cn-east-3"+ - CN East-Shanghai1+ - Provider: HuaweiOBS+ - "cn-east-4"+ - CN East-Qingdao+ - Provider: HuaweiOBS+ - "cn-east-5"+ - CN East-Suzhou+ - Provider: HuaweiOBS- "cn-north-1"- CN North-Beijing1- Provider: HuaweiOBS+ - "cn-north-2"+ - CN North-Beijing2+ - Provider: HuaweiOBS- "cn-north-4"- CN North-Beijing4- Provider: HuaweiOBS+ - "cn-north-9"+ - CN North-Ulanqab+ - Provider: HuaweiOBS+ - "cn-north-11"+ - CN North-Hebei+ - Provider: HuaweiOBS+ - "cn-north-12"+ - CN North-Hohhot+ - Provider: HuaweiOBS- "cn-south-1"- CN South-Guangzhou- Provider: HuaweiOBS- - "ap-southeast-1"- - CN-Hong Kong+ - "cn-south-2"+ - CN South-Shenzhen- Provider: HuaweiOBS- - "sa-argentina-1"- - LA-Buenos Aires1+ - "cn-south-4"+ - CN South-Guangzhou-InvitationOnly- Provider: HuaweiOBS- - "sa-peru-1"- - LA-Lima1+ - "cn-southwest-2"+ - CN Southwest-Guiyang1+ - Provider: HuaweiOBS+ - "cn-southwest-3"+ - CN Southwest-Chongqing+ - Provider: HuaweiOBS+ - "eu-west-101"+ - EU-Dublin+ - Provider: HuaweiOBS+ - "la-north-2"+ - LA-Mexico City2+ - Provider: HuaweiOBS+ - "la-south-2"+ - LA-Santiago+ - Provider: HuaweiOBS+ - "me-east-1"+ - ME-Riyadh- Provider: HuaweiOBS- "na-mexico-1"- LA-Mexico City1- Provider: HuaweiOBS- - "sa-chile-1"- - LA-Santiago2+ - "ru-northwest-2"+ - RU-Moscow2+ - Provider: HuaweiOBS+ - "sa-argentina-1"+ - LA-Buenos Aires1- Provider: HuaweiOBS- "sa-brazil-1"- LA-Sao Paulo1- Provider: HuaweiOBS- - "ru-northwest-2"- - RU-Moscow2+ - "sa-chile-1"+ - LA-Santiago2+ - Provider: HuaweiOBS+ - "sa-peru-1"+ - LA-Lima1+ - Provider: HuaweiOBS+ - "tr-west-1"+ - TR-Istanbul- Provider: HuaweiOBS- "eu-central-2"- Frankfurt, Germany@@ -1401,6 +1468,10 @@ Properties:- "us-east-1"- New York, USA- Provider: ImpossibleCloud+ - "us-east-1"+ - Default region for compatibility.+ - All data is stored in Switzerland.+ - Provider: Infomaniak- "de-fra"- Frankfurt, Germany- Provider: Intercolo@@ -1422,6 +1493,18 @@ Properties:- "us-central-1"- Lenexa, USA- Provider: IONOS+ - "us-east-1"+ - US East (N. Virginia)+ - Provider: Neon,Petabox,Rabata+ - "us-east-2"+ - US East (Ohio)+ - Provider: Neon+ - "eu-central-1"+ - Europe (Frankfurt)+ - Provider: Neon,Petabox+ - "ap-southeast-1"+ - Asia Pacific (Singapore)+ - Provider: Neon,Petabox- "eu-west-2"- Paris, France- Provider: Outscale@@ -1479,15 +1562,6 @@ Properties:- "rbx-archive"- Roubaix, France (Cold Archive)- Provider: OVHcloud- - "us-east-1"- - US East (N. Virginia)- - Provider: Petabox,Rabata- - "eu-central-1"- - Europe (Frankfurt)- - Provider: Petabox- - "ap-southeast-1"- - Asia Pacific (Singapore)- - Provider: Petabox- "me-south-1"- Middle East (Bahrain)- Provider: Petabox@@ -1595,6 +1669,9 @@ Properties:- "pl-waw"- Warsaw, Poland- Provider: Scaleway+ - "it-mil"+ - Milan, Italy+ - Provider: Scaleway- "ru-1"- St. Petersburg- Provider: Selectel,Servercore@@ -1661,7 +1738,7 @@ Properties:- Config: endpoint- Env Var: RCLONE_S3_ENDPOINT-- Provider: AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,ChinaMobile,Cloudflare,Cubbit,DigitalOcean,Dreamhost,Exaba,Fastly,FileLu,FlashBlade,GCS,HCP,Hetzner,HuaweiOBS,IBMCOS,ImpossibleCloud,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Mega,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,Rabata,RackCorp,Rclone,Scaleway,Scality,SeaweedFS,Selectel,Servercore,SpectraLogic,Storj,Synology,TencentCOS,US3,Wasabi,Zadara,Zata,ZeroServices,Other+- Provider: AWS,Alibaba,ArvanCloud,Autonomys,BizflyCloud,Cellar,Ceph,ChinaMobile,Cloudflare,Cubbit,DigitalOcean,Dreamhost,Exaba,Fastly,FileLu,FlashBlade,GCS,HCP,Hetzner,HuaweiOBS,IBMCOS,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Mega,Minio,Neon,Netease,Outscale,OVHcloud,Petabox,Qiniu,Rabata,RackCorp,Rclone,RelAix,Scaleway,Scality,SeaweedFS,Selectel,Servercore,SpectraLogic,Storj,Synology,TencentCOS,US3,Wasabi,Zadara,Zata,ZeroServices,Other- Type: string- Required: false- Examples:@@ -1747,12 +1824,18 @@ Properties:- "s3.ir-tbz-sh1.arvanstorage.ir"- Tabriz Iran (Shahriar)- Provider: ArvanCloud+ - "s3.auto-drive.autonomys.xyz"+ - Auto Drive mainnet S3 endpoint+ - Provider: Autonomys- "hn.ss.bfcplatform.vn"- Hanoi endpoint- Provider: BizflyCloud- "hcm.ss.bfcplatform.vn"- Ho Chi Minh endpoint- Provider: BizflyCloud+ - "cellar-c2.services.clever-cloud.com"+ - Paris, France+ - Provider: Cellar- "eos-wuxi-1.cmecloud.cn"- The default endpoint - a good choice if you are unsure.- East China (Suzhou)@@ -1889,6 +1972,9 @@ Properties:- "eu-central.object.fastlystorage.app"- EU Central- Provider: Fastly+ - "eu-central-1.object.fastlystorage.app"+ - EU Central 1+ - Provider: Fastly- "eu-south-1.object.fastlystorage.app"- EU South 1- Provider: Fastly@@ -1943,50 +2029,104 @@ Properties:- "nbg1.your-objectstorage.com"- Nuremberg- Provider: Hetzner+ - "obs.af-north-1.myhuaweicloud.com"+ - AF-Cairo+ - Provider: HuaweiOBS- "obs.af-south-1.myhuaweicloud.com"- AF-Johannesburg- Provider: HuaweiOBS+ - "obs.ap-southeast-1.myhuaweicloud.com"+ - CN-Hong Kong+ - Provider: HuaweiOBS- "obs.ap-southeast-2.myhuaweicloud.com"- AP-Bangkok- Provider: HuaweiOBS- "obs.ap-southeast-3.myhuaweicloud.com"- AP-Singapore- Provider: HuaweiOBS- - "obs.cn-east-3.myhuaweicloud.com"- - CN East-Shanghai1+ - "obs.ap-southeast-4.myhuaweicloud.com"+ - AP-Jakarta+ - Provider: HuaweiOBS+ - "obs.ap-southeast-5.myhuaweicloud.com"+ - AP-Manila- Provider: HuaweiOBS- "obs.cn-east-2.myhuaweicloud.com"- CN East-Shanghai2- Provider: HuaweiOBS+ - "obs.cn-east-3.myhuaweicloud.com"+ - CN East-Shanghai1+ - Provider: HuaweiOBS+ - "obs.cn-east-4.myhuaweicloud.com"+ - CN East-Qingdao+ - Provider: HuaweiOBS+ - "obs.cn-east-5.myhuaweicloud.com"+ - CN East-Suzhou+ - Provider: HuaweiOBS- "obs.cn-north-1.myhuaweicloud.com"- CN North-Beijing1- Provider: HuaweiOBS+ - "obs.cn-north-2.myhuaweicloud.com"+ - CN North-Beijing2+ - Provider: HuaweiOBS- "obs.cn-north-4.myhuaweicloud.com"- CN North-Beijing4- Provider: HuaweiOBS+ - "obs.cn-north-9.myhuaweicloud.com"+ - CN North-Ulanqab+ - Provider: HuaweiOBS+ - "obs.cn-north-11.myhuaweicloud.com"+ - CN North-Hebei+ - Provider: HuaweiOBS+ - "obs.cn-north-12.myhuaweicloud.com"+ - CN North-Hohhot+ - Provider: HuaweiOBS- "obs.cn-south-1.myhuaweicloud.com"- CN South-Guangzhou- Provider: HuaweiOBS- - "obs.ap-southeast-1.myhuaweicloud.com"- - CN-Hong Kong+ - "obs.cn-south-2.myhuaweicloud.com"+ - CN South-Shenzhen- Provider: HuaweiOBS- - "obs.sa-argentina-1.myhuaweicloud.com"- - LA-Buenos Aires1+ - "obs.cn-south-4.myhuaweicloud.com"+ - CN South-Guangzhou-InvitationOnly- Provider: HuaweiOBS- - "obs.sa-peru-1.myhuaweicloud.com"- - LA-Lima1+ - "obs.cn-southwest-2.myhuaweicloud.com"+ - CN Southwest-Guiyang1+ - Provider: HuaweiOBS+ - "obs.cn-southwest-3.myhuaweicloud.com"+ - CN Southwest-Chongqing+ - Provider: HuaweiOBS+ - "obs.eu-west-101.myhuaweicloud.eu"+ - EU-Dublin+ - Provider: HuaweiOBS+ - "obs.la-north-2.myhuaweicloud.com"+ - LA-Mexico City2+ - Provider: HuaweiOBS+ - "obs.la-south-2.myhuaweicloud.com"+ - LA-Santiago+ - Provider: HuaweiOBS+ - "obs.me-east-1.myhuaweicloud.com"+ - ME-Riyadh- Provider: HuaweiOBS- "obs.na-mexico-1.myhuaweicloud.com"- LA-Mexico City1- Provider: HuaweiOBS- - "obs.sa-chile-1.myhuaweicloud.com"- - LA-Santiago2+ - "obs.ru-northwest-2.myhuaweicloud.com"+ - RU-Moscow2+ - Provider: HuaweiOBS+ - "obs.sa-argentina-1.myhuaweicloud.com"+ - LA-Buenos Aires1- Provider: HuaweiOBS- "obs.sa-brazil-1.myhuaweicloud.com"- LA-Sao Paulo1- Provider: HuaweiOBS- - "obs.ru-northwest-2.myhuaweicloud.com"- - RU-Moscow2+ - "obs.sa-chile-1.myhuaweicloud.com"+ - LA-Santiago2+ - Provider: HuaweiOBS+ - "obs.sa-peru-1.myhuaweicloud.com"+ - LA-Lima1+ - Provider: HuaweiOBS+ - "obs.tr-west-1.myhuaweicloud.com"+ - TR-Istanbul- Provider: HuaweiOBS- "s3.us.cloud-object-storage.appdomain.cloud"- US Cross Region Endpoint@@ -2195,6 +2335,18 @@ Properties:- "us-east-1.storage.impossibleapi.net"- New York, USA- Provider: ImpossibleCloud+ - "s3.swiss-backup01.infomaniak.com"+ - Swiss Backup Location 01+ - Provider: Infomaniak+ - "s3.swiss-backup02.infomaniak.com"+ - Swiss Backup Location 02+ - Provider: Infomaniak+ - "s3.swiss-backup03.infomaniak.com"+ - Swiss Backup Location 03+ - Provider: Infomaniak+ - "s3.swiss-backup04.infomaniak.com"+ - Swiss Backup Location 04+ - Provider: Infomaniak- "de-fra.i3storage.com"- Frankfurt, Germany- Provider: Intercolo@@ -2299,38 +2451,47 @@ Properties:- "br-ne1.magaluobjects.com"- Fortaleza, CE (BR), br-ne1- Provider: Magalu- - "s3.eu-amsterdam.megas4.com"- - Mega S4 Amsterdam+ - "s3.eu-luxembourg-1.megas4.com"+ - Mega S4 Luxembourg 1+ - Provider: Mega+ - "s3.eu-luxembourg-2.megas4.com"+ - Mega S4 Luxembourg 2- Provider: Mega- - "s3.eu-luxembourg.megas4.com"- - Mega S4 Luxembourg+ - "s3.eu-amsterdam-1.megas4.com"+ - Mega S4 Amsterdam 1- Provider: Mega- - "s3.eu-paris.megas4.com"- - Mega S4 Paris+ - "s3.eu-amsterdam-2.megas4.com"+ - Mega S4 Amsterdam 2- Provider: Mega- - "s3.eu-barcelona.megas4.com"- - Mega S4 Barcelona+ - "s3.eu-paris-1.megas4.com"+ - Mega S4 Paris 1- Provider: Mega- - "s3.ca-montreal.megas4.com"- - Mega S4 Montreal+ - "s3.eu-paris-2.megas4.com"+ - Mega S4 Paris 2- Provider: Mega- - "s3.ca-vancouver.megas4.com"- - Mega S4 Vancouver+ - "s3.eu-barcelona-1.megas4.com"+ - Mega S4 Barcelona 1- Provider: Mega- - "s3.ap-tokyo.megas4.com"- - Mega S4 Tokyo+ - "s3.eu-barcelona-2.megas4.com"+ - Mega S4 Barcelona 2- Provider: Mega- - "s3.eu-central-1.s4.mega.io"- - Mega S4 eu-central-1 (Amsterdam, legacy)+ - "s3.ca-montreal-1.megas4.com"+ - Mega S4 Montreal 1- Provider: Mega- - "s3.eu-central-2.s4.mega.io"- - Mega S4 eu-central-2 (Bettembourg, legacy)+ - "s3.ca-montreal-2.megas4.com"+ - Mega S4 Montreal 2- Provider: Mega- - "s3.ca-central-1.s4.mega.io"- - Mega S4 ca-central-1 (Montreal, legacy)+ - "s3.ca-vancouver-1.megas4.com"+ - Mega S4 Vancouver 1- Provider: Mega- - "s3.ca-west-1.s4.mega.io"- - Mega S4 ca-west-1 (Vancouver, legacy)+ - "s3.ca-vancouver-2.megas4.com"+ - Mega S4 Vancouver 2+ - Provider: Mega+ - "s3.ap-tokyo-1.megas4.com"+ - Mega S4 Tokyo 1+ - Provider: Mega+ - "s3.ap-tokyo-2.megas4.com"+ - Mega S4 Tokyo 2- Provider: Mega- "oos.eu-west-2.outscale.com"- Outscale EU West 2 (Paris)@@ -2497,6 +2658,12 @@ Properties:- "nz.s3.rackcorp.com"- Auckland (New Zealand) Endpoint- Provider: RackCorp+ - "s3-gw1.relaix.net"+ - Aachen-1, Germany+ - Provider: RelAix+ - "s3-gw2.relaix.net"+ - Aachen-2, Germany+ - Provider: RelAix- "s3.nl-ams.scw.cloud"- Amsterdam Endpoint- Provider: Scaleway@@ -2506,6 +2673,9 @@ Properties:- "s3.pl-waw.scw.cloud"- Warsaw Endpoint- Provider: Scaleway+ - "s3.it-mil.scw.cloud"+ - Milan Endpoint+ - Provider: Scaleway- "localhost:8333"- SeaweedFS S3 localhost- Provider: SeaweedFS@@ -2742,7 +2912,7 @@ Properties:- Config: location_constraint- Env Var: RCLONE_S3_LOCATION_CONSTRAINT-- Provider: AWS,ArvanCloud,Ceph,ChinaMobile,DigitalOcean,Dreamhost,Exaba,GCS,HCP,Hetzner,IBMCOS,ImpossibleCloud,LyveCloud,Minio,Netease,Qiniu,Rabata,RackCorp,Scality,SeaweedFS,Synology,Wasabi,Zata,ZeroServices,Other+- Provider: AWS,ArvanCloud,Ceph,ChinaMobile,DigitalOcean,Dreamhost,Exaba,GCS,HCP,Hetzner,IBMCOS,ImpossibleCloud,Infomaniak,LyveCloud,Minio,Netease,Qiniu,Rabata,RackCorp,Scality,SeaweedFS,Synology,Wasabi,Zata,ZeroServices,Other- Type: string- Required: false- Examples:@@ -3123,36 +3293,36 @@ Properties:- Config: acl- Env Var: RCLONE_S3_ACL-- Provider: AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IBMCOS,IDrive,ImpossibleCloud,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,Scaleway,Scality,SeaweedFS,TencentCOS,US3,Wasabi,Zata,ZeroServices,Other+- Provider: AWS,Alibaba,ArvanCloud,BizflyCloud,Cellar,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IBMCOS,IDrive,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,RelAix,Scaleway,Scality,SeaweedFS,TencentCOS,US3,Wasabi,Zata,ZeroServices,Other- Type: string- Required: false- Examples:- "private"- Owner gets FULL_CONTROL.- No one else has access rights (default).- - Provider: AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,Scaleway,Scality,SeaweedFS,US3,Wasabi,Zata,ZeroServices,Other+ - Provider: AWS,Alibaba,ArvanCloud,BizflyCloud,Cellar,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,RelAix,Scaleway,Scality,SeaweedFS,US3,Wasabi,Zata,ZeroServices,Other- "public-read"- Owner gets FULL_CONTROL.- The AllUsers group gets READ access.- - Provider: AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,Scaleway,Scality,SeaweedFS,TencentCOS,US3,Wasabi,Zata,ZeroServices,Other+ - Provider: AWS,Alibaba,ArvanCloud,BizflyCloud,Cellar,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,RelAix,Scaleway,Scality,SeaweedFS,TencentCOS,US3,Wasabi,Zata,ZeroServices,Other- "public-read-write"- Owner gets FULL_CONTROL.- The AllUsers group gets READ and WRITE access.- Granting this on a bucket is generally not recommended.- - Provider: AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,Scaleway,Scality,SeaweedFS,TencentCOS,Wasabi,Zata,ZeroServices,Other+ - Provider: AWS,Alibaba,ArvanCloud,BizflyCloud,Cellar,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,RelAix,Scaleway,Scality,SeaweedFS,TencentCOS,Wasabi,Zata,ZeroServices,Other- "authenticated-read"- Owner gets FULL_CONTROL.- The AuthenticatedUsers group gets READ access.- - Provider: AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,Scaleway,Scality,SeaweedFS,TencentCOS,Wasabi,Zata,ZeroServices,Other+ - Provider: AWS,Alibaba,ArvanCloud,BizflyCloud,Cellar,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,RelAix,Scaleway,Scality,SeaweedFS,TencentCOS,Wasabi,Zata,ZeroServices,Other- "bucket-owner-read"- Object owner gets FULL_CONTROL.- Bucket owner gets READ access.- If you specify this canned ACL when creating a bucket, Amazon S3 ignores it.- - Provider: AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,Scaleway,Scality,SeaweedFS,TencentCOS,Wasabi,Zata,ZeroServices,Other+ - Provider: AWS,Alibaba,ArvanCloud,BizflyCloud,Cellar,Ceph,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,RelAix,Scaleway,Scality,SeaweedFS,TencentCOS,Wasabi,Zata,ZeroServices,Other- "bucket-owner-full-control"- Both the object owner and the bucket owner get FULL_CONTROL over the object.- If you specify this canned ACL when creating a bucket, Amazon S3 ignores it.- - Provider: AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,Scaleway,Scality,SeaweedFS,TencentCOS,Wasabi,Zata,ZeroServices,Other+ - Provider: AWS,Alibaba,ArvanCloud,BizflyCloud,Cellar,Ceph,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,RelAix,Scaleway,Scality,SeaweedFS,TencentCOS,Wasabi,Zata,ZeroServices,Other- "private"- Owner gets FULL_CONTROL.- No one else has access rights (default).@@ -3346,7 +3516,7 @@ Properties:### Advanced options-Here are the Advanced options specific to s3 (Amazon S3 Compliant Storage Providers including AWS, Alibaba, ArvanCloud, BizflyCloud, Ceph, ChinaMobile, Cloudflare, Cubbit, DigitalOcean, Dreamhost, Exaba, Fastly, FileLu, FlashBlade, GCS, HCP, Hetzner, HuaweiOBS, IBMCOS, IDrive, ImpossibleCloud, Intercolo, IONOS, Leviia, Liara, Linode, LyveCloud, Magalu, Mega, Minio, Netease, Outscale, OVHcloud, Petabox, Qiniu, Rabata, RackCorp, Rclone, Scaleway, Scality, SeaweedFS, Selectel, Servercore, SpectraLogic, Storj, Synology, TencentCOS, US3, Wasabi, Zadara, Zata, ZeroServices, Other).+Here are the Advanced options specific to s3 (Amazon S3 Compliant Storage Providers including AWS, Alibaba, ArvanCloud, Autonomys, BizflyCloud, Cellar, Ceph, ChinaMobile, Cloudflare, Cubbit, DigitalOcean, Dreamhost, Exaba, Fastly, FileLu, FlashBlade, GCS, HCP, Hetzner, HuaweiOBS, IBMCOS, IDrive, ImpossibleCloud, Infomaniak, Intercolo, IONOS, Leviia, Liara, Linode, LyveCloud, Magalu, Mega, Minio, Neon, Netease, Outscale, OVHcloud, Petabox, Qiniu, Rabata, RackCorp, Rclone, RelAix, Scaleway, Scality, SeaweedFS, Selectel, Servercore, SpectraLogic, Storj, Synology, TencentCOS, US3, Wasabi, Zadara, Zata, ZeroServices, Other).#### --s3-bucket-acl@@ -3365,7 +3535,7 @@ Properties:- Config: bucket_acl- Env Var: RCLONE_S3_BUCKET_ACL-- Provider: AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IBMCOS,IDrive,ImpossibleCloud,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Mega,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,Scaleway,SeaweedFS,Servercore,TencentCOS,US3,Wasabi,Zata,ZeroServices,Other+- Provider: AWS,Alibaba,ArvanCloud,BizflyCloud,Cellar,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IBMCOS,IDrive,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Mega,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,RelAix,Scaleway,SeaweedFS,Servercore,TencentCOS,US3,Wasabi,Zata,ZeroServices,Other- Type: string- Required: false- Examples:@@ -3403,7 +3573,7 @@ Properties:- Config: sse_customer_algorithm- Env Var: RCLONE_S3_SSE_CUSTOMER_ALGORITHM-- Provider: AWS,Ceph,ChinaMobile,Minio+- Provider: AWS,Cellar,Ceph,ChinaMobile,Minio- Type: string- Required: false- Examples:@@ -3416,13 +3586,13 @@ Properties:To use SSE-C you may provide the secret encryption key used to encrypt/decrypt your data.-Alternatively you can provide --sse-customer-key-base64.+Alternatively you can provide --s3-sse-customer-key-base64.Properties:- Config: sse_customer_key- Env Var: RCLONE_S3_SSE_CUSTOMER_KEY-- Provider: AWS,Ceph,ChinaMobile,Minio+- Provider: AWS,Cellar,Ceph,ChinaMobile,Minio- Type: string- Required: false- Examples:@@ -3433,13 +3603,13 @@ Properties:If using SSE-C you must provide the secret encryption key encoded in base64 format to encrypt/decrypt your data.-Alternatively you can provide --sse-customer-key.+Alternatively you can provide --s3-sse-customer-key.Properties:- Config: sse_customer_key_base64- Env Var: RCLONE_S3_SSE_CUSTOMER_KEY_BASE64-- Provider: AWS,Ceph,ChinaMobile,Minio+- Provider: AWS,Cellar,Ceph,ChinaMobile,Minio- Type: string- Required: false- Examples:@@ -3457,7 +3627,7 @@ Properties:- Config: sse_customer_key_md5- Env Var: RCLONE_S3_SSE_CUSTOMER_KEY_MD5-- Provider: AWS,Ceph,ChinaMobile,Minio+- Provider: AWS,Cellar,Ceph,ChinaMobile,Minio- Type: string- Required: false- Examples:@@ -4547,6 +4717,47 @@ info on how to pass options and arguments.These can be run on a running backend using the rc command[backend/command](/rc/#backend-command).+### link++Generate a signed link with response header overrides.++```console+rclone backend link remote: [options] [<arguments>+]+```++This command generates a signed download link for one file, with optional+overrides for the HTTP response headers. Pass the file path as a separate+argument, relative to the remote path.++Usage examples:++```console+rclone backend link s3:bucket path/to/file -o expire=1h+rclone backend link s3:bucket path/to/file -o response-expires="Thu, 01 Jan 1970 00:00:00 GMT"+rclone backend link s3:bucket path/to/file -o response-content-disposition='attachment; filename="download.txt"'+```++The link expires after 7 days by default. Use `-o expire=1h` to change+its lifetime. Durations must be at least 1 second; values above 7 days are+limited to 7 days, as with `rclone link`.++The `response-expires` option sets the HTTP Expires response header,+not the lifetime of the signed link. It must be an HTTP date, such as+`Thu, 01 Jan 1970 00:00:00 GMT`.++The overrides are included in the signature and must not be changed in the+returned URL. They do not modify the object's stored metadata.++Options:++- "expire": How long the link will be valid (default 7d, maximum 7d).+- "response-cache-control": Set the Cache-Control response header.+- "response-content-disposition": Set the Content-Disposition response header.+- "response-content-encoding": Set the Content-Encoding response header.+- "response-content-language": Set the Content-Language response header.+- "response-content-type": Set the Content-Type response header.+- "response-expires": Set the Expires response header to an HTTP date.+### restoreRestore objects from GLACIER or INTELLIGENT-TIERING archive tier.diff --git a/docs/content/sftp.md b/docs/content/sftp.mdindex 1fff363c7..52280c2b1 100644--- a/docs/content/sftp.md+++ b/docs/content/sftp.md@@ -920,7 +920,7 @@ Properties:- Config: encoding- Env Var: RCLONE_SFTP_ENCODING- Type: Encoding-- Default: Slash,Del,Ctl,Dot+- Default: Slash,Dot#### --sftp-set-modtime@@ -1169,6 +1169,34 @@ Properties:- Type: bool- Default: false+#### --sftp-multithread-upload++Set this to use multi-thread (multi-connection) uploads.++Normally rclone uploads a single large file over one SFTP connection.+With this set rclone uploads large files over several SFTP connections+at once (one per chunk), which greatly improves throughput on high+latency links, the same way multi-thread downloads already work. Each+connection writes a different, non-overlapping byte range of the file,+so this requires a server that allows several handles to the same file+with writes at arbitrary offsets, which OpenSSH does.++This is off by default because many SFTP servers only accept sequential+writes (for example some object storage backed SFTP gateways) and would+fail large uploads with "truncate failed" or "invalid offset" style+errors.++This is ignored (uploads stay single-threaded) when+--sftp-disable-concurrent-writes or --sftp-connections is in use.+++Properties:++- Config: multithread_upload+- Env Var: RCLONE_SFTP_MULTITHREAD_UPLOAD+- Type: bool+- Default: false+#### --sftp-idle-timeoutMax time before closing idle connections.diff --git a/docs/content/smb.md b/docs/content/smb.mdindex a95ca4d33..e76b267b5 100644--- a/docs/content/smb.md+++ b/docs/content/smb.md@@ -137,12 +137,14 @@ Properties:SMB username.+Leave blank to use the current user name.+Properties:- Config: user- Env Var: RCLONE_SMB_USER- Type: string-- Default: "$USER"+- Required: false#### --smb-port@@ -219,6 +221,23 @@ Properties:Here are the Advanced options specific to smb (SMB / CIFS).+#### --smb-workstation++Workstation name sent in NTLM authentication.++The NetBIOS name sent as the client workstation in the NTLM AUTHENTICATE+message. Some servers (e.g. domain-joined Windows servers with per-account+workstation restrictions or NTLM allow-lists) require this to match a+specific name. Leave blank to send no workstation name (the default).+++Properties:++- Config: workstation+- Env Var: RCLONE_SMB_WORKSTATION+- Type: string+- Required: false+#### --smb-idle-timeoutMax time before closing idle connections.diff --git a/docs/content/swift.md b/docs/content/swift.mdindex 1124424c7..e61837a34 100644--- a/docs/content/swift.md+++ b/docs/content/swift.md@@ -328,6 +328,14 @@ Properties:- OVH- "https://authenticate.ain.net"- Blomp Cloud Storage+ - "https://swiss-backup01.infomaniak.com/identity/v3"+ - Infomaniak Swiss Backup Location 01+ - "https://swiss-backup02.infomaniak.com/identity/v3"+ - Infomaniak Swiss Backup Location 02+ - "https://swiss-backup03.infomaniak.com/identity/v3"+ - Infomaniak Swiss Backup Location 03+ - "https://swiss-backup04.infomaniak.com/identity/v3"+ - Infomaniak Swiss Backup Location 04#### --swift-user-iddiff --git a/docs/content/webdav.md b/docs/content/webdav.mdindex 8e59bc952..72cfe5c94 100644--- a/docs/content/webdav.md+++ b/docs/content/webdav.md@@ -249,6 +249,13 @@ For example, to set a Cookie use 'Cookie,name=value', or '"Cookie","name=value"'You can set multiple headers, e.g. '"Cookie","name=value","Authorization","xxx"'.+The headers are only sent to the host in the configured URL. If the+server redirects to another host (including a subdomain or a different+port) the headers are not sent to it, or to any further hop in that+redirect chain. The exception is the directory listing and file+download requests when auth_redirect is set, as that keeps all+credentials across redirects.+Properties:diff --git a/docs/content/yandex.md b/docs/content/yandex.mdindex d1558dc28..5b248702c 100644--- a/docs/content/yandex.md+++ b/docs/content/yandex.md@@ -278,6 +278,26 @@ Properties:- Type: Duration- Default: 0s+#### --yandex-app-folder++Use the application folder as the root.++If you registered your own OAuth application with Yandex with only+the "Application Folder" permission (cloud_api:disk.app_folder)+instead of full disk access, then rclone needs to address paths with+the "app:/" prefix instead of the usual "disk:/" prefix, otherwise+all requests fail with a 403 Forbidden error.++Set this to true if your OAuth token only has access to the+application folder.++Properties:++- Config: app_folder+- Env Var: RCLONE_YANDEX_APP_FOLDER+- Type: bool+- Default: false+#### --yandex-descriptionDescription of the remote.diff --git a/lib/transform/transform.md b/lib/transform/transform.mdindex 5e6bab81f..573669bf0 100644--- a/lib/transform/transform.md+++ b/lib/transform/transform.md@@ -218,12 +218,12 @@ rclone convmv "stories/The Quick Brown Fox!.txt" --name-transform "all,command=e```consolerclone convmv "stories/The Quick Brown Fox!" --name-transform "date=-{YYYYMMDD}"-// Output: stories/The Quick Brown Fox!-20260731+// Output: stories/The Quick Brown Fox!-20261106``````consolerclone convmv "stories/The Quick Brown Fox!" --name-transform "date=-{macfriendlytime}"-// Output: stories/The Quick Brown Fox!-2026-07-31 0455PM+// Output: stories/The Quick Brown Fox!-2026-11-06 0235PM``````consolediff --git a/rclone.1 b/rclone.1index 38b3c5526..557777f8f 100644--- a/rclone.1+++ b/rclone.1@@ -15,7 +15,7 @@. ftr VB CB. ftr VBI CBI.\}-.TH "rclone" "1" "Jul 31, 2026" "User Manual" ""+.TH "rclone" "1" "Nov 06, 2026" "User Manual" "".hy.SH NAME.PP@@ -54,6 +54,7 @@ Available commands:gui Open the web based GUI.hashsum Produces a hashsum file for all the objects in the path.help Show help for rclone commands, flags and backends.+ index Write static directory listings into a remote.link Generate public link to file/folder.listremotes List all the remotes in the config file and defined in environment variables.ls List the objects in the path with size and path.@@ -240,6 +241,8 @@ Alibaba Cloud (Aliyun) Object Storage System (OSS).IP \[bu] 2Amazon S3.IP \[bu] 2+Autonomys Auto Drive+.IP \[bu] 2Bizfly Cloud Simple Storage.IP \[bu] 2Backblaze B2@@ -254,6 +257,8 @@ Arvan Cloud Object Storage (AOS).IP \[bu] 2Citrix ShareFile.IP \[bu] 2+Clever Cloud Cellar+.IP \[bu] 2Cloudflare R2.IP \[bu] 2Cloudinary@@ -264,6 +269,8 @@ DigitalOcean Spaces.IP \[bu] 2Digi Storage.IP \[bu] 2+dosya.dev+.IP \[bu] 2Dreamhost.IP \[bu] 2Drime@@ -330,10 +337,16 @@ IDrive e2.IP \[bu] 2Impossible Cloud.IP \[bu] 2+Infomaniak Swiss Backup (S3)+.IP \[bu] 2+Infomaniak Swiss Backup (Swift)+.IP \[bu] 2Intercolo Object Storage.IP \[bu] 2IONOS Cloud.IP \[bu] 2+kDrive+.IP \[bu] 2Koofr.IP \[bu] 2Leviia Object Storage@@ -364,6 +377,8 @@ Microsoft OneDrive.IP \[bu] 2Minio.IP \[bu] 2+Neon Object Storage+.IP \[bu] 2Nextcloud.IP \[bu] 2Blomp Cloud Storage@@ -412,6 +427,8 @@ Rackspace Cloud Files.IP \[bu] 2Rclone Serve S3.IP \[bu] 2+RelAix.Storage+.IP \[bu] 2rsync.net.IP \[bu] 2Scaleway@@ -814,12 +831,6 @@ status (https://repology.org/badge/vertical-allrepos/rclone.svg?columns=3)] (httThe rclone developers maintain a docker image forrclone (https://hub.docker.com/r/rclone/rclone)..PP-\f[B]Note:\f[R] We also now offer a paid version of rclone with-enterprise-grade security and zero CVEs through our partner-SecureBuild (https://securebuild.com/blog/introducing-securebuild).-If you are interested, check out their website and the Rclone-SecureBuild Image (https://securebuild.com/images/rclone).-.PPThese images are built as part of the release process based on a minimalAlpine Linux..PP@@ -961,7 +972,7 @@ Its current version is as below..SS Source installation.PPMake sure you have git and Go (https://golang.org/) installed.-Go version 1.25 or newer is required, the latest release is recommended.+Go version 1.26 or newer is required, the latest release is recommended.You can get it from your package manager, or download it fromgolang.org/dl (https://golang.org/dl/).Then you can run the following:@@ -1399,6 +1410,8 @@ DigitalOcean Spaces (https://rclone.org/s3/#digitalocean-spaces).IP \[bu] 2Digi Storage (https://rclone.org/koofr/#digi-storage).IP \[bu] 2+dosya.dev (https://rclone.org/dosya/)+.IP \[bu] 2Drime (https://rclone.org/drime/).IP \[bu] 2Dropbox (https://rclone.org/dropbox/)@@ -1442,6 +1455,8 @@ Internxt (https://rclone.org/internxt/).IP \[bu] 2Jottacloud (https://rclone.org/jottacloud/).IP \[bu] 2+kDrive (https://rclone.org/kdrive/)+.IP \[bu] 2Koofr (https://rclone.org/koofr/).IP \[bu] 2Linkbox (https://rclone.org/linkbox/)@@ -2901,6 +2916,11 @@ Note that \f[V]ls\f[R] and \f[V]lsl\f[R] recurse by default - useThe other list commands \f[V]lsd\f[R],\f[V]lsf\f[R],\f[V]lsjson\f[R] donot recurse by default - use \f[V]-R\f[R] to make them recurse..PP+On the bucket-based remotes (e.g.+s3, gcs, b2) a directory is a key prefix rather than an object, so a+listing without \f[V]-R\f[R] shows each prefix as one directory entry+and not the objects stored under it.+.PPList commands prefer a recursive method that uses more memory but fewertransactions by default.Use \f[V]--disable ListR\f[R] to suppress the behavior.@@ -2983,9 +3003,9 @@ Does not recurse by default.Use the \f[V]-R\f[R] flag to recurse..PPThis command lists the total size of the directory (if known, -1 if-not), the modification time (if known, the current time if not), the-number of objects in the directory (if known, -1 if not) and the name of-the directory, E.g.+not), the modification time (if known, the \f[V]--default-time\f[R] if+not), the number of objects in the directory (if known, -1 if not) and+the name of the directory, E.g..IP.nf\f[C]@@ -3034,6 +3054,11 @@ Note that \f[V]ls\f[R] and \f[V]lsl\f[R] recurse by default - useThe other list commands \f[V]lsd\f[R],\f[V]lsf\f[R],\f[V]lsjson\f[R] donot recurse by default - use \f[V]-R\f[R] to make them recurse..PP+On the bucket-based remotes (e.g.+s3, gcs, b2) a directory is a key prefix rather than an object, so a+listing without \f[V]-R\f[R] shows each prefix as one directory entry+and not the objects stored under it.+.PPList commands prefer a recursive method that uses more memory but fewertransactions by default.Use \f[V]--disable ListR\f[R] to suppress the behavior.@@ -3153,6 +3178,11 @@ Note that \f[V]ls\f[R] and \f[V]lsl\f[R] recurse by default - useThe other list commands \f[V]lsd\f[R],\f[V]lsf\f[R],\f[V]lsjson\f[R] donot recurse by default - use \f[V]-R\f[R] to make them recurse..PP+On the bucket-based remotes (e.g.+s3, gcs, b2) a directory is a key prefix rather than an object, so a+listing without \f[V]-R\f[R] shows each prefix as one directory entry+and not the objects stored under it.+.PPList commands prefer a recursive method that uses more memory but fewertransactions by default.Use \f[V]--disable ListR\f[R] to suppress the behavior.@@ -4462,6 +4492,7 @@ rclone bisync remote1:path1 remote2:path2 [flags]--force Bypass --max-delete safety check and run the sync. Consider using with --verbose-h, --help help for bisync--ignore-listing-checksum Do not use checksums for listings (add --ignore-checksum to additionally skip post-copy checksum checks)+ --max-delete-renames-aware Exclude tracked renames from the --max-delete safety check (requires --track-renames)--max-lock Duration Consider lock files older than this to be expired (default: 0 (never expire)) (minimum: 2m) (default 0s)--no-cleanup Retain working files (useful for troubleshooting and testing).--no-slow-hash Ignore listing checksums only on backends where they are slow@@ -4853,6 +4884,8 @@ Output completion script for a given shell..PPGenerates a shell completion script for rclone.Run with \f[V]--help\f[R] to list the supported shells.+Enum-valued flags such as \f[V]--log-level\f[R] and+\f[V]--vfs-cache-mode\f[R] also complete their supported values..SS Options.IP.nf@@ -6393,14 +6426,14 @@ rclone convmv \[dq]stories/The Quick Brown Fox!.txt\[dq] --name-transform \[dq]a.nf\f[C]rclone convmv \[dq]stories/The Quick Brown Fox!\[dq] --name-transform \[dq]date=-{YYYYMMDD}\[dq]-// Output: stories/The Quick Brown Fox!-20260731+// Output: stories/The Quick Brown Fox!-20261106\f[R].fi.IP.nf\f[C]rclone convmv \[dq]stories/The Quick Brown Fox!\[dq] --name-transform \[dq]date=-{macfriendlytime}\[dq]-// Output: stories/The Quick Brown Fox!-2026-07-31 0340PM+// Output: stories/The Quick Brown Fox!-2026-11-06 0220PM\f[R].fi.IP@@ -7213,7 +7246,10 @@ Remove a single file from remote.Remove a single file from remote.Unlike \f[V]delete\f[R] it cannot be used to remove a directory and itdoesn\[aq]t obey include/exclude filters - if the specified file exists,-it will always be removed.+it will always be removed from its original location.+.PP+The \f[V]--backup-dir\f[R] and \f[V]--suffix\f[R] flags are honored when+set..IP.nf\f[C]@@ -7231,6 +7267,27 @@ rclone deletefile remote:path [flags]Options shared with other commands are described next.See the global flags page (https://rclone.org/flags/) for global optionsnot listed here.+.SS Sync Options+.PP+Flags used for sync commands+.IP+.nf+\f[C]+ --backup-dir string Make backups into hierarchy based in DIR+ --delete-after When synchronizing, delete files on destination after transferring (default)+ --delete-before When synchronizing, delete files on destination before transferring+ --delete-during When synchronizing, delete files during transfer+ --fix-case Force rename of case insensitive dest to match source+ --ignore-errors Delete even if there are I/O errors+ --list-cutoff int To save memory, sort directory listings on disk above this threshold (default 1000000)+ --max-delete int When synchronizing, limit the number of deletes (default -1)+ --max-delete-size SizeSuffix When synchronizing, limit the total size of deletes (default off)+ --suffix string Suffix to add to changed files+ --suffix-keep-extension Preserve the extension when using --suffix+ --track-renames When synchronizing, track file renames and do a server-side move if possible+ --track-renames-strategy string Strategies to use when synchronizing using track-renames hash|modtime|leaf (default \[dq]hash\[dq])+\f[R]+.fi.SS Important Options.PPImportant flags useful for most commands@@ -7518,11 +7575,6 @@ Flags to control the Remote Control API--rc-template string User-specified template--rc-user string User name for authentication--rc-user-from-header string User name from a defined HTTP header- --rc-web-fetch-url string URL to fetch the releases for webgui (default \[dq]https://api.github.com/repos/rclone/rclone-webui-react/releases/latest\[dq])- --rc-web-gui Launch WebGUI on localhost- --rc-web-gui-force-update Force update to latest version of web gui- --rc-web-gui-no-open-browser Don\[aq]t open the browser automatically- --rc-web-gui-update Check and update to latest version of web gui\f[R].fi.SS See Also@@ -7567,6 +7619,7 @@ Supported hashes are:- blake3- xxh3- xxh128+- crc32c\f[R].fi.PP@@ -7647,6 +7700,421 @@ Flags for listing directories.IP \[bu] 2rclone (https://rclone.org/commands/rclone/) - Show help for rclonecommands, flags and backends.+.SH rclone index+.PP+Write static directory listings into a remote.+.SS Synopsis+.PP+Write a directory listing, \f[V]index.html\f[R] by default, into every+directory of \f[V]remote:path\f[R] so that it can be browsed when it is+served as a static website, for example from an S3 website endpoint, a+bucket behind a CDN, or a plain web server.+.PP+This works like \f[V]rclone sync\f[R]: each run makes the minimum+changes needed to bring the listings into line with the files.+Listings which haven\[aq]t changed aren\[aq]t rewritten, so a run with+no changes makes no uploads.+The listings never show the listing files themselves.+.PP+\f[B]Note\f[R] that \f[V]rclone index\f[R] treats every file with an+output name (\f[V]index.html\f[R] by default) in the directories it+indexes as its own, and will overwrite it or delete it without checking+where it came from.+To protect a hand-written index, exclude its directory with+\f[V]--index-exclude\f[R].+.SS Outputs+.PP+\f[V]--output NAME=FORMAT\f[R] says what to write in each directory and+can be repeated.+NAME is the file name, and FORMAT is one of the built-in formats below+or the path of a Go template file.+.PP+.TS+tab(@);+lw(19.3n) lw(26.6n) lw(24.1n).+T{+Format+T}@T{+MIME type+T}@T{+Contents+T}+_+T{+\f[V]html\f[R]+T}@T{+\f[V]text/html; charset=utf-8\f[R]+T}@T{+The HTML listing from the \f[V]rclone serve http\f[R] template+T}+T{+\f[V]json\f[R]+T}@T{+\f[V]application/json\f[R]+T}@T{+JSON in the format of \f[V]rclone lsjson\f[R] for the directory+T}+T{+\f[V]caddy\f[R]+T}@T{+\f[V]application/json\f[R]+T}@T{+JSON in the format of Caddy\[aq]s \f[V]file_server browse\f[R]+T}+.TE+.PP+The default is \f[V]--output index.html=html\f[R].+The built-in formats are templates, so they can be printed with+\f[V]--print-template FORMAT\f[R], copied and changed.+Outputs whose name ends in \f[V].html\f[R] or \f[V].htm\f[R] are+rendered with Go\[aq]s \f[V]html/template\f[R] and everything else with+\f[V]text/template\f[R], which has a \f[V]json\f[R] function for safe+encoding.+The template data is that of \f[V]rclone serve http\f[R] with+\f[V].Static\f[R] set (see rclone serve+http (https://rclone.org/commands/rclone_serve_http/#template)).+Outputs are uploaded with the MIME type of the built-in format, or that+of the output name\[aq]s extension for template outputs, since static+hosts serve the stored type.+.SS Filters+.PP+The normal filters (\f[V]--include\f[R], \f[V]--exclude\f[R],+\f[V]--filter-from\f[R], \f[V]--max-age\f[R], ...)+control what appears in the listings.+An excluded file isn\[aq]t listed, and an excluded directory isn\[aq]t+listed or indexed.+.PP+The \f[V]--index-include\f[R], \f[V]--index-exclude\f[R],+\f[V]--index-filter\f[R] flags and their \f[V]-from\f[R] variants+control which directories get listings.+They take the same patterns as the normal filters, applied to directory+paths, and \f[V]--index-include\f[R] implies excluding all other+directories.+A directory excluded here still appears in its parent\[aq]s listing, but+nothing is written or deleted inside it, so it can have its own+hand-made index.+\f[V]--index-max-depth\f[R] limits how deep listings are written.+.SS Modification times+.PP+Listings show the modification time of each file.+When using the s3, oracleobjectstorage and swift backends rclone stores+the modification time as metadata which isn\[aq]t returned by the bucket+listing, so \f[B]every object needs a HEAD request\f[R], which can mean+thousands of transactions per run.+Avoid this with one of:+.IP \[bu] 2+\f[V]--use-server-modtime\f[R] uses the time the object was uploaded+instead, which is what the web server sends as \f[V]Last-Modified\f[R].+Recommended for s3 unless the original modification times matter.+.IP \[bu] 2+\f[V]--no-modtime\f[R] shows no times at all, which is the cheapest+option.+.PP+\f[V]--dir-time\f[R] sets the time shown for directories:+.IP \[bu] 2+\f[V]newest\f[R] (default): the newest modification time of anything+below the directory.+Note that this means one new file changes the listing of every directory+above it.+.IP \[bu] 2+\f[V]dir\f[R]: the directory\[aq]s own modification time.+On most file systems writing a listing into a directory changes its+time, so this can take more than one run to settle.+.IP \[bu] 2+\f[V]none\f[R]: no time.+.PP+The listing files are given the newest modification time below their+directory, so the \f[V]Last-Modified\f[R] header of a listing is+meaningful.+.SS Links+.PP+Directory links are \f[V]dir/\f[R] by default, which works on hosts that+serve index documents: S3 website endpoints, GCS and Azure static+websites, nginx and Apache, and R2 custom domains with a URL rewrite+rule.+For hosts that don\[aq]t (plain S3 REST URLs, B2 friendly URLs) use+\f[V]--link-index\f[R] to make directory links \f[V]dir/index.html\f[R]+instead.+.SS Partial runs+.PP+After a sync which changed a known set of files there is no need to walk+the whole remote.+Tell \f[V]rclone index\f[R] what changed and it re-indexes only the+directories that could be affected:+.IP+.nf+\f[C]+rclone index r2:bucket --changed v1.76.0/ --changed version.txt+rclone index r2:bucket --changed-from changes.txt+rclone sync ./public r2:bucket --exclude index.html --combined - | rclone index r2:bucket --changed-combined -+\f[R]+.fi+.IP \[bu] 2+\f[V]--changed PATH\f[R] names a changed file or directory relative to+\f[V]remote:path\f[R] and can be repeated.+A directory, given with a trailing slash or found to be one, is+re-indexed completely.+.IP \[bu] 2+\f[V]--changed-from FILE\f[R] reads one path per line exactly as+written.+.IP \[bu] 2+\f[V]--changed-combined FILE\f[R] reads the \f[V]--combined\f[R] report+of \f[V]rclone sync\f[R]: lines starting with \f[V]+\f[R], \f[V]-\f[R],+\f[V]*\f[R] and \f[V]!\f[R] are changes and \f[V]=\f[R] lines are+ignored.+.IP \[bu] 2+Both files accept \f[V]-\f[R] for standard input and all three flags can+be combined.+.PP+A changed file re-indexes its directory and every directory above it,+since their times and their listings can change too, and on bucket based+storage a directory can cease to exist.+Each of those is one directory listing.+With \f[V]--dir-time newest\f[R] the directories above also need the+times of their unchanged subdirectories, which are read from the+modification times of those subdirectories\[aq] listings at the cost of+one small request each.+With \f[V]--use-server-modtime\f[R] that is the time the listing was+written rather than the time of the newest file, so partial runs can+drift by the indexing delay until the next full run.+\f[V]--dir-time dir\f[R] and \f[V]none\f[R] need no such lookups.+.PP+Directories which are neither named nor above a named path are left+alone, as are directories deeper than \f[V]--max-depth\f[R].+An empty change list does nothing, and naming the root does a full run.+There is no automatic cutover to a full run since that depends on the+size of the whole remote: a partial run costs one listing per affected+directory, a full run one request per thousand objects on S3-like+storage or one per directory elsewhere.+The number of directories a partial run will list is logged with+\f[V]-v\f[R], and \f[V]--changed-max-dirs\f[R] makes it fall back to a+full run above that number.+A good pattern is a partial run after each upload and a scheduled full+run to catch anything missed.+.SS Site icon+.PP+The listings carry no icon or branding of their own.+To give a site an icon, upload a \f[V]favicon.ico\f[R] to the root of+the remote, which browsers fetch from there for every page.+Exclude it from the listings with \f[V]--exclude /favicon.ico\f[R] so it+stays on the site but isn\[aq]t shown.+.SS Deleting listings+.PP+On backends which can\[aq]t have empty directories (S3, B2 and other+bucket-based storage) a directory only exists while it has files in it.+Once the last file in a directory has been removed its listing is+deleted, since the directory only existed because of it.+Backends which can have empty directories keep a listing for each empty+directory.+.SS Using with sync+.PP+A sync from a local build deletes files on the destination which+aren\[aq]t in the source, which includes the listings.+Either index the local build before syncing:+.IP+.nf+\f[C]+rclone index ./public+rclone sync ./public r2:bucket+\f[R]+.fi+.PP+or exclude the listings from the sync so it neither uploads nor deletes+them, then index the remote:+.IP+.nf+\f[C]+rclone sync ./public r2:bucket --exclude index.html+rclone index r2:bucket+\f[R]+.fi+.SS Examples+.PP+An S3 website bucket, using the upload times so no object needs a HEAD+request:+.IP+.nf+\f[C]+rclone index s3:my-site --use-server-modtime+\f[R]+.fi+.PP+HTML and rclone JSON listings, setting \f[V]Cache-Control\f[R] on them:+.IP+.nf+\f[C]+rclone index s3:bucket --output index.html=html --output index.json=json --header-upload \[dq]Cache-Control: max-age=300\[dq]+\f[R]+.fi+.PP+Only list release files, and don\[aq]t write listings under+\f[V]/private\f[R]:+.IP+.nf+\f[C]+rclone index r2:bucket --include \[dq]*.{zip,deb,rpm}\[dq] --index-exclude \[dq]/private/**\[dq]+\f[R]+.fi+.PP+A B2 bucket served from its friendly URLs:+.IP+.nf+\f[C]+rclone index b2:bucket --link-index+\f[R]+.fi+.PP+See what would change:+.IP+.nf+\f[C]+rclone index r2:bucket --dry-run -v+\f[R]+.fi+.PP+Listings which haven\[aq]t changed aren\[aq]t normally rewritten, so+changing the headers or metadata set on them needs+\f[V]--index-rewrite\f[R], which writes every listing a run would render+whether it changed or not:+.IP+.nf+\f[C]+rclone index r2:bucket --index-rewrite --header-upload \[dq]Cache-Control: max-age=3600\[dq]+\f[R]+.fi+.IP+.nf+\f[C]+rclone index remote:path [flags]+\f[R]+.fi+.SS Options+.IP+.nf+\f[C]+ --changed stringArray Only re-index directories affected by this changed path (directories end in /)+ --changed-combined stringArray Read changed paths from a sync --combined report (use - to read from stdin)+ --changed-from stringArray Read changed paths from file, one per line (use - to read from stdin)+ --changed-max-dirs int Do a full index if a partial one would list more than this many directories (0 for no limit)+ --dir-time newest|dir|none How to work out the time shown for a directory (default newest)+ -h, --help help for index+ --index-exclude stringArray Don\[aq]t write listings in directories matching pattern+ --index-exclude-from stringArray Read directory exclude patterns from file (use - to read from stdin)+ --index-filter stringArray Add a rule for which directories get listings+ --index-filter-from stringArray Read directory rules from a file (use - to read from stdin)+ --index-include stringArray Only write listings in directories matching pattern+ --index-include-from stringArray Read directory include patterns from file (use - to read from stdin)+ --index-max-depth int Only write listings this many directories deep (-1 for no limit) (default -1)+ --index-rewrite Write every listing even if it is unchanged+ --link-index Link to dir/NAME rather than dir/ for hosts without index documents+ --no-modtime Don\[aq]t show modification times in listings+ --output stringArray Listing to write in each directory as NAME=FORMAT (default [index.html=html])+ --print-template string Print the built-in template for FORMAT and exit+\f[R]+.fi+.PP+Options shared with other commands are described next.+See the global flags page (https://rclone.org/flags/) for global options+not listed here.+.SS Copy Options+.PP+Flags for anything which can copy a file+.IP+.nf+\f[C]+ --check-first Do all the checks before starting transfers+ -c, --checksum Check for changes with size & checksum (if available, or fallback to size only)+ --compare-dest stringArray Include additional server-side paths during comparison+ --copy-dest stringArray Implies --compare-dest but also copies files from paths into destination+ --cutoff-mode HARD|SOFT|CAUTIOUS Mode to stop transfers when reaching the max transfer limit HARD|SOFT|CAUTIOUS (default HARD)+ --ignore-case-sync Ignore case when synchronizing+ --ignore-checksum Skip post copy check of checksums+ --ignore-existing Skip all files that exist on destination+ --ignore-size Ignore size when skipping use modtime or checksum+ -I, --ignore-times Don\[aq]t skip items that match size and time - transfer all unconditionally+ --immutable Do not modify files, fail if existing files have been modified+ --inplace Download directly to destination file instead of atomic download to temp/rename+ -l, --links Translate symlinks to/from regular files with a \[aq].rclonelink\[aq] extension+ --max-backlog int Maximum number of objects in sync or check backlog (default 10000)+ --max-duration Duration Maximum duration rclone will transfer data for (default 0s)+ --max-transfer SizeSuffix Maximum size of data to transfer (default off)+ -M, --metadata If set, preserve metadata when copying objects+ --modify-window Duration Max time diff to be considered the same (default 1ns)+ --multi-thread-chunk-size SizeSuffix Chunk size for multi-thread downloads / uploads, if not set by filesystem (default 64Mi)+ --multi-thread-cutoff SizeSuffix Use multi-thread downloads for files above this size (default 256Mi)+ --multi-thread-streams int Number of streams to use for multi-thread downloads (default 4)+ --multi-thread-write-buffer-size SizeSuffix In memory buffer size for writing when in multi-thread mode (default 128Ki)+ --name-transform stringArray Transform paths during the copy process+ --no-check-dest Don\[aq]t check the destination, copy regardless+ --no-traverse Don\[aq]t traverse destination file system on copy+ --no-update-dir-modtime Don\[aq]t update directory modification times+ --no-update-modtime Don\[aq]t update destination modtime if files identical+ --order-by string Instructions on how to order the transfers, e.g. \[aq]size,descending\[aq]+ --partial-suffix string Add partial-suffix to temporary file name when --inplace is not used (default \[dq].partial\[dq])+ --refresh-times Refresh the modtime of remote files+ --server-side-across-configs Allow server-side operations (e.g. copy) to work across different configs+ --size-only Skip based on size only, not modtime or checksum+ --streaming-upload-cutoff SizeSuffix Cutoff for switching to chunked upload if file size is unknown, upload starts after reaching cutoff or when file ends (default 100Ki)+ -u, --update Skip files that are newer on the destination+\f[R]+.fi+.SS Important Options+.PP+Important flags useful for most commands+.IP+.nf+\f[C]+ -n, --dry-run Do a trial run with no permanent changes+ -i, --interactive Enable interactive mode+ -v, --verbose count Print lots more stuff (repeat for more)+\f[R]+.fi+.SS Filter Options+.PP+Flags for filtering directory listings+.IP+.nf+\f[C]+ --delete-excluded Delete files on dest excluded from sync+ --exclude stringArray Exclude files matching pattern+ --exclude-from stringArray Read file exclude patterns from file (use - to read from stdin)+ --exclude-if-present stringArray Exclude directories if filename is present+ --files-from stringArray Read list of source-file names from file (use - to read from stdin)+ --files-from-raw stringArray Read list of source-file names from file without any processing of lines (use - to read from stdin)+ --files-from0 stringArray Read list of source-file names from file using NUL as separator (use - to read from stdin)+ -f, --filter stringArray Add a file filtering rule+ --filter-from stringArray Read file filtering patterns from a file (use - to read from stdin)+ --hash-filter string Partition filenames by hash k/n or randomly \[at]/n+ --ignore-case Ignore case in filters (case insensitive)+ --include stringArray Include files matching pattern+ --include-from stringArray Read file include patterns from file (use - to read from stdin)+ --max-age Duration Only transfer files younger than this in s or suffix ms|s|m|h|d|w|M|y (default off)+ --max-depth int If set limits the recursion depth to this (default -1)+ --max-size SizeSuffix Only transfer files smaller than this in KiB or suffix B|K|M|G|T|P (default off)+ --metadata-exclude stringArray Exclude metadatas matching pattern+ --metadata-exclude-from stringArray Read metadata exclude patterns from file (use - to read from stdin)+ --metadata-filter stringArray Add a metadata filtering rule+ --metadata-filter-from stringArray Read metadata filtering patterns from a file (use - to read from stdin)+ --metadata-include stringArray Include metadatas matching pattern+ --metadata-include-from stringArray Read metadata include patterns from file (use - to read from stdin)+ --min-age Duration Only transfer files older than this in s or suffix ms|s|m|h|d|w|M|y (default off)+ --min-size SizeSuffix Only transfer files bigger than this in KiB or suffix B|K|M|G|T|P (default off)+\f[R]+.fi+.SS Listing Options+.PP+Flags for listing directories+.IP+.nf+\f[C]+ --default-time Time Time to show if modtime is unknown for files and directories (default 2000-01-01T00:00:00Z)+ --fast-list Use recursive list if available; uses more memory but fewer transactions+\f[R]+.fi+.SS See Also+.IP \[bu] 2+rclone (https://rclone.org/commands/rclone/) - Show help for rclone+commands, flags and backends..SH rclone link.PPGenerate public link to file/folder.@@ -7942,6 +8410,11 @@ Note that \f[V]ls\f[R] and \f[V]lsl\f[R] recurse by default - useThe other list commands \f[V]lsd\f[R],\f[V]lsf\f[R],\f[V]lsjson\f[R] donot recurse by default - use \f[V]-R\f[R] to make them recurse..PP+On the bucket-based remotes (e.g.+s3, gcs, b2) a directory is a key prefix rather than an object, so a+listing without \f[V]-R\f[R] shows each prefix as one directory entry+and not the objects stored under it.+.PPList commands prefer a recursive method that uses more memory but fewertransactions by default.Use \f[V]--disable ListR\f[R] to suppress the behavior.@@ -8155,6 +8628,11 @@ Note that \f[V]ls\f[R] and \f[V]lsl\f[R] recurse by default - useThe other list commands \f[V]lsd\f[R],\f[V]lsf\f[R],\f[V]lsjson\f[R] donot recurse by default - use \f[V]-R\f[R] to make them recurse..PP+On the bucket-based remotes (e.g.+s3, gcs, b2) a directory is a key prefix rather than an object, so a+listing without \f[V]-R\f[R] shows each prefix as one directory entry+and not the objects stored under it.+.PPList commands prefer a recursive method that uses more memory but fewertransactions by default.Use \f[V]--disable ListR\f[R] to suppress the behavior.@@ -8592,6 +9070,13 @@ For details, seevfs-case-sensitivity (https://rclone.org/commands/rclone_mount/#vfs-case-sensitivity)..SS NFS mount.PP+For macOS (and other platforms where this path is supported), prefer the+dedicated rclone nfsmount (https://rclone.org/commands/rclone_nfsmount/)+command.+It starts the NFS server and performs the mount for you.+\f[V]rclone mount\f[R] itself still uses FUSE (macFUSE/FUSE-T) and does+not switch to NFS via a flag.+.PPThis method spins up an NFS server using servenfs (https://rclone.org/commands/rclone_serve_nfs/) command and mountsit to the specified mountpoint.@@ -8993,8 +9478,8 @@ at all times.The buffered data is bound to one open file and won\[aq]t be shared..PPThis flag is a upper limit for the used memory per open file.-The buffer will only use memory for data that is downloaded but not not-yet read.+The buffer will only use memory for data that is downloaded but not yet+read.If the buffer is empty, only a small amount of memory will be used..PPThe maximum memory used by rclone for buffering can be up to@@ -9056,7 +9541,8 @@ This cache flushing strategy is efficient and more relevant files arelikely to remain cached..PPThe \f[V]--vfs-cache-max-age\f[R] will evict files from the cache after-the set time since last access has passed.+the set time since last access has passed; it is based on access time,+not on when the file was first added to the cache.The default value of 1 hour will start evicting files from cache thathaven\[aq]t been accessed for 1 hour.When a cached file is accessed the 1 hour timer is reset to 0 and will@@ -9229,8 +9715,8 @@ Rclone reads \f[V]--vfs-read-chunk-streams\f[R] chunks of size\f[V]--vfs-read-chunk-size\f[R] concurrently.The size for each read will stay constant..PP-This improves performance performance massively on high latency links or-very high bandwidth links to high performance object stores.+This improves performance massively on high latency links or very high+bandwidth links to high performance object stores..PPSome experimentation will be needed to find the optimum values of\f[V]--vfs-read-chunk-size\f[R] and \f[V]--vfs-read-chunk-streams\f[R]@@ -10319,6 +10805,13 @@ For details, seevfs-case-sensitivity (https://rclone.org/commands/rclone_mount/#vfs-case-sensitivity)..SS NFS mount.PP+For macOS (and other platforms where this path is supported), prefer the+dedicated rclone nfsmount (https://rclone.org/commands/rclone_nfsmount/)+command.+It starts the NFS server and performs the mount for you.+\f[V]rclone mount\f[R] itself still uses FUSE (macFUSE/FUSE-T) and does+not switch to NFS via a flag.+.PPThis method spins up an NFS server using servenfs (https://rclone.org/commands/rclone_serve_nfs/) command and mountsit to the specified mountpoint.@@ -10721,8 +11214,8 @@ at all times.The buffered data is bound to one open file and won\[aq]t be shared..PPThis flag is a upper limit for the used memory per open file.-The buffer will only use memory for data that is downloaded but not not-yet read.+The buffer will only use memory for data that is downloaded but not yet+read.If the buffer is empty, only a small amount of memory will be used..PPThe maximum memory used by rclone for buffering can be up to@@ -10784,7 +11277,8 @@ This cache flushing strategy is efficient and more relevant files arelikely to remain cached..PPThe \f[V]--vfs-cache-max-age\f[R] will evict files from the cache after-the set time since last access has passed.+the set time since last access has passed; it is based on access time,+not on when the file was first added to the cache.The default value of 1 hour will start evicting files from cache thathaven\[aq]t been accessed for 1 hour.When a cached file is accessed the 1 hour timer is reset to 0 and will@@ -10957,8 +11451,8 @@ Rclone reads \f[V]--vfs-read-chunk-streams\f[R] chunks of size\f[V]--vfs-read-chunk-size\f[R] concurrently.The size for each read will stay constant..PP-This improves performance performance massively on high latency links or-very high bandwidth links to high performance object stores.+This improves performance massively on high latency links or very high+bandwidth links to high performance object stores..PPSome experimentation will be needed to find the optimum values of\f[V]--vfs-read-chunk-size\f[R] and \f[V]--vfs-read-chunk-streams\f[R]@@ -11622,6 +12116,15 @@ responses, will overriding existing values.The flag may be repeated to add multiple headers.Use the format \f[V]Header-Name: value\f[R]..PP+\f[V]--rc-allow-origin\f[R] enables cross-origin (CORS) requests from+browsers by setting the \f[V]Access-Control-Allow-Origin\f[R] header on+every response to the origin given, e.g.+\f[V]--rc-allow-origin https://app.example.com\f[R].+Browser CORS preflight OPTIONS requests are answered by rclone without+needing authentication as browsers never send credentials with them.+Setting this to \f[V]*\f[R] allows any origin which is a security risk+on an authenticated server.+.PP\f[V]--rc-baseurl\f[R] controls the URL prefix that rclone serves from.By default rclone will serve from the root.If you used \f[V]--rc-baseurl \[dq]/rclone\[dq]\f[R] then rclone would@@ -11744,6 +12247,51 @@ T}@T{Currently unused.T}T{+\&.Path+T}@T{+T}@T{+The path of the directory from the root with leading and trailing+slashes, e.g.+\[aq]/dir/\[aq].+T}+T{+\&.IsRoot+T}@T{+T}@T{+Boolean for if this is the root directory.+T}+T{+\&.UpLink+T}@T{+T}@T{+The link to the parent directory.+T}+T{+\&.Static+T}@T{+T}@T{+Boolean for if the listing is being written as a static page by rclone+index.+T}+T{+\&.NumDirs+T}@T{+T}@T{+The number of directories in the listing.+T}+T{+\&.NumFiles+T}@T{+T}@T{+The number of files in the listing.+T}+T{+\&.TotalSize+T}@T{+T}@T{+The total size in bytes of the files in the listing.+T}+T{\&.BreadcrumbT}@T{T}@T{@@ -11797,6 +12345,13 @@ T}@T{T}@T{The UTC timestamp of an entry.T}+T{+T}@T{+\&.MimeType+T}@T{+The MIME type of an entry guessed from its name, or+\[aq]inode/directory\[aq].+T}.TE.PPThe server also makes the following functions available so that they can@@ -11870,8 +12425,21 @@ htpasswd -B htpasswd anotherUser\f[R].fi.PP+Use the same hash format and cost for every entry in the file.+If the entries differ, the time taken to reject a wrong password can+reveal which usernames exist.+.PPThe password file can be updated while rclone is running..PP+When authentication is enabled every request must be authenticated,+including HTTP OPTIONS requests, as the answer to an OPTIONS request can+reveal whether a path exists.+Health checks or monitoring which use unauthenticated OPTIONS requests+will need to send credentials.+The exception is browser CORS preflight requests when+\f[V]--rc-allow-origin\f[R] is set, which are answered without+authentication as browsers never send credentials with them.+.PPUse \f[V]--rc-realm\f[R] to set the authentication realm..PPUse \f[V]--rc-salt\f[R] to change the password hashing salt from the@@ -11925,11 +12493,6 @@ Flags to control the Remote Control API--rc-template string User-specified template--rc-user string User name for authentication--rc-user-from-header string User name from a defined HTTP header- --rc-web-fetch-url string URL to fetch the releases for webgui (default \[dq]https://api.github.com/repos/rclone/rclone-webui-react/releases/latest\[dq])- --rc-web-gui Launch WebGUI on localhost- --rc-web-gui-force-update Force update to latest version of web gui- --rc-web-gui-no-open-browser Don\[aq]t open the browser automatically- --rc-web-gui-update Check and update to latest version of web gui\f[R].fi.SS See Also@@ -12263,8 +12826,8 @@ at all times.The buffered data is bound to one open file and won\[aq]t be shared..PPThis flag is a upper limit for the used memory per open file.-The buffer will only use memory for data that is downloaded but not not-yet read.+The buffer will only use memory for data that is downloaded but not yet+read.If the buffer is empty, only a small amount of memory will be used..PPThe maximum memory used by rclone for buffering can be up to@@ -12326,7 +12889,8 @@ This cache flushing strategy is efficient and more relevant files arelikely to remain cached..PPThe \f[V]--vfs-cache-max-age\f[R] will evict files from the cache after-the set time since last access has passed.+the set time since last access has passed; it is based on access time,+not on when the file was first added to the cache.The default value of 1 hour will start evicting files from cache thathaven\[aq]t been accessed for 1 hour.When a cached file is accessed the 1 hour timer is reset to 0 and will@@ -12499,8 +13063,8 @@ Rclone reads \f[V]--vfs-read-chunk-streams\f[R] chunks of size\f[V]--vfs-read-chunk-size\f[R] concurrently.The size for each read will stay constant..PP-This improves performance performance massively on high latency links or-very high bandwidth links to high performance object stores.+This improves performance massively on high latency links or very high+bandwidth links to high performance object stores..PPSome experimentation will be needed to find the optimum values of\f[V]--vfs-read-chunk-size\f[R] and \f[V]--vfs-read-chunk-streams\f[R]@@ -13028,8 +13592,8 @@ at all times.The buffered data is bound to one open file and won\[aq]t be shared..PPThis flag is a upper limit for the used memory per open file.-The buffer will only use memory for data that is downloaded but not not-yet read.+The buffer will only use memory for data that is downloaded but not yet+read.If the buffer is empty, only a small amount of memory will be used..PPThe maximum memory used by rclone for buffering can be up to@@ -13091,7 +13655,8 @@ This cache flushing strategy is efficient and more relevant files arelikely to remain cached..PPThe \f[V]--vfs-cache-max-age\f[R] will evict files from the cache after-the set time since last access has passed.+the set time since last access has passed; it is based on access time,+not on when the file was first added to the cache.The default value of 1 hour will start evicting files from cache thathaven\[aq]t been accessed for 1 hour.When a cached file is accessed the 1 hour timer is reset to 0 and will@@ -13264,8 +13829,8 @@ Rclone reads \f[V]--vfs-read-chunk-streams\f[R] chunks of size\f[V]--vfs-read-chunk-size\f[R] concurrently.The size for each read will stay constant..PP-This improves performance performance massively on high latency links or-very high bandwidth links to high performance object stores.+This improves performance massively on high latency links or very high+bandwidth links to high performance object stores..PPSome experimentation will be needed to find the optimum values of\f[V]--vfs-read-chunk-size\f[R] and \f[V]--vfs-read-chunk-streams\f[R]@@ -13726,8 +14291,8 @@ at all times.The buffered data is bound to one open file and won\[aq]t be shared..PPThis flag is a upper limit for the used memory per open file.-The buffer will only use memory for data that is downloaded but not not-yet read.+The buffer will only use memory for data that is downloaded but not yet+read.If the buffer is empty, only a small amount of memory will be used..PPThe maximum memory used by rclone for buffering can be up to@@ -13789,7 +14354,8 @@ This cache flushing strategy is efficient and more relevant files arelikely to remain cached..PPThe \f[V]--vfs-cache-max-age\f[R] will evict files from the cache after-the set time since last access has passed.+the set time since last access has passed; it is based on access time,+not on when the file was first added to the cache.The default value of 1 hour will start evicting files from cache thathaven\[aq]t been accessed for 1 hour.When a cached file is accessed the 1 hour timer is reset to 0 and will@@ -13962,8 +14528,8 @@ Rclone reads \f[V]--vfs-read-chunk-streams\f[R] chunks of size\f[V]--vfs-read-chunk-size\f[R] concurrently.The size for each read will stay constant..PP-This improves performance performance massively on high latency links or-very high bandwidth links to high performance object stores.+This improves performance massively on high latency links or very high+bandwidth links to high performance object stores..PPSome experimentation will be needed to find the optimum values of\f[V]--vfs-read-chunk-size\f[R] and \f[V]--vfs-read-chunk-streams\f[R]@@ -14243,9 +14809,12 @@ This config generated must have this extra parameter.IP \[bu] 2\f[V]_root\f[R] - root to use for the backend.PP-And it may have this parameter+And it may have these parameters.IP \[bu] 2\f[V]_obscure\f[R] - comma separated strings for parameters to obscure+.IP \[bu] 2+\f[V]_secret_access_key\f[R] - the secret for S3 access key auth (see+below).PPIf password authentication was used by the client, input to the proxyprocess (on STDIN) would look similar to this:@@ -14254,7 +14823,8 @@ process (on STDIN) would look similar to this:\f[C]{\[dq]user\[dq]: \[dq]me\[dq],- \[dq]pass\[dq]: \[dq]mypassword\[dq]+ \[dq]pass\[dq]: \[dq]mypassword\[dq],+ \[dq]client_ip\[dq]: \[dq]192.168.1.1\[dq]}\f[R].fi@@ -14266,11 +14836,51 @@ process (on STDIN) would look similar to this:\f[C]{\[dq]user\[dq]: \[dq]me\[dq],- \[dq]public_key\[dq]: \[dq]AAAAB3NzaC1yc2EAAAADAQABAAABAQDuwESFdAe14hVS6omeyX7edc...JQdf\[dq]+ \[dq]public_key\[dq]: \[dq]AAAAB3NzaC1yc2EAAAADAQABAAABAQDuwESFdAe14hVS6omeyX7edc...JQdf\[dq],+ \[dq]client_ip\[dq]: \[dq]192.168.1.1\[dq]+}+\f[R]+.fi+.PP+If the client authenticated with an S3 access key+(\f[V]rclone serve s3\f[R]), the client never sends its secret, only a+signature made with it, so the input contains just the access key ID as+the \f[V]user\f[R] with no \f[V]pass\f[R] or \f[V]public_key\f[R]:+.IP+.nf+\f[C]+{+ \[dq]user\[dq]: \[dq]AKIAIOSFODNN7EXAMPLE\[dq],+ \[dq]client_ip\[dq]: \[dq]192.168.1.1\[dq]}\f[R].fi.PP+In this case the program must look up the secret access key for that+access key ID and return it in the \f[V]_secret_access_key\f[R] field of+the output.+Rclone then uses that secret to verify the signature on the request,+refusing the request if it does not match.+This means the proxy program is the source of truth for both the+credentials and the backend they map to.+If the program does not return \f[V]_secret_access_key\f[R] or returns+it empty the request is refused.+.PP+The program\[aq]s answer for an access key ID is cached (see below) but+is checked with the program again after 5 minutes even if the access key+ID is in constant use, so revoking an access key ID in the program takes+effect within 5 minutes.+A rotated secret takes effect on the first request signed with it.+.PP+The \f[V]client_ip\f[R] key holds the IP address the client connected+from, without a port number.+It can be used to restrict logins to certain networks, or to log+authentication attempts centrally.+It is omitted if the client has no IP address, for example when+connecting over a unix socket.+Note that if rclone is behind a reverse proxy this will be the address+of the reverse proxy and not the original client.+.PPAnd as an example return this on STDOUT.IP.nf@@ -14301,12 +14911,21 @@ the \f[V]user\f[R] be \f[V]user\[at]example.com\f[R] and then set theFor security you\[aq]d probably want to restrict the \f[V]host\f[R] to alimited list..PP-An internal cache of backends is keyed on the \f[V]user\f[R] and a hash-of the \f[V]pass\f[R] or \f[V]public_key\f[R].-This means that if a user\[aq]s password or public-key changes, or the-proxy returns different config parameters (eg a rotated-\f[V]api_key\f[R]), a fresh backend will be created on the next request-rather than the cached one being reused.+An internal cache of backends is keyed on the \f[V]user\f[R], the kind+of credential presented, a hash of the \f[V]pass\f[R] or+\f[V]public_key\f[R], and the \f[V]client_ip\f[R].+This means that if a user\[aq]s password or public-key changes, the+client connects from a new IP address, or the proxy returns different+config parameters (eg a rotated \f[V]api_key\f[R]), a fresh backend will+be created on the next request rather than the cached one being reused.+A backend created for a public key login is never reused for a password+login, so a public key cannot be sent as a password to skip the program.+.PP+A cached answer is checked with the program again after 5 minutes even+if the user is in constant use, so revoking a user in the program takes+effect within 5 minutes.+For \f[V]serve sftp\f[R] this applies to new logins - an existing+connection is not closed..PPThis can be used to build general purpose proxies to any kind of backendthat rclone supports.@@ -14453,6 +15072,15 @@ responses, will overriding existing values.The flag may be repeated to add multiple headers.Use the format \f[V]Header-Name: value\f[R]..PP+\f[V]--allow-origin\f[R] enables cross-origin (CORS) requests from+browsers by setting the \f[V]Access-Control-Allow-Origin\f[R] header on+every response to the origin given, e.g.+\f[V]--allow-origin https://app.example.com\f[R].+Browser CORS preflight OPTIONS requests are answered by rclone without+needing authentication as browsers never send credentials with them.+Setting this to \f[V]*\f[R] allows any origin which is a security risk+on an authenticated server.+.PP\f[V]--baseurl\f[R] controls the URL prefix that rclone serves from.By default rclone will serve from the root.If you used \f[V]--baseurl \[dq]/rclone\[dq]\f[R] then rclone would@@ -14573,6 +15201,51 @@ T}@T{Currently unused.T}T{+\&.Path+T}@T{+T}@T{+The path of the directory from the root with leading and trailing+slashes, e.g.+\[aq]/dir/\[aq].+T}+T{+\&.IsRoot+T}@T{+T}@T{+Boolean for if this is the root directory.+T}+T{+\&.UpLink+T}@T{+T}@T{+The link to the parent directory.+T}+T{+\&.Static+T}@T{+T}@T{+Boolean for if the listing is being written as a static page by rclone+index.+T}+T{+\&.NumDirs+T}@T{+T}@T{+The number of directories in the listing.+T}+T{+\&.NumFiles+T}@T{+T}@T{+The number of files in the listing.+T}+T{+\&.TotalSize+T}@T{+T}@T{+The total size in bytes of the files in the listing.+T}+T{\&.BreadcrumbT}@T{T}@T{@@ -14626,6 +15299,13 @@ T}@T{T}@T{The UTC timestamp of an entry.T}+T{+T}@T{+\&.MimeType+T}@T{+The MIME type of an entry guessed from its name, or+\[aq]inode/directory\[aq].+T}.TE.PPThe server also makes the following functions available so that they can@@ -14698,8 +15378,21 @@ htpasswd -B htpasswd anotherUser\f[R].fi.PP+Use the same hash format and cost for every entry in the file.+If the entries differ, the time taken to reject a wrong password can+reveal which usernames exist.+.PPThe password file can be updated while rclone is running..PP+When authentication is enabled every request must be authenticated,+including HTTP OPTIONS requests, as the answer to an OPTIONS request can+reveal whether a path exists.+Health checks or monitoring which use unauthenticated OPTIONS requests+will need to send credentials.+The exception is browser CORS preflight requests when+\f[V]--allow-origin\f[R] is set, which are answered without+authentication as browsers never send credentials with them.+.PPUse \f[V]--realm\f[R] to set the authentication realm..PPUse \f[V]--salt\f[R] to change the password hashing salt from the@@ -14777,8 +15470,8 @@ at all times.The buffered data is bound to one open file and won\[aq]t be shared..PPThis flag is a upper limit for the used memory per open file.-The buffer will only use memory for data that is downloaded but not not-yet read.+The buffer will only use memory for data that is downloaded but not yet+read.If the buffer is empty, only a small amount of memory will be used..PPThe maximum memory used by rclone for buffering can be up to@@ -14840,7 +15533,8 @@ This cache flushing strategy is efficient and more relevant files arelikely to remain cached..PPThe \f[V]--vfs-cache-max-age\f[R] will evict files from the cache after-the set time since last access has passed.+the set time since last access has passed; it is based on access time,+not on when the file was first added to the cache.The default value of 1 hour will start evicting files from cache thathaven\[aq]t been accessed for 1 hour.When a cached file is accessed the 1 hour timer is reset to 0 and will@@ -15013,8 +15707,8 @@ Rclone reads \f[V]--vfs-read-chunk-streams\f[R] chunks of size\f[V]--vfs-read-chunk-size\f[R] concurrently.The size for each read will stay constant..PP-This improves performance performance massively on high latency links or-very high bandwidth links to high performance object stores.+This improves performance massively on high latency links or very high+bandwidth links to high performance object stores..PPSome experimentation will be needed to find the optimum values of\f[V]--vfs-read-chunk-size\f[R] and \f[V]--vfs-read-chunk-streams\f[R]@@ -15294,9 +15988,12 @@ This config generated must have this extra parameter.IP \[bu] 2\f[V]_root\f[R] - root to use for the backend.PP-And it may have this parameter+And it may have these parameters.IP \[bu] 2\f[V]_obscure\f[R] - comma separated strings for parameters to obscure+.IP \[bu] 2+\f[V]_secret_access_key\f[R] - the secret for S3 access key auth (see+below).PPIf password authentication was used by the client, input to the proxyprocess (on STDIN) would look similar to this:@@ -15305,7 +16002,8 @@ process (on STDIN) would look similar to this:\f[C]{\[dq]user\[dq]: \[dq]me\[dq],- \[dq]pass\[dq]: \[dq]mypassword\[dq]+ \[dq]pass\[dq]: \[dq]mypassword\[dq],+ \[dq]client_ip\[dq]: \[dq]192.168.1.1\[dq]}\f[R].fi@@ -15317,11 +16015,51 @@ process (on STDIN) would look similar to this:\f[C]{\[dq]user\[dq]: \[dq]me\[dq],- \[dq]public_key\[dq]: \[dq]AAAAB3NzaC1yc2EAAAADAQABAAABAQDuwESFdAe14hVS6omeyX7edc...JQdf\[dq]+ \[dq]public_key\[dq]: \[dq]AAAAB3NzaC1yc2EAAAADAQABAAABAQDuwESFdAe14hVS6omeyX7edc...JQdf\[dq],+ \[dq]client_ip\[dq]: \[dq]192.168.1.1\[dq]}\f[R].fi.PP+If the client authenticated with an S3 access key+(\f[V]rclone serve s3\f[R]), the client never sends its secret, only a+signature made with it, so the input contains just the access key ID as+the \f[V]user\f[R] with no \f[V]pass\f[R] or \f[V]public_key\f[R]:+.IP+.nf+\f[C]+{+ \[dq]user\[dq]: \[dq]AKIAIOSFODNN7EXAMPLE\[dq],+ \[dq]client_ip\[dq]: \[dq]192.168.1.1\[dq]+}+\f[R]+.fi+.PP+In this case the program must look up the secret access key for that+access key ID and return it in the \f[V]_secret_access_key\f[R] field of+the output.+Rclone then uses that secret to verify the signature on the request,+refusing the request if it does not match.+This means the proxy program is the source of truth for both the+credentials and the backend they map to.+If the program does not return \f[V]_secret_access_key\f[R] or returns+it empty the request is refused.+.PP+The program\[aq]s answer for an access key ID is cached (see below) but+is checked with the program again after 5 minutes even if the access key+ID is in constant use, so revoking an access key ID in the program takes+effect within 5 minutes.+A rotated secret takes effect on the first request signed with it.+.PP+The \f[V]client_ip\f[R] key holds the IP address the client connected+from, without a port number.+It can be used to restrict logins to certain networks, or to log+authentication attempts centrally.+It is omitted if the client has no IP address, for example when+connecting over a unix socket.+Note that if rclone is behind a reverse proxy this will be the address+of the reverse proxy and not the original client.+.PPAnd as an example return this on STDOUT.IP.nf@@ -15352,12 +16090,21 @@ the \f[V]user\f[R] be \f[V]user\[at]example.com\f[R] and then set theFor security you\[aq]d probably want to restrict the \f[V]host\f[R] to alimited list..PP-An internal cache of backends is keyed on the \f[V]user\f[R] and a hash-of the \f[V]pass\f[R] or \f[V]public_key\f[R].-This means that if a user\[aq]s password or public-key changes, or the-proxy returns different config parameters (eg a rotated-\f[V]api_key\f[R]), a fresh backend will be created on the next request-rather than the cached one being reused.+An internal cache of backends is keyed on the \f[V]user\f[R], the kind+of credential presented, a hash of the \f[V]pass\f[R] or+\f[V]public_key\f[R], and the \f[V]client_ip\f[R].+This means that if a user\[aq]s password or public-key changes, the+client connects from a new IP address, or the proxy returns different+config parameters (eg a rotated \f[V]api_key\f[R]), a fresh backend will+be created on the next request rather than the cached one being reused.+A backend created for a public key login is never reused for a password+login, so a public key cannot be sent as a password to skip the program.+.PP+A cached answer is checked with the program again after 5 minutes even+if the user is in constant use, so revoking a user in the program takes+effect within 5 minutes.+For \f[V]serve sftp\f[R] this applies to new logins - an existing+connection is not closed..PPThis can be used to build general purpose proxies to any kind of backendthat rclone supports.@@ -15655,8 +16402,8 @@ at all times.The buffered data is bound to one open file and won\[aq]t be shared..PPThis flag is a upper limit for the used memory per open file.-The buffer will only use memory for data that is downloaded but not not-yet read.+The buffer will only use memory for data that is downloaded but not yet+read.If the buffer is empty, only a small amount of memory will be used..PPThe maximum memory used by rclone for buffering can be up to@@ -15718,7 +16465,8 @@ This cache flushing strategy is efficient and more relevant files arelikely to remain cached..PPThe \f[V]--vfs-cache-max-age\f[R] will evict files from the cache after-the set time since last access has passed.+the set time since last access has passed; it is based on access time,+not on when the file was first added to the cache.The default value of 1 hour will start evicting files from cache thathaven\[aq]t been accessed for 1 hour.When a cached file is accessed the 1 hour timer is reset to 0 and will@@ -15891,8 +16639,8 @@ Rclone reads \f[V]--vfs-read-chunk-streams\f[R] chunks of size\f[V]--vfs-read-chunk-size\f[R] concurrently.The size for each read will stay constant..PP-This improves performance performance massively on high latency links or-very high bandwidth links to high performance object stores.+This improves performance massively on high latency links or very high+bandwidth links to high performance object stores..PPSome experimentation will be needed to find the optimum values of\f[V]--vfs-read-chunk-size\f[R] and \f[V]--vfs-read-chunk-streams\f[R]@@ -16364,6 +17112,15 @@ responses, will overriding existing values.The flag may be repeated to add multiple headers.Use the format \f[V]Header-Name: value\f[R]..PP+\f[V]--allow-origin\f[R] enables cross-origin (CORS) requests from+browsers by setting the \f[V]Access-Control-Allow-Origin\f[R] header on+every response to the origin given, e.g.+\f[V]--allow-origin https://app.example.com\f[R].+Browser CORS preflight OPTIONS requests are answered by rclone without+needing authentication as browsers never send credentials with them.+Setting this to \f[V]*\f[R] allows any origin which is a security risk+on an authenticated server.+.PP\f[V]--baseurl\f[R] controls the URL prefix that rclone serves from.By default rclone will serve from the root.If you used \f[V]--baseurl \[dq]/rclone\[dq]\f[R] then rclone would@@ -16466,8 +17223,21 @@ htpasswd -B htpasswd anotherUser\f[R].fi.PP+Use the same hash format and cost for every entry in the file.+If the entries differ, the time taken to reject a wrong password can+reveal which usernames exist.+.PPThe password file can be updated while rclone is running..PP+When authentication is enabled every request must be authenticated,+including HTTP OPTIONS requests, as the answer to an OPTIONS request can+reveal whether a path exists.+Health checks or monitoring which use unauthenticated OPTIONS requests+will need to send credentials.+The exception is browser CORS preflight requests when+\f[V]--allow-origin\f[R] is set, which are answered without+authentication as browsers never send credentials with them.+.PPUse \f[V]--realm\f[R] to set the authentication realm..PPUse \f[V]--salt\f[R] to change the password hashing salt from the@@ -16536,6 +17306,24 @@ docs (https://docs.aws.amazon.com/general/latest/gr/signature-version-4.html)).If \f[V]--auth-key\f[R] is not provided then \f[V]serve s3\f[R] willallow anonymous access..PP+Alternatively \f[V]--auth-proxy\f[R] can be used to look up the secret+for each access key ID and choose the backend it maps to (see Auth Proxy+below).+When an auth proxy is in use \f[V]--auth-key\f[R] is ignored and every+request must be signed with the secret the proxy returns for its access+key ID.+Each access key ID is a separate user, whose multipart uploads and+object metadata are private to it, even if the proxy maps several access+key IDs to the same backend.+.PP+As on AWS, every \f[V]x-amz-*\f[R] header in an authenticated request+must be in its signed headers list; a request with an unsigned+\f[V]x-amz-*\f[R] header is refused with \f[V]AccessDenied\f[R].+S3 clients sign these headers as a matter of course, but a header added+after signing (for example by a proxy in front of the server, or+\f[V]x-amz-meta-*\f[R] headers sent with a presigned URL which only+signs the host) is refused.+.PPLike all rclone flags \f[V]--auth-key\f[R] can be set via environmentvariables, in this case \f[V]RCLONE_AUTH_KEY\f[R].Since this flag can be repeated, the input to \f[V]RCLONE_AUTH_KEY\f[R]@@ -16577,6 +17365,15 @@ Use \f[V]--etag-hash\f[R] if you want to change the hash uses for theNote that using anything other than \f[V]MD5\f[R] (the default) islikely to cause problems for S3 clients which rely on the Etag being theMD5.+Use \f[V]--etag-hash auto\f[R] to use the best hash the backend supports+- with \f[V]--auth-proxy\f[R] that of each user\[aq]s backend.+.PP+The \f[V]ETag\f[R] of an object made by a multipart upload is its hash,+like any other object, rather than the \f[V]-N\f[R] suffixed+\f[V]ETag\f[R] AWS S3 gives these.+If the remote doesn\[aq]t have the MD5 of objects, \f[V]HeadObject\f[R]+and listings only return the MD5 of objects uploaded through+\f[V]serve s3\f[R] while the object\[aq]s metadata is kept (see below)..SS Quickstart.PPFor a simple set up, to serve \f[V]remote:path\f[R] over s3, run the@@ -16626,32 +17423,84 @@ access_key_id = ACCESS_KEY_IDsecret_access_key = SECRET_ACCESS_KEY\f[R].fi+.SS Object uploads (PUT)+.PP+A \f[V]PutObject\f[R] upload only ever changes the object at its key+atomically, on success, a failed or interrupted PUT neither removes nor+overwrites the object already stored at the key, and never leaves a+partial object visible at it.+.PP+Remotes that upload atomically (e.g.+object stores such as \f[V]s3\f[R]) are streamed straight to the+destination.+On remotes where a partial upload would otherwise be visible (e.g.+\f[V]local\f[R]), and whenever \f[V]--vfs-cache-mode\f[R] is+\f[V]writes\f[R] or above, the upload is written to a temporary object+that is renamed into place on success; these remotes need to support a+server-side move or copy for this (nearly all do - without move or copy+the upload is written directly and a failed PUT may leave a partial+object at the key).+If \f[V]serve s3\f[R] is killed part-way through an upload the temporary+object (named with a leading \f[V].rclone_temp_put_\f[R]) may be left+behind; it is hidden from S3 listings but must be removed manually..SS Multipart uploads.PP-By default \f[V]serve s3\f[R] \f[B]streams\f[R] each multipart upload,-in part-number order, into a single \f[V]PutStream\f[R] upload to the-underlying remote, so the whole file is never buffered in memory --memory use stays bounded by the parts in flight.+Multipart uploads are written, in part-number order, to a temporary+object which is renamed into place, server-side, on completion, so the+upload is atomic.+The object at the key only ever changes on a successful completion.+A failed or aborted upload never affects any object already stored under+that name and a partly-uploaded object never becomes visible under it.+.PP+With the default \f[V]--vfs-cache-mode off\f[R] \f[V]serve s3\f[R]+\f[B]streams\f[R] each multipart upload, in part-number order, into a+single streaming upload to the underlying remote, so the whole file is+never buffered in memory: only parts which arrive out of order are+buffered, within the limits described below.The remote then performs its own internal upload (for example its ownmultipart upload, still with bounded memory).-This works for any remote that supports \f[V]PutStream\f[R], which is-nearly all of them, including through \f[V]crypt\f[R].-.PP-The upload is atomic so the destination object only ever changes on a-successful completion.-A failed or aborted upload never affects any object already stored under-that name.-Remotes that upload atomically already (object stores such as-\f[V]s3\f[R]) are streamed straight to the destination.-On remotes where a partial upload would otherwise be visible (such as-\f[V]local\f[R]), the parts are streamed to a temporary object that is-moved into place, server-side, on completion; these remotes therefore-also need to support a server-side move or copy.+Remotes that don\[aq]t support streaming uploads (those that must know+the file size before the upload starts, such as \f[V]onedrive\f[R],+\f[V]pcloud\f[R], \f[V]jottacloud\f[R], \f[V]mailru\f[R],+\f[V]opendrive\f[R], \f[V]putio\f[R], \f[V]protondrive\f[R] and+\f[V]zoho\f[R]) have the parts spooled to a temporary file on \f[B]local+disk\f[R] instead, and uploaded with the size then known on completion,+so they need local disk space for the largest objects in flight rather+than memory.+.PP+With \f[V]--vfs-cache-mode writes\f[R] (or \f[V]full\f[R]) the parts are+written to a temporary file in the VFS cache and uploaded by the VFS+write-back - see Multipart uploads and the VFS cache below.+.PP+The rename into place needs the remote to support a server-side move or+copy, which nearly all do.+It is a cheap rename on most remotes, but on object stores without a+real rename (such as \f[V]s3\f[R] itself) the move is performed as a+server-side copy and delete of the whole object, which can take time and+API calls for large objects.+Concurrent multipart uploads of the same key (which S3 permits) are+safe.+Each writes its own temporary object and the last to complete wins.+.PP+On the few remotes that support neither server side move nor copy, the+parts are written straight to the destination object instead and never+buffered in memory.+This is at some cost in atomicity - the incomplete object is visible+under its final name while the upload is in flight, as it also is for a+plain object PUT on such remotes, and concurrent multipart uploads of+the same key write to the same object and can interleave.+A failed or aborted upload still leaves any pre-existing object+untouched provided the remote uploads atomically and the VFS cache is+off; on a remote where partial uploads are visible it may leave partial+data at the key (like a plain PUT there), and with+\f[V]--vfs-cache-mode writes\f[R] (or \f[V]full\f[R]) a write to the+cache cannot be abandoned, so an aborted upload\[aq]s partial data is+written back to the remote as if it had completed..PP\f[B]Features\f[R].IP \[bu] 2-The whole object is never buffered in memory; memory use is bounded by-the parts in flight, not the upload size.+The whole object is never buffered in memory; only parts arriving out of+order are, within fixed limits, whatever the upload size..IP \[bu] 2Parts can be any size.Clients that don\[aq]t produce uniform-sized parts work fine - for@@ -16665,27 +17514,39 @@ encrypted as one continuous stream..IP \[bu] 2The destination object only ever changes atomically, on completion: anaborted or failed upload leaves any pre-existing object of the same name-untouched, and a partly-uploaded object never becomes visible.+untouched, and a partly-uploaded object never becomes visible (except on+the few remotes with no server-side move or copy, as above)..IP \[bu] 2-Backend-agnostic - it only needs the remote to support-\f[V]PutStream\f[R] (plus a server-side move or copy on remotes that-don\[aq]t upload atomically).+Multipart uploads go through the VFS like any other upload, so they show+in rclone\[aq]s transfer stats and obey \f[V]--bwlimit\f[R].+.IP \[bu] 2+Backend-agnostic - it only needs the remote to support a server-side+move or copy for the rename into place, which nearly all do; a remote+without streaming upload support spools to local disk as above..PP\f[B]Limitations\f[R].IP \[bu] 2Parts must arrive in ascending, contiguous part-number order (1, 2, 3,\&...).+The part the stream needs next is written straight through as it+arrives, so it is never buffered, whatever its size.Parts the client uploads concurrently or out of order are buffered untiltheir turn.The memory used for this buffering is capped, per upload, by\f[V]--multipart-streaming-buffer-limit\f[R] (default \f[V]256M\f[R],-\f[V]0\f[R] for no limit): a part that would take the buffer over the-limit is stalled until the stream drains, so a client that uploads-faster than the remote can accept sees backpressure rather than+\f[V]0\f[R] for no limit), and across all the uploads of each user (each+access key) by \f[V]--multipart-streaming-buffer-total\f[R] (default+\f[V]1G\f[R], \f[V]0\f[R] for no limit), counting each buffered part in+the whole 1 MiB memory pages it occupies: a part that would take the+buffer over either limit is stalled until the stream drains (or, if it+is bigger than the limits, until it is the next part), so a client that+uploads faster than the remote can accept sees backpressure rather thanunbounded server memory use.-Since a stalled part holds its HTTP request open, clients whose upload-concurrency times chunk size exceeds the limit may need a longer read-timeout when the remote is slow.+A stalled part holds its HTTP request open, and one stalled for more+than a minute is failed with a \f[V]SlowDown\f[R] error, which S3+clients retry.+Clients whose upload concurrency times part size exceeds the limits will+see these stalls and retries when the remote is slower than the client.Non-contiguous part numbers are rejected on completion.Configure the client to upload in part order, ideally with lowconcurrency, for the lowest memory use.@@ -16693,50 +17554,276 @@ concurrency, for the lowest memory use.A part uploaded again before completion - typically a client retryingafter a timeout - is accepted: if the earlier copy is still buffered itis replaced, and if it has already been streamed an identical re-upload-is a no-op.+is a no-op (a re-upload which arrives while the part is being streamed+waits for it to finish first).+If a part fails part way through being streamed (for example the+client\[aq]s connection drops) the client\[aq]s retry of it carries on+where it stopped, provided it has the same content.What isn\[aq]t possible is replacing a part that has already been-streamed with \f[I]different\f[R] content - that is rejected.-A failure in the stream to the remote itself still aborts the whole-upload and the client must start it again.+streamed, in whole or in part, with \f[I]different\f[R] content - that+is rejected.+A part which arrives in full but is then rejected, for example because+it doesn\[aq]t match the \f[V]Content-MD5\f[R] the client declared, has+already been streamed, so the whole upload is failed and the client must+start it again.+So is a failure in the stream to the remote itself.(The remote\[aq]s own upload still retries its internal chunks.).IP \[bu] 2+A \f[V]CompleteMultipartUpload\f[R] which fails because the parts it+lists don\[aq]t match those uploaded, or there is a gap in the part+numbers, leaves the upload open, as on S3, so the client can upload any+missing parts and complete it again.+If writing the completed object to the remote fails, the upload is+failed and must be started again.+.IP \[bu] 2Parts are serialised into one stream, so ingest from the client is-effectively single-threaded, although the remote\[aq]s own upload still-runs concurrently.-.IP \[bu] 2-On remotes that don\[aq]t upload atomically (such as \f[V]local\f[R]),-the completed object is moved into place with a server-side operation.-This is a cheap rename on most such remotes.-On these remotes, if \f[V]serve s3\f[R] is killed part-way through an-upload the temporary object (named with a leading-\f[V].rclone_multipart_upload_\f[R]) may be left behind; it is hidden-from S3 listings but must be removed manually.+effectively single-threaded.+When streaming, the remote\[aq]s own upload runs concurrently with the+parts arriving; with the local disk spool or the VFS cache the upload to+the remote only starts on completion.+.IP \[bu] 2+If \f[V]serve s3\f[R] is killed part-way through an upload the temporary+object (named with a leading \f[V].rclone_temp_multipart_\f[R]) may be+left behind; it is hidden from S3 listings but must be removed manually.+.SS Multipart uploads and the VFS cache+.PP+With \f[V]--vfs-cache-mode writes\f[R] (or \f[V]full\f[R]) multipart+uploads do not stream to the remote at all.+The parts are written, in part-number order, to a temporary file in the+VFS cache.+On completion the file is renamed into place and uploaded by the VFS+write-back, exactly like a plain object PUT.+This needs no streaming upload support from the remote.+The rename normally happens in the cache before the upload has started,+but the VFS requires the remote to support a server-side move or copy to+rename files at all (and uses one if the temporary file has already been+written back, e.g.+with \f[V]--vfs-write-back 0\f[R]).+On remotes without either, the parts are written to the cache directly+under the final key instead: the upload still never touches memory, but+it loses its atomicity - the in-flight upload is visible at the key, and+an aborted upload cannot be abandoned once in the cache, so its partial+data is written back to the remote as if it were a completed object.+.PP+Remotes that benefit from \f[V]--vfs-cache-mode writes\f[R]:+.IP \[bu] 2+\f[B]Remotes over slow or unreliable links.\f[R] A failure in a streamed+upload aborts the whole multipart upload and the client must start again+from the first part; a failed write-back upload is retried by the VFS+(see \f[V]--vfs-cache-max-age\f[R] and friends) without the client being+involved.+Ingest from the client also runs at local disk speed rather than being+throttled to the remote\[aq]s pace.+.IP \[bu] 2+\f[B]Workloads that read back or overwrite what they just wrote.\f[R]+The completed object stays in the cache, so subsequent+\f[V]GET\f[R]/\f[V]HEAD\f[R] requests are served locally, and plain PUTs+and multipart uploads to the same key go through the same cache entry so+the last write wins regardless of upload style.+.PP+The trade-offs of the VFS cache:+.IP \[bu] 2+The whole object lands on local disk, so the cache+(\f[V]--cache-dir\f[R]) needs space for the largest objects in flight;+\f[V]--vfs-cache-max-size\f[R] cannot evict files which are still being+uploaded.+.IP \[bu] 2+The \f[V]200 OK\f[R] for \f[V]CompleteMultipartUpload\f[R] means the+data is safely in the \f[B]local cache\f[R], not yet on the remote - the+same durability the cache gives plain PUTs.+If an acknowledgement must mean the data has reached the remote (for+example WAL archiving), use the default \f[V]--vfs-cache-mode off\f[R].+.IP \[bu] 2+The upload to the remote only starts on completion, rather than+overlapping with the parts arriving, so the data reaches the remote+later than with streaming.+.IP \[bu] 2+If \f[V]serve s3\f[R] is killed part-way through an upload, the+temporary file survives in the cache and the VFS cache recovery uploads+it to the remote on restart as a temporary object (named with a leading+\f[V].rclone_temp_multipart_\f[R]); as with the streaming path, it is+hidden from S3 listings but must be removed manually.+.SS Cleaning up temporary objects+.PP+If \f[V]serve s3\f[R] is killed part-way through an upload it can leave+a temporary object behind, named with a leading \f[V].rclone_temp_\f[R].+This whole prefix is reserved: any object whose name (the last+\f[V]/\f[R]-separated segment of its key) starts with+\f[V].rclone_temp_\f[R] is hidden from S3 listings, so don\[aq]t give+real objects such names - an existing object with such a name disappears+from listings (though it stays accessible directly by its key: only+listings hide reserved names, \f[V]GET\f[R], \f[V]HEAD\f[R] and+\f[V]DELETE\f[R] of the exact key still work).+A temporary object never holds acknowledged data - uploads whose+temporary object survived were never confirmed to the client - so old+ones are safe to delete:+.IP+.nf+\f[C]+rclone delete --min-age 24h --include \[dq].rclone_temp_*\[dq] remote:path+\f[R]+.fi+.PP+The \f[V]--min-age\f[R] protects uploads which are still in progress:+make sure it is longer than your longest upload, especially if several+\f[V]serve s3\f[R] instances share the same remote.+.PP+rclone v1.75 named its temporary multipart objects+\f[V].rclone_multipart_upload_*\f[R]; leftovers from an older server are+also hidden from listings and can be cleaned up the same way.+.SS Abandoned uploads+.PP+A client which starts a multipart upload and vanishes without either+completing or aborting it would otherwise hold on to its resources+forever.+.PP+An incomplete multipart upload which has had no activity for+\f[V]--multipart-expiry\f[R] (default \f[V]24h\f[R]) is therefore+aborted and cleaned up, exactly as if the client had called+\f[V]AbortMultipartUpload\f[R], and a \f[V]NOTICE\f[R] is logged.+.PP+An upload with a part still being received is never expired, however+slowly the part is arriving, and each completed part restarts the clock,+so the expiry only needs to outlast the client\[aq]s pauses+\f[I]between\f[R] parts, not the whole upload.+.PP+Late operations on an expired upload fail with \f[V]NoSuchUpload\f[R],+as they do on real S3 when a lifecycle rule has aborted the upload.+Set \f[V]--multipart-expiry 0\f[R] to keep incomplete uploads forever -+but note that abandoned uploads are then never cleaned up and go on+counting towards \f[V]--multipart-max-uploads\f[R] below, so enough of+them will stop the user starting any new multipart upload until the+server is restarted.+.PP+Each multipart upload in progress holds server resources - for example+the buffers the remote\[aq]s own upload uses once the first part has+arrived - so each user (each access key) can have no more than+\f[V]--multipart-max-uploads\f[R] (default \f[V]1000\f[R], \f[V]0\f[R]+for no limit) in progress at once.+Starting another fails with a \f[V]SlowDown\f[R] error, which S3 clients+retry, and a one-off \f[V]NOTICE\f[R] is logged..SS Disabling streaming.PP-If you pass \f[V]--disable-multipart-streaming\f[R], or the remote-doesn\[aq]t support \f[V]PutStream\f[R] (or doesn\[aq]t upload-atomically and can\[aq]t move or copy server-side), multipart uploads+If you pass \f[V]--disable-multipart-streaming\f[R], multipart uploadsare instead \f[B]buffered in memory\f[R] by the underlying S3 library:every part is held in memory and the whole object is written out in one-go when the upload completes (the previous behaviour).+go when the upload completes.This removes the in-order/contiguous-part restriction above, so partscan be uploaded in any order, but \f[B]memory use grows with the size ofthe upload\f[R], so it is only suitable for small objects.A one-off \f[V]NOTICE\f[R] is logged the first time this happens.-.PP-Alternatively, if the client is an rclone \f[V]s3\f[R] remote (like the-\f[V][serves3]\f[R] example above), you can set-\f[V]use_multipart_uploads = false\f[R] on it so it uploads each object-as a single stream and skips multipart uploads altogether.+This flag is the only thing that makes multipart uploads buffer in+memory - it is never done because of missing remote capabilities.+The limits on multipart uploads above+(\f[V]--multipart-streaming-buffer-limit\f[R],+\f[V]--multipart-streaming-buffer-total\f[R] and+\f[V]--multipart-max-uploads\f[R]) do not apply to uploads buffered in+memory, so don\[aq]t use this flag if the server is open to clients you+don\[aq]t trust.+Consider \f[V]--vfs-cache-mode writes\f[R] instead, which buffers the+upload in the VFS cache on disk and takes precedence over+\f[V]--disable-multipart-streaming\f[R].+.SS Memory use+.PP+\f[V]serve s3\f[R] never buffers whole objects in memory (except with+\f[V]--disable-multipart-streaming\f[R], above), and limits the memory+each kind of request can use, so that a client, even a hostile one,+can\[aq]t make it use unbounded memory.+.IP \[bu] 2+\f[B]Object data.\f[R] \f[V]PutObject\f[R] bodies and multipart parts+arriving in order are streamed straight through to the remote.+Parts arriving out of order are buffered, up to+\f[V]--multipart-streaming-buffer-limit\f[R] per upload and+\f[V]--multipart-streaming-buffer-total\f[R] per user.+.IP \[bu] 2+\f[B]Uploads in progress.\f[R] Each user can have at most+\f[V]--multipart-max-uploads\f[R] multipart uploads in progress.+An upload holds little memory until its first part arrives.+After that, with \f[V]--vfs-cache-mode off\f[R], it holds a+\f[V]--streaming-upload-cutoff\f[R] sized buffer plus whatever the+remote\[aq]s own upload buffers, which for many remotes is up to their+chunk size times their upload concurrency (for example+\f[V]--s3-chunk-size\f[R] times \f[V]--s3-upload-concurrency\f[R]).+Plain \f[V]PutObject\f[R] uploads use the same while they are in+progress.+.IP \[bu] 2+\f[B]Metadata.\f[R] The metadata of at most+\f[V]--metadata-max-objects\f[R] objects of each user is kept, each+limited to 2 KB.+.IP \[bu] 2+\f[B]Request bodies.\f[R] The XML bodies of requests such as+\f[V]CompleteMultipartUpload\f[R] and \f[V]DeleteObjects\f[R] are+limited to 10 MiB, and \f[V]DeleteObjects\f[R] to 1000 keys.+A browser form (\f[V]POST\f[R]) upload holds at most 1 MiB in memory,+spooling the rest to a temporary file, and is limited to 5 GB.+.IP \[bu] 2+\f[B]Listings.\f[R] Each page of a \f[V]ListObjects\f[R] listing reads+only the directories needed to fill it, and returns at most 1000 keys.+Directory listings are cached by the VFS directory cache.+.PP+The largest of these is usually the remote\[aq]s upload buffers, which+can be up to \f[V]--multipart-max-uploads\f[R] times the buffers of one+upload for each user if clients start many uploads and send part of+each.+The limits apply to each user separately, so that one user can\[aq]t+stop others uploading, which means the memory use grows with the number+of users.+To limit it, lower \f[V]--multipart-max-uploads\f[R], the remote\[aq]s+chunk size or upload concurrency, or set \f[V]--max-buffer-memory\f[R]+to limit the memory used by the buffers of all uploads to remotes which+use rclone\[aq]s buffer pool (including \f[V]s3\f[R],+\f[V]azureblob\f[R] and \f[V]b2\f[R]).+Buffered multipart parts come from the same pool, so if you set+\f[V]--max-buffer-memory\f[R] make it comfortably bigger than+\f[V]--multipart-streaming-buffer-total\f[R] times the number of users+uploading at once, otherwise buffered parts can use up all the memory+the remote needs to upload the parts which would free them.+.PP+The number of connections is not limited, so if the server is open to+clients you don\[aq]t trust, use \f[V]--auth-key\f[R] or+\f[V]--auth-proxy\f[R] and consider putting it behind a reverse proxy+which limits connections.+.SS Design goals+.PP+These goals explain many of the limitations below.+.IP \[bu] 2+\f[B]Objects are plain files on the remote.\f[R] A bucket is a directory+in the root and a key is a path in it, so the remote stays usable by+other tools, such as \f[V]rclone mount\f[R] or \f[V]rclone sync\f[R].+So keys must be paths (see Object names).+.IP \[bu] 2+\f[B]No database.\f[R] \f[V]serve s3\f[R] stores nothing but the+objects.+What the remote can\[aq]t hold, such as user metadata, tags, the MD5+\f[V]ETag\f[R] of an object on a remote without MD5 and the directory+markers put, is kept in memory as a best effort, and lost on restart.+.IP \[bu] 2+\f[B]Works with any rclone remote.\f[R] What \f[V]serve s3\f[R] can do+depends on what the remote supports, such as the hashes used for+\f[V]ETag\f[R]s and modification times, rather than needing a particular+backend.+.IP \[bu] 2+\f[B]Bounded memory.\f[R] Uploads, including multipart uploads, are+streamed to the remote and the memory each request can use is limited+(see Memory use).+Nothing is written to local disk unless the VFS cache is used or the+remote can\[aq]t stream uploads (see Multipart uploads).+.IP \[bu] 2+\f[B]Refuse rather than ignore.\f[R] Unsupported features are refused,+or answered as S3 answers when the feature isn\[aq]t configured, so a+client is never told something was done which wasn\[aq]t (see+Compatibility).+.IP \[bu] 2+\f[B]A storage gateway, not a full S3.\f[R] \f[V]serve s3\f[R]+implements the parts of the S3 API clients use to store and fetch+objects.+Versioning, ACLs, bucket policies, lifecycle rules and the like are out+of scope.+.IP \[bu] 2+\f[B]Compatible with rclone\[aq]s s3 backend.\f[R] This includes its+directory markers, which the tests check..SS Bugs.PP-Multipart server side copies do not work (see-#7454 (https://github.com/rclone/rclone/issues/7454)).-These take a very long time and eventually fail.-The default threshold for multipart server side copies is 5G which is-the maximum it can be, so files above this side will fail to be server-side copied.-.PPFor a current list of \f[V]serve s3\f[R] bugs see the serves3 (https://github.com/rclone/rclone/labels/serve%20s3) bug category onGitHub.@@ -16744,25 +17831,55 @@ GitHub..PP\f[V]serve s3\f[R] will treat all directories in the root as buckets andignore all files in the root.-You can use \f[V]CreateBucket\f[R] to create folders under the root, but-you can\[aq]t create empty folders under other folders not in the root.+You can use \f[V]CreateBucket\f[R] to create folders under the root, and+put a directory marker to create an empty folder below them (see Object+names)..PPWhen using \f[V]PutObject\f[R] or \f[V]DeleteObject\f[R], rclone willautomatically create or clean up empty folders.If you don\[aq]t want to clean up empty folders automatically, use-\f[V]--no-cleanup\f[R].-.PP-When using \f[V]ListObjects\f[R], rclone will use \f[V]/\f[R] when the-delimiter is empty.-This reduces backend requests with no effect on most operations, but if-the delimiter is something other than \f[V]/\f[R] and empty, rclone will-do a full recursive search of the backend, which can take some time.-.PP-Versioning is not currently supported.+\f[V]--directory-markers\f[R] (see Object names).+The deprecated \f[V]--no-cleanup\f[R] does the same.+.PP+When using \f[V]ListObjects\f[R] with the delimiter \f[V]/\f[R], rclone+lists the directory the prefix is in, showing its subdirectories as+common prefixes, including empty ones.+Without a delimiter the listing is recursive: rclone walks the directory+tree below the prefix, reading only as many directories as it needs to+fill each page, and lists empty directories as their directory markers+(see below).+Any other delimiter is supported too, but needs the whole directory tree+below the prefix walking, unless the delimiter is in the directory+names..PP-Metadata will only be saved in memory other than the rclone-\f[V]mtime\f[R] metadata which will be set as the modification time of-the file.+Versioning is not supported (see Design goals).+.PP+Conditional writes, \f[V]PutObject\f[R] and+\f[V]CompleteMultipartUpload\f[R] with \f[V]If-None-Match\f[R] or+\f[V]If-Match\f[R], are supported.+The conditions are checked before the object is written, and again as it+is moved into place when it is written to a temporary object first.+This happens for multipart uploads and, for \f[V]PutObject\f[R], with+remotes which can upload partial files (such as local disk) or with+\f[V]--vfs-cache-mode minimal\f[R] or above.+The second check is atomic against other writes through+\f[V]serve s3\f[R], but neither is atomic against changes made to the+remote some other way.+If the remote doesn\[aq]t support the \f[V]--etag-hash\f[R], objects+have no \f[V]ETag\f[R] so \f[V]If-Match\f[R] with an \f[V]ETag\f[R]+always fails.+.PP+Metadata will only be saved in memory (see Design goals) other than the+rclone \f[V]mtime\f[R] metadata which will be set as the modification+time of the file.+The metadata of at most \f[V]--metadata-max-objects\f[R] objects+(default \f[V]100000\f[R], \f[V]0\f[R] for no limit) of each user (each+access key) is kept; after that the metadata of the user\[aq]s least+recently used object is forgotten.+The metadata of an object is ignored once the object has been changed+some other way, for example by another \f[V]--auth-proxy\f[R] user or+directly on the backend.+Object tags are kept with the metadata in the same way..SS Object names.PP\f[V]serve s3\f[R] stores objects as files in the backend, so object@@ -16770,52 +17887,273 @@ keys are mapped to file paths rather than treated as the opaque stringsAWS S3 allows.Keys must be in canonical path form: keys that contain \f[V]..\f[R] or\f[V].\f[R] path segments, repeated slashes (\f[V]//\f[R]), or a leading-or trailing slash are rejected with a \f[V]400 Bad Request\f[R]+slash are rejected with a \f[V]400 Bad Request\f[R](\f[V]InvalidArgument\f[R]) instead of being normalised, sincenormalising them could alias two distinct keys to the same file orresolve a key outside its bucket.This matches the behaviour of other S3 servers such as MinIO.-.SS Supported operations.PP-\f[V]serve s3\f[R] currently supports the following operations.-.IP \[bu] 2-Bucket-.RS 2-.IP \[bu] 2-\f[V]ListBuckets\f[R]-.IP \[bu] 2-\f[V]CreateBucket\f[R]-.IP \[bu] 2-\f[V]DeleteBucket\f[R]-.RE-.IP \[bu] 2-Object-.RS 2-.IP \[bu] 2-\f[V]HeadObject\f[R]-.IP \[bu] 2-\f[V]ListObjects\f[R]-.IP \[bu] 2-\f[V]GetObject\f[R]-.IP \[bu] 2+A key ending in \f[V]/\f[R], such as \f[V]dir/\f[R], is a directory+marker, standing for the directory \f[V]dir\f[R].+Putting an empty directory marker creates the directory (one with data+is rejected with \f[V]400 Bad Request\f[R]), and deleting it removes the+directory if it is empty.+Directory markers are found by \f[V]HeadObject\f[R], \f[V]GetObject\f[R]+and \f[V]CopyObject\f[R] for any directory, but only the markers of+empty directories, and of the directories a marker was put for, are+listed as objects, as other directories may not have been made by+putting a marker.+The markers put are remembered in memory with the metadata (see+Limitations), so after a restart, or once forgotten, the marker of a+directory holding other keys is no longer listed.+While a marker is remembered its directory isn\[aq]t removed when the+last key below it is deleted.+.PP+With \f[V]--directory-markers\f[R] every directory is listed as a+directory marker, so the markers are listed after a restart too, as+needed by clients which rely on them, such as rclone\[aq]s s3 backend+with \f[V]--s3-directory-markers\f[R].+As every directory has a marker, a directory isn\[aq]t removed when the+last key below it is deleted.+Deleting the marker of a directory which holds other keys stops it being+listed, and the directory is then removed with the last key below it,+but this is only remembered in memory, like the markers put without+\f[V]--directory-markers\f[R].+.PP+As keys are file paths, a key can\[aq]t be both an object and a+directory holding other objects, for example \f[V]a/b\f[R] and+\f[V]a/b/c\f[R].+Writing an object which would need this fails with+\f[V]400 Bad Request\f[R] (\f[V]InvalidRequest\f[R]).+.SS Compatibility+.PP+\f[V]serve s3\f[R] implements the parts of the S3 API that S3 clients+commonly use to store and fetch objects.+Features it doesn\[aq]t support are refused with+\f[V]501 NotImplemented\f[R], or answered as AWS S3 answers for a bucket+without that feature configured, rather than being silently ignored.+.PP+.TS+tab(@);+lw(23.3n) lw(30.0n) lw(16.7n).+T{+Feature+T}@T{+Supported+T}@T{+Notes+T}+_+T{+\f[V]ListBuckets\f[R], \f[V]CreateBucket\f[R], \f[V]DeleteBucket\f[R],+\f[V]HeadBucket\f[R], \f[V]GetBucketLocation\f[R]+T}@T{+Yes+T}@T{+Buckets are the directories in the root of the remote.+T}+T{+\f[V]ListObjects\f[R], \f[V]ListObjectsV2\f[R]+T}@T{+Yes+T}@T{+Prefixes, delimiters, markers, \f[V]start-after\f[R], continuation+tokens, \f[V]fetch-owner\f[R] and \f[V]encoding-type=url\f[R] all work.+Delimiters other than \f[V]/\f[R] need the whole directory tree walking.+T}+T{+\f[V]HeadObject\f[R], \f[V]GetObject\f[R]+T}@T{+Yes+T}@T{+Including \f[V]Range\f[R], \f[V]partNumber\f[R], the+\f[V]response-*\f[R] header overrides and the \f[V]If-Match\f[R],+\f[V]If-None-Match\f[R], \f[V]If-Modified-Since\f[R] and+\f[V]If-Unmodified-Since\f[R] conditions.+T}+T{\f[V]PutObject\f[R]-.IP \[bu] 2-\f[V]DeleteObject\f[R]-.IP \[bu] 2-\f[V]DeleteObjects\f[R]-.IP \[bu] 2-\f[V]CreateMultipartUpload\f[R]-.IP \[bu] 2-\f[V]CompleteMultipartUpload\f[R]-.IP \[bu] 2-\f[V]AbortMultipartUpload\f[R]-.IP \[bu] 2+T}@T{+Yes+T}@T{+Including \f[V]Content-MD5\f[R], \f[V]aws-chunked\f[R] uploads and+uploads with no \f[V]Content-Length\f[R].+T}+T{\f[V]CopyObject\f[R]-.IP \[bu] 2-\f[V]UploadPart\f[R]-.RE+T}@T{+Yes+T}@T{+Including the metadata and tagging directives and the+\f[V]x-amz-copy-source-if-*\f[R] conditions.+T}+T{+\f[V]DeleteObject\f[R], \f[V]DeleteObjects\f[R]+T}@T{+Yes+T}@T{+T}+T{+Multipart uploads+T}@T{+Yes+T}@T{+\f[V]CreateMultipartUpload\f[R], \f[V]UploadPart\f[R],+\f[V]UploadPartCopy\f[R], \f[V]CompleteMultipartUpload\f[R],+\f[V]AbortMultipartUpload\f[R], \f[V]ListMultipartUploads\f[R] and+\f[V]ListParts\f[R].+The \f[V]ETag\f[R] of the object is its hash, not the \f[V]-N\f[R]+suffixed one AWS gives.+The 5 MiB minimum part size isn\[aq]t enforced.+T}+T{+Conditional writes+T}@T{+Partial+T}@T{+\f[V]If-Match\f[R] and \f[V]If-None-Match\f[R] on \f[V]PutObject\f[R]+and \f[V]CompleteMultipartUpload\f[R] are checked, but not atomically+against changes made to the remote some other way (see Limitations).+T}+T{+Authentication+T}@T{+Partial+T}@T{+AWS Signature Version 4, in headers, presigned URLs and browser form+(\f[V]POST\f[R]) uploads.+Signature Version 2 isn\[aq]t supported.+T}+T{+Anonymous access+T}@T{+Partial+T}@T{+All requests are allowed without \f[V]--auth-key\f[R] or+\f[V]--auth-proxy\f[R], and none are with them, as buckets and objects+can\[aq]t be made public.+T}+T{+Metadata+T}@T{+Partial+T}@T{+\f[V]x-amz-meta-*\f[R], the storage class and the standard headers such+as \f[V]Content-Type\f[R] are kept in memory only (see Limitations),+except the modification time.+The storage class is only recorded, so doesn\[aq]t change how the object+is stored.+T}+T{+Directory markers+T}@T{+Partial+T}@T{+Keys ending in \f[V]/\f[R] are stored as directories, so can\[aq]t hold+data.+A marker put for a directory holding other keys is only listed while it+is remembered in memory, unless \f[V]--directory-markers\f[R] is set+(see Object names).+T}+T{+Checksums+T}@T{+Partial+T}@T{+The \f[V]x-amz-checksum-*\f[R] of a \f[V]PutObject\f[R] (CRC32, CRC32C,+CRC64NVME, SHA1 or SHA256) is checked and returned with the metadata.+The \f[V]Content-MD5\f[R] and \f[V]x-amz-checksum-*\f[R] of the bodies+of other requests, such as \f[V]DeleteObjects\f[R], are checked.+The checksums of multipart uploads aren\[aq]t supported.+T}+T{+\f[V]GetObjectAttributes\f[R]+T}@T{+Partial+T}@T{+Everything but the checksums of multipart uploads.+T}+T{+Object tagging+T}@T{+Partial+T}@T{+\f[V]GetObjectTagging\f[R], \f[V]PutObjectTagging\f[R] and+\f[V]DeleteObjectTagging\f[R] work, but tags are kept in memory with the+metadata.+T}+T{+Bucket owner+T}@T{+No+T}@T{+\f[V]serve s3\f[R] has no AWS account IDs, so+\f[V]x-amz-expected-bucket-owner\f[R] and+\f[V]x-amz-source-expected-bucket-owner\f[R] are ignored.+T}+T{+ACLs+T}@T{+Partial+T}@T{+\f[V]GetBucketAcl\f[R] and \f[V]GetObjectAcl\f[R] give the owner full+control.+Only the \f[V]private\f[R] ACL can be set.+T}+T{+CORS+T}@T{+Partial+T}@T{+Preflight (\f[V]OPTIONS\f[R]) requests are answered, allowing any+origin.+CORS configuration (\f[V]PutBucketCors\f[R]) isn\[aq]t supported.+T}+T{+Server side encryption+T}@T{+No+T}@T{+SSE-S3 (\f[V]AES256\f[R]) is accepted but the data isn\[aq]t encrypted.+SSE-KMS and SSE-C are refused.+T}+T{+Versioning+T}@T{+No+T}@T{+Buckets are reported as unversioned and \f[V]ListObjectVersions\f[R]+lists each object as version \f[V]null\f[R].+Enabling versioning is refused.+T}+T{+Object lock, retention and legal hold+T}@T{+No+T}@T{+T}+T{+Bucket configuration+T}@T{+No+T}@T{+Bucket policies, tagging, lifecycle, logging, encryption, website,+replication, notifications and the like.+Getting them fails as for a bucket without them configured.+T}+T{+\f[V]SelectObjectContent\f[R], \f[V]RestoreObject\f[R], torrents+T}@T{+No+T}@T{+T}+.TE.PP-Other operations will return error \f[V]Unimplemented\f[R].+The ceph s3-tests (https://github.com/ceph/s3-tests) can be run against+\f[V]serve s3\f[R] to check its compatibility.+See+cmd/serve/s3/s3tests (https://github.com/rclone/rclone/tree/master/cmd/serve/s3/s3tests)+for how..SS Authentication.PPBy default this will serve files without needing a login.@@ -16851,8 +18189,21 @@ htpasswd -B htpasswd anotherUser\f[R].fi.PP+Use the same hash format and cost for every entry in the file.+If the entries differ, the time taken to reject a wrong password can+reveal which usernames exist.+.PPThe password file can be updated while rclone is running..PP+When authentication is enabled every request must be authenticated,+including HTTP OPTIONS requests, as the answer to an OPTIONS request can+reveal whether a path exists.+Health checks or monitoring which use unauthenticated OPTIONS requests+will need to send credentials.+The exception is browser CORS preflight requests when+\f[V]--allow-origin\f[R] is set, which are answered without+authentication as browsers never send credentials with them.+.PPUse \f[V]--realm\f[R] to set the authentication realm..PPUse \f[V]--salt\f[R] to change the password hashing salt from the@@ -16889,6 +18240,15 @@ responses, will overriding existing values.The flag may be repeated to add multiple headers.Use the format \f[V]Header-Name: value\f[R]..PP+\f[V]--allow-origin\f[R] enables cross-origin (CORS) requests from+browsers by setting the \f[V]Access-Control-Allow-Origin\f[R] header on+every response to the origin given, e.g.+\f[V]--allow-origin https://app.example.com\f[R].+Browser CORS preflight OPTIONS requests are answered by rclone without+needing authentication as browsers never send credentials with them.+Setting this to \f[V]*\f[R] allows any origin which is a security risk+on an authenticated server.+.PP\f[V]--baseurl\f[R] controls the URL prefix that rclone serves from.By default rclone will serve from the root.If you used \f[V]--baseurl \[dq]/rclone\[dq]\f[R] then rclone would@@ -17029,8 +18389,8 @@ at all times.The buffered data is bound to one open file and won\[aq]t be shared..PPThis flag is a upper limit for the used memory per open file.-The buffer will only use memory for data that is downloaded but not not-yet read.+The buffer will only use memory for data that is downloaded but not yet+read.If the buffer is empty, only a small amount of memory will be used..PPThe maximum memory used by rclone for buffering can be up to@@ -17092,7 +18452,8 @@ This cache flushing strategy is efficient and more relevant files arelikely to remain cached..PPThe \f[V]--vfs-cache-max-age\f[R] will evict files from the cache after-the set time since last access has passed.+the set time since last access has passed; it is based on access time,+not on when the file was first added to the cache.The default value of 1 hour will start evicting files from cache thathaven\[aq]t been accessed for 1 hour.When a cached file is accessed the 1 hour timer is reset to 0 and will@@ -17265,8 +18626,8 @@ Rclone reads \f[V]--vfs-read-chunk-streams\f[R] chunks of size\f[V]--vfs-read-chunk-size\f[R] concurrently.The size for each read will stay constant..PP-This improves performance performance massively on high latency links or-very high bandwidth links to high performance object stores.+This improves performance massively on high latency links or very high+bandwidth links to high performance object stores..PPSome experimentation will be needed to find the optimum values of\f[V]--vfs-read-chunk-size\f[R] and \f[V]--vfs-read-chunk-streams\f[R]@@ -17517,6 +18878,155 @@ total 1048578If the file has no metadata it will be returned as \f[V]{}\f[R] and ifthere is an error reading the metadata the error will be returned as\f[V]{\[dq]error\[dq]:\[dq]error string\[dq]}\f[R].+.SS Auth Proxy+.PP+If you supply the parameter \f[V]--auth-proxy /path/to/program\f[R] then+rclone will use that program to generate backends on the fly which then+are used to authenticate incoming requests.+This uses a simple JSON based protocol with input on STDIN and output on+STDOUT.+.PP+\f[B]PLEASE NOTE:\f[R] \f[V]--auth-proxy\f[R] and+\f[V]--authorized-keys\f[R] cannot be used together, if+\f[V]--auth-proxy\f[R] is set the authorized keys option will be+ignored.+.PP+There is an example program+bin/test_proxy.py (https://github.com/rclone/rclone/blob/master/bin/test_proxy.py)+in the rclone source code.+.PP+The program\[aq]s job is to take a \f[V]user\f[R] and \f[V]pass\f[R] on+the input and turn those into the config for a backend on STDOUT in JSON+format.+This config will have any default parameters for the backend added, but+it won\[aq]t use configuration from environment variables or command+line options - it is the job of the proxy program to make a complete+config.+.PP+This config generated must have this extra parameter+.IP \[bu] 2+\f[V]_root\f[R] - root to use for the backend+.PP+And it may have these parameters+.IP \[bu] 2+\f[V]_obscure\f[R] - comma separated strings for parameters to obscure+.IP \[bu] 2+\f[V]_secret_access_key\f[R] - the secret for S3 access key auth (see+below)+.PP+If password authentication was used by the client, input to the proxy+process (on STDIN) would look similar to this:+.IP+.nf+\f[C]+{+ \[dq]user\[dq]: \[dq]me\[dq],+ \[dq]pass\[dq]: \[dq]mypassword\[dq],+ \[dq]client_ip\[dq]: \[dq]192.168.1.1\[dq]+}+\f[R]+.fi+.PP+If public-key authentication was used by the client, input to the proxy+process (on STDIN) would look similar to this:+.IP+.nf+\f[C]+{+ \[dq]user\[dq]: \[dq]me\[dq],+ \[dq]public_key\[dq]: \[dq]AAAAB3NzaC1yc2EAAAADAQABAAABAQDuwESFdAe14hVS6omeyX7edc...JQdf\[dq],+ \[dq]client_ip\[dq]: \[dq]192.168.1.1\[dq]+}+\f[R]+.fi+.PP+If the client authenticated with an S3 access key+(\f[V]rclone serve s3\f[R]), the client never sends its secret, only a+signature made with it, so the input contains just the access key ID as+the \f[V]user\f[R] with no \f[V]pass\f[R] or \f[V]public_key\f[R]:+.IP+.nf+\f[C]+{+ \[dq]user\[dq]: \[dq]AKIAIOSFODNN7EXAMPLE\[dq],+ \[dq]client_ip\[dq]: \[dq]192.168.1.1\[dq]+}+\f[R]+.fi+.PP+In this case the program must look up the secret access key for that+access key ID and return it in the \f[V]_secret_access_key\f[R] field of+the output.+Rclone then uses that secret to verify the signature on the request,+refusing the request if it does not match.+This means the proxy program is the source of truth for both the+credentials and the backend they map to.+If the program does not return \f[V]_secret_access_key\f[R] or returns+it empty the request is refused.+.PP+The program\[aq]s answer for an access key ID is cached (see below) but+is checked with the program again after 5 minutes even if the access key+ID is in constant use, so revoking an access key ID in the program takes+effect within 5 minutes.+A rotated secret takes effect on the first request signed with it.+.PP+The \f[V]client_ip\f[R] key holds the IP address the client connected+from, without a port number.+It can be used to restrict logins to certain networks, or to log+authentication attempts centrally.+It is omitted if the client has no IP address, for example when+connecting over a unix socket.+Note that if rclone is behind a reverse proxy this will be the address+of the reverse proxy and not the original client.+.PP+And as an example return this on STDOUT+.IP+.nf+\f[C]+{+ \[dq]type\[dq]: \[dq]sftp\[dq],+ \[dq]_root\[dq]: \[dq]\[dq],+ \[dq]_obscure\[dq]: \[dq]pass\[dq],+ \[dq]user\[dq]: \[dq]me\[dq],+ \[dq]pass\[dq]: \[dq]mypassword\[dq],+ \[dq]host\[dq]: \[dq]sftp.example.com\[dq]+}+\f[R]+.fi+.PP+This would mean that an SFTP backend would be created on the fly for the+\f[V]user\f[R] and \f[V]pass\f[R]/\f[V]public_key\f[R] returned in the+output to the host given.+Note that since \f[V]_obscure\f[R] is set to \f[V]pass\f[R], rclone will+obscure the \f[V]pass\f[R] parameter before creating the backend (which+is required for sftp backends).+.PP+The program can manipulate the supplied \f[V]user\f[R] in any way, for+example to make proxy to many different sftp backends, you could make+the \f[V]user\f[R] be \f[V]user\[at]example.com\f[R] and then set the+\f[V]host\f[R] to \f[V]example.com\f[R] in the output and the user to+\f[V]user\f[R].+For security you\[aq]d probably want to restrict the \f[V]host\f[R] to a+limited list.+.PP+An internal cache of backends is keyed on the \f[V]user\f[R], the kind+of credential presented, a hash of the \f[V]pass\f[R] or+\f[V]public_key\f[R], and the \f[V]client_ip\f[R].+This means that if a user\[aq]s password or public-key changes, the+client connects from a new IP address, or the proxy returns different+config parameters (eg a rotated \f[V]api_key\f[R]), a fresh backend will+be created on the next request rather than the cached one being reused.+A backend created for a public key login is never reused for a password+login, so a public key cannot be sent as a password to skip the program.+.PP+A cached answer is checked with the program again after 5 minutes even+if the user is in constant use, so revoking a user in the program takes+effect within 5 minutes.+For \f[V]serve sftp\f[R] this applies to new logins - an existing+connection is not closed.+.PP+This can be used to build general purpose proxies to any kind of backend+that rclone supports..IP.nf\f[C]@@ -17536,7 +19046,8 @@ rclone serve s3 remote:path [flags]--client-ca string Client certificate authority to verify clients with--dir-cache-time Duration Time to cache directory entries for (default 5m0s)--dir-perms FileMode Directory permissions (default 777)- --disable-multipart-streaming Buffer multipart uploads in memory instead of streaming them to the backend (see the Multipart uploads docs section)+ --directory-markers List every directory as a directory marker object+ --disable-multipart-streaming Buffer multipart uploads in memory instead of streaming them to the backend--etag-hash string Which hash to use for the ETag, or auto or blank for off (default \[dq]MD5\[dq])--file-perms FileMode File permissions (default 666)--force-path-style If true use path style access if false use virtual hosted style (default true)@@ -17546,10 +19057,14 @@ rclone serve s3 remote:path [flags]--key string TLS PEM Private key--link-perms FileMode Link permissions (default 666)--max-header-bytes int Maximum size of request header (default 4096)+ --metadata-max-objects int Maximum number of objects of a user whose metadata is kept in memory, 0 for unlimited (default 100000)--min-tls-version string Minimum TLS version that is acceptable (default \[dq]tls1.0\[dq])- --multipart-streaming-buffer-limit SizeSuffix Maximum memory buffered per streamed multipart upload for parts arriving out of order, 0 for unlimited (see the Multipart uploads docs section) (default 256Mi)+ --multipart-expiry Duration Abort incomplete multipart uploads idle for longer than this, 0 to keep forever (default 1d)+ --multipart-max-uploads int Maximum number of streamed multipart uploads a user can have in progress at once, 0 for unlimited (default 1000)+ --multipart-streaming-buffer-limit SizeSuffix Maximum memory buffered per streamed multipart upload for parts arriving out of order, 0 for unlimited (default 256Mi)+ --multipart-streaming-buffer-total SizeSuffix Maximum memory buffered by all the streamed multipart uploads of a user for parts arriving out of order, 0 for unlimited (default 1Gi)--no-checksum Don\[aq]t compare checksums on up/download- --no-cleanup Not to cleanup empty folder after object is deleted+ --no-cleanup Deprecated: use --directory-markers instead--no-modtime Don\[aq]t read/write the modification time (can speed things up)--no-seek Don\[aq]t allow seeking in files--pass string Password for authentication@@ -17784,8 +19299,8 @@ at all times.The buffered data is bound to one open file and won\[aq]t be shared..PPThis flag is a upper limit for the used memory per open file.-The buffer will only use memory for data that is downloaded but not not-yet read.+The buffer will only use memory for data that is downloaded but not yet+read.If the buffer is empty, only a small amount of memory will be used..PPThe maximum memory used by rclone for buffering can be up to@@ -17847,7 +19362,8 @@ This cache flushing strategy is efficient and more relevant files arelikely to remain cached..PPThe \f[V]--vfs-cache-max-age\f[R] will evict files from the cache after-the set time since last access has passed.+the set time since last access has passed; it is based on access time,+not on when the file was first added to the cache.The default value of 1 hour will start evicting files from cache thathaven\[aq]t been accessed for 1 hour.When a cached file is accessed the 1 hour timer is reset to 0 and will@@ -18020,8 +19536,8 @@ Rclone reads \f[V]--vfs-read-chunk-streams\f[R] chunks of size\f[V]--vfs-read-chunk-size\f[R] concurrently.The size for each read will stay constant..PP-This improves performance performance massively on high latency links or-very high bandwidth links to high performance object stores.+This improves performance massively on high latency links or very high+bandwidth links to high performance object stores..PPSome experimentation will be needed to find the optimum values of\f[V]--vfs-read-chunk-size\f[R] and \f[V]--vfs-read-chunk-streams\f[R]@@ -18301,9 +19817,12 @@ This config generated must have this extra parameter.IP \[bu] 2\f[V]_root\f[R] - root to use for the backend.PP-And it may have this parameter+And it may have these parameters.IP \[bu] 2\f[V]_obscure\f[R] - comma separated strings for parameters to obscure+.IP \[bu] 2+\f[V]_secret_access_key\f[R] - the secret for S3 access key auth (see+below).PPIf password authentication was used by the client, input to the proxyprocess (on STDIN) would look similar to this:@@ -18312,7 +19831,8 @@ process (on STDIN) would look similar to this:\f[C]{\[dq]user\[dq]: \[dq]me\[dq],- \[dq]pass\[dq]: \[dq]mypassword\[dq]+ \[dq]pass\[dq]: \[dq]mypassword\[dq],+ \[dq]client_ip\[dq]: \[dq]192.168.1.1\[dq]}\f[R].fi@@ -18324,11 +19844,51 @@ process (on STDIN) would look similar to this:\f[C]{\[dq]user\[dq]: \[dq]me\[dq],- \[dq]public_key\[dq]: \[dq]AAAAB3NzaC1yc2EAAAADAQABAAABAQDuwESFdAe14hVS6omeyX7edc...JQdf\[dq]+ \[dq]public_key\[dq]: \[dq]AAAAB3NzaC1yc2EAAAADAQABAAABAQDuwESFdAe14hVS6omeyX7edc...JQdf\[dq],+ \[dq]client_ip\[dq]: \[dq]192.168.1.1\[dq]}\f[R].fi.PP+If the client authenticated with an S3 access key+(\f[V]rclone serve s3\f[R]), the client never sends its secret, only a+signature made with it, so the input contains just the access key ID as+the \f[V]user\f[R] with no \f[V]pass\f[R] or \f[V]public_key\f[R]:+.IP+.nf+\f[C]+{+ \[dq]user\[dq]: \[dq]AKIAIOSFODNN7EXAMPLE\[dq],+ \[dq]client_ip\[dq]: \[dq]192.168.1.1\[dq]+}+\f[R]+.fi+.PP+In this case the program must look up the secret access key for that+access key ID and return it in the \f[V]_secret_access_key\f[R] field of+the output.+Rclone then uses that secret to verify the signature on the request,+refusing the request if it does not match.+This means the proxy program is the source of truth for both the+credentials and the backend they map to.+If the program does not return \f[V]_secret_access_key\f[R] or returns+it empty the request is refused.+.PP+The program\[aq]s answer for an access key ID is cached (see below) but+is checked with the program again after 5 minutes even if the access key+ID is in constant use, so revoking an access key ID in the program takes+effect within 5 minutes.+A rotated secret takes effect on the first request signed with it.+.PP+The \f[V]client_ip\f[R] key holds the IP address the client connected+from, without a port number.+It can be used to restrict logins to certain networks, or to log+authentication attempts centrally.+It is omitted if the client has no IP address, for example when+connecting over a unix socket.+Note that if rclone is behind a reverse proxy this will be the address+of the reverse proxy and not the original client.+.PPAnd as an example return this on STDOUT.IP.nf@@ -18359,12 +19919,21 @@ the \f[V]user\f[R] be \f[V]user\[at]example.com\f[R] and then set theFor security you\[aq]d probably want to restrict the \f[V]host\f[R] to alimited list..PP-An internal cache of backends is keyed on the \f[V]user\f[R] and a hash-of the \f[V]pass\f[R] or \f[V]public_key\f[R].-This means that if a user\[aq]s password or public-key changes, or the-proxy returns different config parameters (eg a rotated-\f[V]api_key\f[R]), a fresh backend will be created on the next request-rather than the cached one being reused.+An internal cache of backends is keyed on the \f[V]user\f[R], the kind+of credential presented, a hash of the \f[V]pass\f[R] or+\f[V]public_key\f[R], and the \f[V]client_ip\f[R].+This means that if a user\[aq]s password or public-key changes, the+client connects from a new IP address, or the proxy returns different+config parameters (eg a rotated \f[V]api_key\f[R]), a fresh backend will+be created on the next request rather than the cached one being reused.+A backend created for a public key login is never reused for a password+login, so a public key cannot be sent as a password to skip the program.+.PP+A cached answer is checked with the program again after 5 minutes even+if the user is in constant use, so revoking a user in the program takes+effect within 5 minutes.+For \f[V]serve sftp\f[R] this applies to new logins - an existing+connection is not closed..PPThis can be used to build general purpose proxies to any kind of backendthat rclone supports.@@ -18597,6 +20166,15 @@ responses, will overriding existing values.The flag may be repeated to add multiple headers.Use the format \f[V]Header-Name: value\f[R]..PP+\f[V]--allow-origin\f[R] enables cross-origin (CORS) requests from+browsers by setting the \f[V]Access-Control-Allow-Origin\f[R] header on+every response to the origin given, e.g.+\f[V]--allow-origin https://app.example.com\f[R].+Browser CORS preflight OPTIONS requests are answered by rclone without+needing authentication as browsers never send credentials with them.+Setting this to \f[V]*\f[R] allows any origin which is a security risk+on an authenticated server.+.PP\f[V]--baseurl\f[R] controls the URL prefix that rclone serves from.By default rclone will serve from the root.If you used \f[V]--baseurl \[dq]/rclone\[dq]\f[R] then rclone would@@ -18717,6 +20295,51 @@ T}@T{Currently unused.T}T{+\&.Path+T}@T{+T}@T{+The path of the directory from the root with leading and trailing+slashes, e.g.+\[aq]/dir/\[aq].+T}+T{+\&.IsRoot+T}@T{+T}@T{+Boolean for if this is the root directory.+T}+T{+\&.UpLink+T}@T{+T}@T{+The link to the parent directory.+T}+T{+\&.Static+T}@T{+T}@T{+Boolean for if the listing is being written as a static page by rclone+index.+T}+T{+\&.NumDirs+T}@T{+T}@T{+The number of directories in the listing.+T}+T{+\&.NumFiles+T}@T{+T}@T{+The number of files in the listing.+T}+T{+\&.TotalSize+T}@T{+T}@T{+The total size in bytes of the files in the listing.+T}+T{\&.BreadcrumbT}@T{T}@T{@@ -18770,6 +20393,13 @@ T}@T{T}@T{The UTC timestamp of an entry.T}+T{+T}@T{+\&.MimeType+T}@T{+The MIME type of an entry guessed from its name, or+\[aq]inode/directory\[aq].+T}.TE.PPThe server also makes the following functions available so that they can@@ -18842,8 +20472,21 @@ htpasswd -B htpasswd anotherUser\f[R].fi.PP+Use the same hash format and cost for every entry in the file.+If the entries differ, the time taken to reject a wrong password can+reveal which usernames exist.+.PPThe password file can be updated while rclone is running..PP+When authentication is enabled every request must be authenticated,+including HTTP OPTIONS requests, as the answer to an OPTIONS request can+reveal whether a path exists.+Health checks or monitoring which use unauthenticated OPTIONS requests+will need to send credentials.+The exception is browser CORS preflight requests when+\f[V]--allow-origin\f[R] is set, which are answered without+authentication as browsers never send credentials with them.+.PPUse \f[V]--realm\f[R] to set the authentication realm..PPUse \f[V]--salt\f[R] to change the password hashing salt from the@@ -18921,8 +20564,8 @@ at all times.The buffered data is bound to one open file and won\[aq]t be shared..PPThis flag is a upper limit for the used memory per open file.-The buffer will only use memory for data that is downloaded but not not-yet read.+The buffer will only use memory for data that is downloaded but not yet+read.If the buffer is empty, only a small amount of memory will be used..PPThe maximum memory used by rclone for buffering can be up to@@ -18984,7 +20627,8 @@ This cache flushing strategy is efficient and more relevant files arelikely to remain cached..PPThe \f[V]--vfs-cache-max-age\f[R] will evict files from the cache after-the set time since last access has passed.+the set time since last access has passed; it is based on access time,+not on when the file was first added to the cache.The default value of 1 hour will start evicting files from cache thathaven\[aq]t been accessed for 1 hour.When a cached file is accessed the 1 hour timer is reset to 0 and will@@ -19157,8 +20801,8 @@ Rclone reads \f[V]--vfs-read-chunk-streams\f[R] chunks of size\f[V]--vfs-read-chunk-size\f[R] concurrently.The size for each read will stay constant..PP-This improves performance performance massively on high latency links or-very high bandwidth links to high performance object stores.+This improves performance massively on high latency links or very high+bandwidth links to high performance object stores..PPSome experimentation will be needed to find the optimum values of\f[V]--vfs-read-chunk-size\f[R] and \f[V]--vfs-read-chunk-streams\f[R]@@ -19438,9 +21082,12 @@ This config generated must have this extra parameter.IP \[bu] 2\f[V]_root\f[R] - root to use for the backend.PP-And it may have this parameter+And it may have these parameters.IP \[bu] 2\f[V]_obscure\f[R] - comma separated strings for parameters to obscure+.IP \[bu] 2+\f[V]_secret_access_key\f[R] - the secret for S3 access key auth (see+below).PPIf password authentication was used by the client, input to the proxyprocess (on STDIN) would look similar to this:@@ -19449,7 +21096,8 @@ process (on STDIN) would look similar to this:\f[C]{\[dq]user\[dq]: \[dq]me\[dq],- \[dq]pass\[dq]: \[dq]mypassword\[dq]+ \[dq]pass\[dq]: \[dq]mypassword\[dq],+ \[dq]client_ip\[dq]: \[dq]192.168.1.1\[dq]}\f[R].fi@@ -19461,11 +21109,51 @@ process (on STDIN) would look similar to this:\f[C]{\[dq]user\[dq]: \[dq]me\[dq],- \[dq]public_key\[dq]: \[dq]AAAAB3NzaC1yc2EAAAADAQABAAABAQDuwESFdAe14hVS6omeyX7edc...JQdf\[dq]+ \[dq]public_key\[dq]: \[dq]AAAAB3NzaC1yc2EAAAADAQABAAABAQDuwESFdAe14hVS6omeyX7edc...JQdf\[dq],+ \[dq]client_ip\[dq]: \[dq]192.168.1.1\[dq]}\f[R].fi.PP+If the client authenticated with an S3 access key+(\f[V]rclone serve s3\f[R]), the client never sends its secret, only a+signature made with it, so the input contains just the access key ID as+the \f[V]user\f[R] with no \f[V]pass\f[R] or \f[V]public_key\f[R]:+.IP+.nf+\f[C]+{+ \[dq]user\[dq]: \[dq]AKIAIOSFODNN7EXAMPLE\[dq],+ \[dq]client_ip\[dq]: \[dq]192.168.1.1\[dq]+}+\f[R]+.fi+.PP+In this case the program must look up the secret access key for that+access key ID and return it in the \f[V]_secret_access_key\f[R] field of+the output.+Rclone then uses that secret to verify the signature on the request,+refusing the request if it does not match.+This means the proxy program is the source of truth for both the+credentials and the backend they map to.+If the program does not return \f[V]_secret_access_key\f[R] or returns+it empty the request is refused.+.PP+The program\[aq]s answer for an access key ID is cached (see below) but+is checked with the program again after 5 minutes even if the access key+ID is in constant use, so revoking an access key ID in the program takes+effect within 5 minutes.+A rotated secret takes effect on the first request signed with it.+.PP+The \f[V]client_ip\f[R] key holds the IP address the client connected+from, without a port number.+It can be used to restrict logins to certain networks, or to log+authentication attempts centrally.+It is omitted if the client has no IP address, for example when+connecting over a unix socket.+Note that if rclone is behind a reverse proxy this will be the address+of the reverse proxy and not the original client.+.PPAnd as an example return this on STDOUT.IP.nf@@ -19496,12 +21184,21 @@ the \f[V]user\f[R] be \f[V]user\[at]example.com\f[R] and then set theFor security you\[aq]d probably want to restrict the \f[V]host\f[R] to alimited list..PP-An internal cache of backends is keyed on the \f[V]user\f[R] and a hash-of the \f[V]pass\f[R] or \f[V]public_key\f[R].-This means that if a user\[aq]s password or public-key changes, or the-proxy returns different config parameters (eg a rotated-\f[V]api_key\f[R]), a fresh backend will be created on the next request-rather than the cached one being reused.+An internal cache of backends is keyed on the \f[V]user\f[R], the kind+of credential presented, a hash of the \f[V]pass\f[R] or+\f[V]public_key\f[R], and the \f[V]client_ip\f[R].+This means that if a user\[aq]s password or public-key changes, the+client connects from a new IP address, or the proxy returns different+config parameters (eg a rotated \f[V]api_key\f[R]), a fresh backend will+be created on the next request rather than the cached one being reused.+A backend created for a public key login is never reused for a password+login, so a public key cannot be sent as a password to skip the program.+.PP+A cached answer is checked with the program again after 5 minutes even+if the user is in constant use, so revoking a user in the program takes+effect within 5 minutes.+For \f[V]serve sftp\f[R] this applies to new logins - an existing+connection is not closed..PPThis can be used to build general purpose proxies to any kind of backendthat rclone supports.@@ -19927,7 +21624,7 @@ This command runs a series of uploads and downloads to the remote,measuring and printing the speed of each test using varying file sizesand numbers of files..PP-Test time can be innaccurate with small file caps and large files.+Test time can be inaccurate with small file caps and large files.As it uses the results of an initial test to determine how many files touse in each subsequent test..PP@@ -19999,13 +21696,19 @@ time instead of the current time.Times may be specified as one of:.IP \[bu] 2\[aq]YYMMDD\[aq] - e.g.-17.10.30+171030+.IP \[bu] 2+\[aq]YYYY-MM-DD\[aq] - e.g.+2006-01-02.IP \[bu] 2\[aq]YYYY-MM-DDTHH:MM:SS\[aq] - e.g.2006-01-02T15:04:05.IP \[bu] 2\[aq]YYYY-MM-DDTHH:MM:SS.SSS\[aq] - e.g.2006-01-02T15:04:05.123456789+.IP \[bu] 2+\[aq]YYYY-MM-DDTHH:MM:SSZ\[aq] - e.g.+2006-01-02T15:04:05Z.PPNote that value of \f[V]--timestamp\f[R] is in UTC.If you want local time then add the \f[V]--localtime\f[R] flag.@@ -20525,6 +22228,11 @@ This is mostly a problem on Windows, where the console traditionallyuses a non-Unicode character set - defined by the so-called \[dq]codepage\[dq]..PP+If you want to configure a remote using environment variables then its+name becomes part of the environment variable name, which in practice+restricts the characters you can use in it.+See the Config file section under Environment variables for details.+.PPDo not use single character names on Windows as it creates ambiguitywith Windows drives\[aq] names, e.g.: remote called \f[V]C\f[R] isindistinguishable from \f[V]C\f[R] drive.@@ -20579,6 +22287,8 @@ Not all global variables make sense to be overridden like this as theconfig is only applied during the remote creation.Here is a non exhaustive list of ones which might be useful:.IP \[bu] 2+\f[V]allow_https_downgrade\f[R]+.IP \[bu] 2\f[V]bind_addr\f[R].IP \[bu] 2\f[V]ca_cert\f[R]@@ -21011,6 +22721,143 @@ precedence if supplied in the metadata over reading the.PPHashes are not included in system metadata as there is a well definedway of reading those already.+.SS Performance+.PP+Rclone\[aq]s defaults are fairly conservative.+They are chosen to work on small machines, slow links and with cloud+providers which rate limit heavily.+If you have plenty of bandwidth, memory and a fast storage system then+rclone can usually be made to run a great deal faster by adjusting a few+flags.+.PP+The right values depend on the size of your files, your network and the+storage systems at each end, so treat the numbers below as starting+points.+Change one thing at a time and measure the result.+.SS Measuring+.PP+Use \f[V]rclone copy\f[R] with -P/--progress to watch the transfer rate+while you are experimenting.+For unattended runs log the stats instead, e.g.+\f[V]--stats 1m --stats-one-line --stats-log-level NOTICE\f[R].+.PP+A transfer is always limited by something: the network bandwidth, the+rate at which the storage system will accept requests, disk IO, CPU or+memory.+Tuning is working out which limit you are hitting and raising it until+you hit the next one.+Watch the network, disk, CPU and memory use of the machine rclone is+running on as well as rclone\[aq]s stats.+.SS Lots of files+.PP+The time taken to transfer small files is dominated by the round trip+time of the requests rather than by the bandwidth, so the way to make+them go faster is to do more of them at once.+.PP+--transfers sets the number of files transferred in parallel.+The default is 4.+Double this until the transfer rate stops increasing or the storage+system starts returning errors.+Values of 32 to 64 are common and values in the hundreds work well with+the large object storage systems (e.g.+s3, azureblob, googlecloudstorage, b2) when transferring lots of small+files.+.PP+--checkers sets the number of directory listings and file comparisons+done in parallel.+The default is 8.+Raising this speeds up syncs where most of the files are already present+on the destination and stops the checking being the bottleneck when+\f[V]--transfers\f[R] is large.+16 to 64 is a reasonable range.+.PP+Not all storage systems like lots of parallel requests.+If you see rate limiting errors (e.g.+HTTP 429 or 503 errors) or lots of low level retries in the log then you+have gone too far, so reduce \f[V]--transfers\f[R] and+\f[V]--checkers\f[R] or use --tpslimit to cap the request rate.+Local disks, especially spinning ones, and network file systems may get+slower rather than faster with too much parallelism.+.SS Large files+.PP+A single stream often won\[aq]t fill a fast link, so rclone can transfer+a large file as several chunks in parallel.+These are called multi-thread transfers and are used for files bigger+than --multi-thread-cutoff on backends which support them.+.PP+The number of chunks transferred in parallel is set by+--multi-thread-streams (default 4) and this works for all backends so+increase this first to make large files transfer faster.+.PP+The size of the chunks is set by the backend, usually with a+\f[V]--backend-chunk-size\f[R] flag, e.g.+\f[V]--s3-chunk-size\f[R].+Backends without a natural chunk size (e.g.+local, smb) use --multi-thread-chunk-size.+Larger chunks mean fewer requests which increases throughput on big+files.+See the docs for the backend you are using for details.+.PP+Chunked uploads are usually buffered in memory, so in the worst case+rclone will use+.IP+.nf+\f[C]+--transfers * upload concurrency * chunk size+\f[R]+.fi+.PP+of memory for buffers.+For example \f[V]--transfers 256\f[R] with an upload concurrency of 8+and a chunk size of 16 MiB could use up to 32 GiB.+When you raise \f[V]--transfers\f[R] to speed up small files you may+therefore need to lower the upload concurrency, or set+--max-buffer-memory which limits the total memory used for buffers and+makes transfers wait when the limit is reached.+.PP+--use-mmap makes rclone return buffer memory to the OS as soon as it has+finished with it which is helpful when using lots of transfers or large+chunks.+.SS A mixture of large and small files+.PP+By default rclone transfers files in roughly the order it finds them.+With a mixture of file sizes this means that at times all the transfers+are large files, which fill the bandwidth but make few requests, and at+other times all the transfers are small files, which make lots of+requests but use little of the bandwidth.+.PP+Using --order-by \f[V]size,mixed,50\f[R] makes half of the transfers+take the smallest files in the backlog and half of them take the+largest.+This keeps both large and small files moving at once so the bandwidth+and the request rate are both kept full for the whole of the run.+.PP+This works best if rclone has lots of files to choose from so use it+with a big --max-backlog, e.g.+\f[V]--max-backlog 2000000\f[R].+Each item in the backlog uses in the order of 1 KiB of memory.+.SS Doing less work+.PP+Fewer requests to the storage system means a faster transfer.+.IP \[bu] 2+If the source and destination are on the same remote then rclone will+use server-side copy where it is supported so the data doesn\[aq]t need+to be downloaded and uploaded again.+.IP \[bu] 2+--fast-list lists bucket-based remotes with fewer transactions at the+cost of using more memory.+This may be faster or slower than the default so it is worth testing.+.IP \[bu] 2+--no-traverse stops rclone listing the destination which helps when+copying a small number of files into a large destination.+.IP \[bu] 2+--size-only or --checksum avoid reading the modification time.+On some object storage systems (e.g.+s3, swift) this needs an extra request for each file.+Alternatively use --use-server-modtime with \f[V]--update\f[R].+.IP \[bu] 2+--no-check-dest skips checking the destination entirely.+Only use this if you know that none of the files are there already..SS Options.PPRclone has a number of options to control its behaviour.@@ -21098,8 +22945,8 @@ The following suffixes are valid:.IP \[bu] 2\f[V]y\f[R] - Years.PP-Examples: \[dq]10\[dq], \[dq]300ms\[dq], \[dq]-1.5h\[dq] or-\[dq]2h45m\[dq].+Examples: \[dq]10\[dq], \[dq]300ms\[dq], \[dq]-1.5h\[dq],+\[dq]2h45m\[dq] or \[dq]1d12h\[dq]..SS Size options.PPOptions that take a SizeSuffix parameter can be specified as an integer@@ -21113,6 +22960,21 @@ These are the binary units, e.g..PPSee also --human-readable..SS Main options+.SS --allow-https-downgrade+.PP+By default rclone refuses to follow a redirect from an+\f[V]https://\f[R] URL to a plain \f[V]http://\f[R] URL, as any+credentials or other headers in the request would then be sent+unencrypted, and the response could be altered by anyone on the network+path.+The request fails with an error which mentions this flag.+.PP+Set \f[V]--allow-https-downgrade\f[R] to follow such redirects, for+example when downloading from a mirror which redirects to plain HTTP.+Some backends still refuse to follow them, as doing so would send their+credentials in cleartext.+.PP+This option defaults to \f[V]false\f[R]..SS --backup-dir string.PPWhen using sync (https://rclone.org/commands/rclone_sync/),@@ -21270,7 +23132,7 @@ Is equivalent to this:.IP.nf\f[C]---bwlimit \[dq]Mon-00:00,512Mon-12:00,1M Tue-12:00,1M Wed-12:00,1M Thu-12:00,1M Fri-12:00,1M Sat-12:00,1M Sun-12:00,1M Sun-20:00,off\[dq]+--bwlimit \[dq]Mon-00:00,512 Mon-12:00,1M Tue-12:00,1M Wed-12:00,1M Thu-12:00,1M Fri-12:00,1M Sat-12:00,1M Sun-12:00,1M Sun-20:00,off\[dq]\f[R].fi.PP@@ -21333,7 +23195,8 @@ Each \f[V]--transfer\f[R] will use this much memory for buffering..PPWhen using \f[V]mount\f[R] or \f[V]cmount\f[R] each open file descriptorwill use this much memory for buffering.-See the mount (https://rclone.org/commands/rclone_mount/#file-buffering)+See the+mount (https://rclone.org/commands/rclone_mount/#vfs-file-buffering)documentation for more details..PPSet to \f[V]0\f[R] to disable the buffering for the minimum memory@@ -21647,7 +23510,7 @@ See the dedupe (https://rclone.org/commands/rclone_dedupe/) command formore information as to what these options mean..SS --default-time Time.PP-If a file or directory does have a modification time rclone can read+If a file or directory does not have a modification time rclone can readthen rclone will display this fixed time instead..PPThe default is \f[V]2000-01-01 00:00:00 UTC\f[R].@@ -21830,6 +23693,13 @@ downloads use \f[V]--header-download\f[R].This flag is supported for all HTTP based backends even those notsupported by \f[V]--header-upload\f[R] and \f[V]--header-download\f[R]so may be used as a workaround for those with care.+.PP+The headers are only sent to the host of the original request.+If the server redirects to another host (including a subdomain or a+different port), or redirects an \f[V]https\f[R] request to+\f[V]http\f[R], the headers are not sent to it, or to any further hop in+that redirect chain.+Rclone logs this at debug level (\f[V]-vv\f[R])..IP.nf\f[C]@@ -22042,6 +23912,12 @@ Note that on the local file system if you don\[aq]t useAnd if you do use \f[V]--inplace\f[R] you won\[aq]t be able to update inuse executables..PP+On macOS, if rclone clones a file (see+--local-no-clone (https://rclone.org/local/#local-no-clone)) and the+destination already exists, the existing file is replaced with a new+one, even with \f[V]--inplace\f[R], so hard links to it will be broken.+Use \f[V]--local-no-clone\f[R] to keep them.+.PPNote also that versions of rclone prior to v1.63.0 behave as if the\f[V]--inplace\f[R] flag is always supplied..SS -i, --interactive@@ -22103,7 +23979,7 @@ like symlinks under Windows).Ignored files won\[aq]t be copied, moved or deleted in a sync..PPIf you supply this flag then rclone will copy symbolic links from any-supported backend backend, and store them as text files, with a+supported backend, and store them as text files, with a\f[V].rclonelink\f[R] suffix in the destination..PPThe text file will contain the target of the symbolic link.@@ -22705,9 +24581,9 @@ overview (https://rclone.org/overview/#optional-features) as\f[V]MultithreadUpload\f[R].(They need to implement either the \f[V]OpenWriterAt\f[R] or\f[V]OpenChunkWriter\f[R] internal interfaces).-These include include, \f[V]local\f[R], \f[V]s3\f[R],-\f[V]azureblob\f[R], \f[V]b2\f[R], \f[V]oracleobjectstorage\f[R] and-\f[V]smb\f[R] at the time of writing.+These include \f[V]local\f[R], \f[V]s3\f[R], \f[V]azureblob\f[R],+\f[V]b2\f[R], \f[V]oracleobjectstorage\f[R] and \f[V]smb\f[R] at the+time of writing..PPOn the local disk, rclone preallocates the file (using\f[V]fallocate(FALLOC_FL_KEEP_SIZE)\f[R] on unix or@@ -22736,7 +24612,7 @@ In the worst case memory usage can be at maximum \f[V]--transfers\f[R] *\f[V]--multi-thread-chunk-size\f[R] * \f[V]--multi-thread-streams\f[R]or specifically for the s3 backend \f[V]--transfers\f[R] *\f[V]--s3-chunk-size\f[R] * \f[V]--s3-concurrency\f[R].-However you can use the the+However you can use the--max-buffer-memory (https://rclone.org/docs/#max-buffer-memory) flag tocontrol the maximum memory used here..PP@@ -22867,8 +24743,11 @@ The first part describes what aspect is being measured:\f[V]name\f[R] - order by the full path of the files.IP \[bu] 2\f[V]modtime\f[R] - order by the modification date of the files+.IP \[bu] 2+\f[V]pattern\f[R] - order by a list of path glob patterns.PP-This can have a modifier appended with a comma:+The \f[V]size\f[R], \f[V]name\f[R] and \f[V]modtime\f[R] comparisons can+have a modifier appended with a comma:.IP \[bu] 2\f[V]ascending\f[R] or \f[V]asc\f[R] - order so that the smallest (oroldest) is processed first@@ -22892,6 +24771,15 @@ will be processed continuously..PPIf no modifier is supplied then the order is \f[V]ascending\f[R]..PP+For \f[V]pattern\f[R], append one or more comma-separated patterns.+Patterns use the filter pattern+syntax (https://rclone.org/filtering/#patterns).+Files matching the first pattern are processed first, followed by files+matching the second pattern, and so on.+Files which do not match a pattern are processed last.+Within each group, files are ordered by full path.+Patterns cannot contain a comma.+.PPFor example.IP \[bu] 2\f[V]--order-by size,desc\f[R] - send the largest files first@@ -22900,6 +24788,9 @@ For example.IP \[bu] 2\f[V]--order-by name\f[R] - send the files with alphabetically by pathfirst+.IP \[bu] 2+\f[V]--order-by \[aq]pattern,*.tar.gz,*.md5\[aq]\f[R] - send archives+before checksum files.PPIf the \f[V]--order-by\f[R] flag is not supplied or it is supplied withan empty string then the default ordering will be used which is as@@ -22946,6 +24837,8 @@ If one of the arguments has a space in then enclose it in\f[V]\[dq]\f[R], if you want a literal \f[V]\[dq]\f[R] in an argumentthen enclose the argument in \f[V]\[dq]\f[R] and double the\f[V]\[dq]\f[R].+This includes the command itself: if the path to the executable contains+a space, it must be quoted too.See CSV encoding (https://godoc.org/encoding/csv) for more info..PPEg@@ -22955,6 +24848,7 @@ Eg--password-command \[dq]echo hello\[dq]--password-command \[aq]echo \[dq]hello with space\[dq]\[aq]--password-command \[aq]echo \[dq]hello with \[dq]\[dq]quotes\[dq]\[dq] and space\[dq]\[aq]+--password-command \[aq]\[dq]/path with a space/get-password.sh\[dq]\[aq]\f[R].fi.PP@@ -24177,9 +26071,13 @@ backend.To find the name of the environment variable, you need to set, take\f[V]RCLONE_CONFIG_\f[R] + name of remote + \f[V]_\f[R] + name of configfile option and make it all uppercase.-Note one implication here is the remote\[aq]s name must be convertible-into a valid environment variable name, so it can only contain letters,-digits, or the \f[V]_\f[R] (underscore) character.+Note that the name of the remote is only converted to uppercase -+characters such as \f[V]-\f[R] and \f[V].\f[R] are not changed to+\f[V]_\f[R] - so a remote called \f[V]my-remote\f[R] needs+\f[V]RCLONE_CONFIG_MY-REMOTE_TYPE\f[R].+Many shells only allow letters, digits and \f[V]_\f[R] (underscore) in+environment variable names, so it is best to use only those characters+in the name of a remote you want to configure this way..PPFor example, to configure an S3 remote named \f[V]mys3:\f[R] without aconfig file (using unix ways of setting environment variables):@@ -24489,7 +26387,7 @@ Rclone matching rules follow a glob style:* matches any sequence of non-separator (/) characters** matches any sequence of characters including / separators? matches any single non-separator (/) character-[ [ ! ] { character-range } ]+[ [ ! | \[ha] ] { character-range } ]character class (must be non-empty){ pattern-list }pattern alternatives@@ -25873,6 +27771,10 @@ open..IP \[bu] 2The URL already contains the username & password, in which case the GUIwill use those values and log you in automatically.+.PP+The rc address is passed to the GUI via the \f[V]url\f[R] query+parameter (the URL printed in the terminal); append+\f[V]?url=http://...\f[R] to point it at a different rc server..SS Security.PPIt\[aq]s important to think first about what rclone has access to and@@ -26086,10 +27988,20 @@ Default Off..PPSet the allowed Access-Control-Allow-Origin for rc requests..PP-Can be used with --rc-web-gui if the rclone is running on different IP-than the web-gui.+This enables cross-origin (CORS) requests from browsers on the origin+given, for example \f[V]--rc-allow-origin https://app.example.com\f[R].+Browser CORS preflight OPTIONS requests are answered by rclone without+authentication as browsers never send credentials with them.+All other requests, including other OPTIONS requests, must be+authenticated when authentication is enabled.+.PP+The \f[V]rclone gui\f[R] command sets this automatically to the origin+the GUI is served from if it is not set explicitly.+.PP+Setting this to \f[V]*\f[R] allows any origin which is a security risk+on an authenticated server..PP-Default is IP address on which rc is running.+Default is off (no CORS)..SS --rc-web-fetch-url.PPSet the URL to fetch the rclone-web-gui files from.@@ -26133,6 +28045,9 @@ use these methods.The alternative is to use \f[V]--rc-user\f[R] and \f[V]--rc-pass\f[R]and use these credentials in the request..PP+Without authorisation or this flag, POST request bodies are limited to+64 KiB.+.PPDefault Off..SS --rc-baseurl.PP@@ -26221,6 +28136,35 @@ Rather thanrclone rc operations/list --json \[aq]{\[dq]fs\[dq]: \[dq]/tmp\[dq], \[dq]remote\[dq]: \[dq]test\[dq], \[dq]opt\[dq]: {\[dq]showHash\[dq]: true}}\[aq]\f[R].fi+.SS Paths on Windows+.PP+Prefer forward slashes (\f[V]/\f[R]) in Windows paths passed to the rc+API, for example \f[V]C:/path1/path2\f[R].+If you use backslashes (\f[V]\[rs]\f[R]) in a JSON string, escape each+one as \f[V]\[rs]\[rs]\f[R].+These JSON objects contain equivalent Windows paths:+.IP+.nf+\f[C]+{\[dq]remote\[dq]: \[dq]C:/path1/path2\[dq]}+\f[R]+.fi+.IP+.nf+\f[C]+{\[dq]remote\[dq]: \[dq]C:\[rs]\[rs]path1\[rs]\[rs]path2\[dq]}+\f[R]+.fi+.PP+This applies to JSON request bodies and to JSON-encoded object+parameters, such as \f[V]parameters\f[R] in \f[V]config/create\f[R],+even when that JSON is sent in a query string or form data.+An unescaped backslash can cause a JSON parsing error or be interpreted+as an escape sequence, such as \f[V]\[rs]t\f[R] for a tab.+.PP+String parameters sent directly in a query string or form data, without+JSON encoding, do not need JSON escaping.+Use your HTTP client\[aq]s URL encoding for query strings and form data..SS Special parameters.PPThe rc interface supports some special parameters which apply to@@ -26426,6 +28370,18 @@ Or flat:If you wish to check the \f[V]_config\f[R] assignment has workedproperly then calling \f[V]options/local\f[R] will show what the valuegot set to.+.PP+Setting \f[V]LogLevel\f[R] sets the log level for the logs made by the+rc call only, for example to debug a single call.+.IP+.nf+\f[C]+\[dq]_config\[dq]:{\[dq]LogLevel\[dq]: \[dq]DEBUG\[dq]}+\f[R]+.fi+.PP+Note that not all of rclone\[aq]s logs support this yet.+In particular the backends log at the global log level..SS Setting filter flags with _filter.PPIf you wish to set filters for the duration of an rc call only then pass@@ -27425,6 +29381,7 @@ Returns the following values:\[dq]totalTransfers\[dq]: total number of transfers in the group,\[dq]transferTime\[dq] : total time spent on running jobs,\[dq]transfers\[dq]: number of transferred files,+ \[dq]updatedDirs\[dq]: number of directories updated (modtime or metadata set),\[dq]transferring\[dq]: an array of currently active file transfers:[{@@ -27744,8 +29701,6 @@ jobids - array of integer job ids (starting at 1 on each restart)runningIds - array of integer job ids that are running.IP \[bu] 2finishedIds - array of integer job ids that are finished-.PP-\f[B]Authentication is not required for this call.\f[R].SS job/status: Reads the status of the job ID.PPParameters:@@ -27779,8 +29734,6 @@ output - output of the job as would have been returned if calledsynchronously.IP \[bu] 2progress - output of the progress related to the underlying job-.PP-\f[B]Authentication is not required for this call.\f[R].SS job/stop: Stop the running job.PPParameters:@@ -28193,6 +30146,37 @@ instead:rclone rc --loopback operations/fsinfo fs=remote:\f[R].fi+.SS operations/getfile: Get a single file into a JSON response.+.PP+This takes the following parameters:+.IP \[bu] 2+fs - a remote name string e.g.+\[dq]drive:path/to/file\[dq] or \[dq]drive:path/to/dir\[dq]+.IP \[bu] 2+remote - (optional) a path within that remote e.g.+\[dq]file.txt\[dq]+.IP \[bu] 2+offset - (optional) start reading at offset N (or from end if negative)+(default 0)+.IP \[bu] 2+count - (optional) only read N bytes (default -1)+.IP \[bu] 2+head - (optional) only read the first N bytes (default 0)+.IP \[bu] 2+tail - (optional) only read the last N bytes (default 0)+.IP \[bu] 2+maxSize - (optional) maximum size limit in bytes (can only lower the+built-in maximum) (default 2097152 / 2MiB)+.IP \[bu] 2+base64 - (optional) if true, returns the file contents base64-encoded+(default false)+.PP+Returns:+.IP \[bu] 2+result - contents of the file as string or base64-encoded string+.PP+For streaming file content over HTTP with Range support, use the+--rc-serve flag instead..SS operations/hashsum: Produces a hashsum file for all the objects in the path..PPProduces a hash file for all the objects in the path using the hash@@ -28291,6 +30275,47 @@ $ rclone rc --loopback operations/hashsumfile fs=/ remote=/bin/bash hashType=MD5.PPSee the hashsum (https://rclone.org/commands/rclone_hashsum/) commandfor more information on the above.+.SS operations/index: Write static directory listings into a remote+.PP+This takes the following parameters:+.IP \[bu] 2+fs - a remote name string e.g.+\[dq]drive:\[dq]+.IP \[bu] 2+opt - a dictionary of options to control the indexing (optional)+.RS 2+.IP \[bu] 2+output - list of listings to write in each directory as+\[dq]NAME=FORMAT\[dq] (default [\[dq]index.html=html\[dq]])+.IP \[bu] 2+rules - which directories get listings, e.g.+{\[dq]ExcludeRule\[dq]: [\[dq]/private/**\[dq]]}+.IP \[bu] 2+maxDepth - only make listings this many directories deep (default -1 for+no limit)+.IP \[bu] 2+linkIndex - link to \[dq]dir/NAME\[dq] rather than \[dq]dir/\[dq]+.IP \[bu] 2+dirTime - how to work out directory times: \[dq]newest\[dq],+\[dq]dir\[dq] or \[dq]none\[dq]+.IP \[bu] 2+noModTime - don\[aq]t show modification times+.IP \[bu] 2+rewrite - write every listing even if it is unchanged+.IP \[bu] 2+changed - list of changed paths for a partial run, directories ending in+\[dq]/\[dq]+.IP \[bu] 2+changedFrom - list of files of changed paths, one per line+.IP \[bu] 2+changedCombined - list of files in the sync --combined report format+.IP \[bu] 2+changedMaxDirs - do a full run if a partial one would list more+directories than this+.RE+.PP+See the index (https://rclone.org/commands/rclone_index/) command for+more information on the above..SS operations/list: List the given remote and path in JSON format.PPThis takes the following parameters:@@ -28586,98 +30611,6 @@ And this sets NOTICE level logs (normal without -v)rclone rc options/set --json \[aq]{\[dq]main\[dq]: {\[dq]LogLevel\[dq]: \[dq]NOTICE\[dq]}}\[aq]\f[R].fi-.SS pluginsctl/addPlugin: Add a plugin using url-.PP-Used for adding a plugin to the webgui.-.PP-This takes the following parameters:-.IP \[bu] 2-url - http url of the github repo where the plugin is hosted-(http://github.com/rclone/rclone-webui-react).-.PP-Example:-.PP-rclone rc pluginsctl/addPlugin-.SS pluginsctl/getPluginsForType: Get plugins with type criteria-.PP-This shows all possible plugins by a mime type.-.PP-This takes the following parameters:-.IP \[bu] 2-type - supported mime type by a loaded plugin e.g.-(video/mp4, audio/mp3).-.IP \[bu] 2-pluginType - filter plugins based on their type e.g.-(DASHBOARD, FILE_HANDLER, TERMINAL).-.PP-Returns:-.IP \[bu] 2-loadedPlugins - list of current production plugins.-.IP \[bu] 2-testPlugins - list of temporarily loaded development plugins, usually-running on a different server.-.PP-Example:-.PP-rclone rc pluginsctl/getPluginsForType type=video/mp4-.SS pluginsctl/listPlugins: Get the list of currently loaded plugins-.PP-This allows you to get the currently enabled plugins and their details.-.PP-This takes no parameters and returns:-.IP \[bu] 2-loadedPlugins - list of current production plugins.-.IP \[bu] 2-testPlugins - list of temporarily loaded development plugins, usually-running on a different server.-.PP-E.g.-.PP-rclone rc pluginsctl/listPlugins-.SS pluginsctl/listTestPlugins: Show currently loaded test plugins-.PP-Allows listing of test plugins with the rclone.test set to true in-package.json of the plugin.-.PP-This takes no parameters and returns:-.IP \[bu] 2-loadedTestPlugins - list of currently available test plugins.-.PP-E.g.-.IP-.nf-\f[C]-rclone rc pluginsctl/listTestPlugins-\f[R]-.fi-.SS pluginsctl/removePlugin: Remove a loaded plugin-.PP-This allows you to remove a plugin using it\[aq]s name.-.PP-This takes parameters:-.IP \[bu] 2-name - name of the plugin in the format-\f[V]author\f[R]/\f[V]plugin_name\f[R].-.PP-E.g.-.PP-rclone rc pluginsctl/removePlugin name=rclone/video-plugin-.SS pluginsctl/removeTestPlugin: Remove a test plugin-.PP-This allows you to remove a plugin using it\[aq]s name.-.PP-This takes the following parameters:-.IP \[bu] 2-name - name of the plugin in the format-\f[V]author\f[R]/\f[V]plugin_name\f[R].-.PP-Example:-.IP-.nf-\f[C]-rclone rc pluginsctl/removeTestPlugin name=rclone/rclone-webui-react-\f[R]-.fi.SS rc/error: This returns an error.PPThis returns an error with the input as part of its error string.@@ -28890,6 +30823,11 @@ path2 (required) - (string) a remote directory string e.g..IP \[bu] 2dryRun - (bool) dry-run mode.IP \[bu] 2+maxDelete - (int) Safety check on maximum percentage of deleted files+allowed.+If exceeded, the bisync run will abort.+(default: 50%)+.IP \[bu] 2backupDir1 - (string) --backup-dir for Path1.Must be a non-overlapping path on the same remote..PD 0@@ -28981,6 +30919,12 @@ ignoreListingChecksum - (bool) Do not use checksums for listings (add.P.PD.IP \[bu] 2+maxDeleteRenamesAware - (bool) Exclude tracked renames from the+--max-delete safety check (requires --track-renames)+.PD 0+.P+.PD+.IP \[bu] 2maxLock - (Duration) Consider lock files older than this to be expired(default: 0 (never expire)) (minimum: 2m).PD 0@@ -29054,6 +30998,45 @@ dstFs - a remote name string e.g.\[dq]drive:dst\[dq] for the destination.IP \[bu] 2createEmptySrcDirs - create empty src directories on destination if set+.IP \[bu] 2+combined - make a combined report of changes (default false)+.IP \[bu] 2+missingOnSrc - report all files missing from the source (default false)+.IP \[bu] 2+missingOnDst - report all files missing from the destination (default+false)+.IP \[bu] 2+match - report all matching files (default false)+.IP \[bu] 2+differ - report all non-matching files (default false)+.IP \[bu] 2+error - report all files with errors (hashing or reading) (default+false)+.IP \[bu] 2+destAfter - report all files that exist on the destination post-copy+(default false)+.PP+Returns:+.IP \[bu] 2+combined - array of strings of combined report of changes+.IP \[bu] 2+missingOnSrc - array of strings of all files missing from the source+.IP \[bu] 2+missingOnDst - array of strings of all files missing from the+destination+.IP \[bu] 2+match - array of strings of all matching files+.IP \[bu] 2+differ - array of strings of all non-matching files+.IP \[bu] 2+error - array of strings of all files with errors (hashing or reading)+.IP \[bu] 2+destAfter - array of strings of all files that exist on the destination+post-copy+.PP+Each report is only returned if its parameter is set to true.+If the operation fails the reports are only available if it was run with+\f[V]_async\f[R], as part of the job output..PPSee the copy (https://rclone.org/commands/rclone_copy/) command for moreinformation on the above.@@ -29070,6 +31053,45 @@ dstFs - a remote name string e.g.createEmptySrcDirs - create empty src directories on destination if set.IP \[bu] 2deleteEmptySrcDirs - delete empty src directories if set+.IP \[bu] 2+combined - make a combined report of changes (default false)+.IP \[bu] 2+missingOnSrc - report all files missing from the source (default false)+.IP \[bu] 2+missingOnDst - report all files missing from the destination (default+false)+.IP \[bu] 2+match - report all matching files (default false)+.IP \[bu] 2+differ - report all non-matching files (default false)+.IP \[bu] 2+error - report all files with errors (hashing or reading) (default+false)+.IP \[bu] 2+destAfter - report all files that exist on the destination post-move+(default false)+.PP+Returns:+.IP \[bu] 2+combined - array of strings of combined report of changes+.IP \[bu] 2+missingOnSrc - array of strings of all files missing from the source+.IP \[bu] 2+missingOnDst - array of strings of all files missing from the+destination+.IP \[bu] 2+match - array of strings of all matching files+.IP \[bu] 2+differ - array of strings of all non-matching files+.IP \[bu] 2+error - array of strings of all files with errors (hashing or reading)+.IP \[bu] 2+destAfter - array of strings of all files that exist on the destination+post-move+.PP+Each report is only returned if its parameter is set to true.+If the operation fails the reports are only available if it was run with+\f[V]_async\f[R], as part of the job output..PPSee the move (https://rclone.org/commands/rclone_move/) command for moreinformation on the above.@@ -29084,6 +31106,45 @@ dstFs - a remote name string e.g.\[dq]drive:dst\[dq] for the destination.IP \[bu] 2createEmptySrcDirs - create empty src directories on destination if set+.IP \[bu] 2+combined - make a combined report of changes (default false)+.IP \[bu] 2+missingOnSrc - report all files missing from the source (default false)+.IP \[bu] 2+missingOnDst - report all files missing from the destination (default+false)+.IP \[bu] 2+match - report all matching files (default false)+.IP \[bu] 2+differ - report all non-matching files (default false)+.IP \[bu] 2+error - report all files with errors (hashing or reading) (default+false)+.IP \[bu] 2+destAfter - report all files that exist on the destination post-sync+(default false)+.PP+Returns:+.IP \[bu] 2+combined - array of strings of combined report of changes+.IP \[bu] 2+missingOnSrc - array of strings of all files missing from the source+.IP \[bu] 2+missingOnDst - array of strings of all files missing from the+destination+.IP \[bu] 2+match - array of strings of all matching files+.IP \[bu] 2+differ - array of strings of all non-matching files+.IP \[bu] 2+error - array of strings of all files with errors (hashing or reading)+.IP \[bu] 2+destAfter - array of strings of all files that exist on the destination+post-sync+.PP+Each report is only returned if its parameter is set to true.+If the operation fails the reports are only available if it was run with+\f[V]_async\f[R], as part of the job output..PPSee the sync (https://rclone.org/commands/rclone_sync/) command for moreinformation on the above.@@ -29358,9 +31419,12 @@ status - the HTTP status codepath - the path of the call.SS CORS.PP-The sever implements basic CORS support and allows all origins for that.-The response to a preflight OPTIONS request will echo the requested-\[dq]Access-Control-Request-Headers\[dq] back.+The server implements basic CORS support when+\f[V]--rc-allow-origin\f[R] is set.+Browser CORS preflight OPTIONS requests are then answered by rclone+without authentication with the \f[V]Access-Control-*\f[R] headers.+Without \f[V]--rc-allow-origin\f[R] no CORS headers are sent and a+preflight request is authenticated like any other request..SS Using POST with URL parameters only.IP.nf@@ -30922,6 +32986,7 @@ Flags for general networking and HTTP stuff..IP.nf\f[C]+ --allow-https-downgrade Follow redirects from HTTPS to HTTP (insecure)--bind string Local address to bind to for outgoing connections, IPv4, IPv6 or name--bwlimit BwTimetable Bandwidth limit in KiB/s, or use suffix B|K|M|G|T|P or a full timetable--bwlimit-file BwTimetable Bandwidth limit per file in KiB/s, or use suffix B|K|M|G|T|P or a full timetable@@ -30945,7 +33010,7 @@ Flags for general networking and HTTP stuff.--tpslimit float Limit HTTP transactions per second to this--tpslimit-burst int Max burst of transactions for --tpslimit (default 1)--use-cookies Enable session cookiejar- --user-agent string Set the user-agent to a specified string (default \[dq]rclone/v1.75.0\[dq])+ --user-agent string Set the user-agent to a specified string (default \[dq]rclone/v1.76.0\[dq])\f[R].fi.SS Performance@@ -31126,11 +33191,6 @@ Flags to control the Remote Control API.--rc-template string User-specified template--rc-user string User name for authentication--rc-user-from-header string User name from a defined HTTP header- --rc-web-fetch-url string URL to fetch the releases for webgui (default \[dq]https://api.github.com/repos/rclone/rclone-webui-react/releases/latest\[dq])- --rc-web-gui Launch WebGUI on localhost- --rc-web-gui-force-update Force update to latest version of web gui- --rc-web-gui-no-open-browser Don\[aq]t open the browser automatically- --rc-web-gui-update Check and update to latest version of web gui\f[R].fi.SS Metrics@@ -31192,8 +33252,10 @@ Backend-only flags (these can be set in the config file also).--azureblob-encoding Encoding The encoding for the backend (default Slash,BackSlash,Del,Ctl,RightPeriod,InvalidUtf8)--azureblob-endpoint string Endpoint for the service--azureblob-env-auth Read credentials from runtime (environment variables, CLI or MSI)+ --azureblob-hns Tristate Set if the storage account has a hierarchical namespace (default unset)--azureblob-key string Storage Account Shared Key--azureblob-list-chunk int Size of blob list (default 5000)+ --azureblob-list-parallelism int Number of parallel shards to list a directory with--azureblob-msi-client-id string Object ID of the user-assigned MSI to use, if any--azureblob-msi-mi-res-id string Azure resource ID of the user-assigned MSI to use, if any--azureblob-msi-object-id string Object ID of the user-assigned MSI to use, if any@@ -31206,6 +33268,7 @@ Backend-only flags (these can be set in the config file also).--azureblob-tenant string ID of the service principal\[aq]s tenant. Also called its directory ID--azureblob-upload-concurrency int Concurrency for multipart uploads (default 16)--azureblob-upload-cutoff string Cutoff for switching to chunked upload (<= 256 MiB) (deprecated)+ --azureblob-use-arrow-list Use the Apache Arrow listing format--azureblob-use-az Use Azure CLI tool az for authentication--azureblob-use-copy-blob Whether to use the Copy Blob API when copying to the same storage account (default true)--azureblob-use-emulator Uses local storage emulator if provided as \[aq]true\[aq]@@ -31338,6 +33401,12 @@ Backend-only flags (these can be set in the config file also).--doi-doi string The DOI or the doi.org URL--doi-doi-resolver-api-url string The URL of the DOI resolver API to use--doi-provider string DOI provider+ --dosya-api-key string Your API Key, get it from https://dosya.dev/profile under \[dq]API keys\[dq]+ --dosya-api-url string API base URL (default \[dq]https://api.dosya.dev\[dq])+ --dosya-description string Description of the remote+ --dosya-encoding Encoding The encoding for the backend (default Slash,LtGt,DoubleQuote,SingleQuote,BackSlash,Del,Ctl,LeftSpace,RightSpace,InvalidUtf8,Dot)+ --dosya-upload-concurrency int Concurrency for multipart uploads (default 4)+ --dosya-workspace-id string Your workspace ID--drime-access-token string API Access token--drime-chunk-size SizeSuffix Chunk size to use for uploading (default 5Mi)--drime-description string Description of the remote@@ -31560,6 +33629,8 @@ Backend-only flags (these can be set in the config file also).--http-headers CommaSepList Set HTTP headers for all transactions--http-no-escape Do not escape URL metacharacters in path names--http-no-head Don\[aq]t use HEAD requests+ --http-no-head-size SizeSuffix Set the reported file size for --http-no-head (default off)+ --http-no-head-time Time Set a default modification time for --http-no-head (default off)--http-no-slash Set this if the site doesn\[aq]t end directories with /--http-url string URL of HTTP host to connect to--huaweidrive-auth-url string Auth server URL@@ -31619,6 +33690,17 @@ Backend-only flags (these can be set in the config file also).--jottacloud-token-url string Token server url--jottacloud-trashed-only Only show files that are in the trash--jottacloud-upload-resume-limit SizeSuffix Files bigger than this can be resumed if the upload fail\[aq]s (default 10Mi)+ --kdrive-auth-url string Auth server URL+ --kdrive-chunk-size SizeSuffix Files above this size are uploaded in chunks of this size (default 20Mi)+ --kdrive-client-id string OAuth Client Id+ --kdrive-description string Description of the remote+ --kdrive-drive-id string ID of the kDrive to use+ --kdrive-encoding Encoding The encoding for the backend (default Slash,Del,Ctl,LeftSpace,RightSpace,InvalidUtf8,Dot)+ --kdrive-endpoint string The API endpoint to use+ --kdrive-root-folder-id string The folder to use as root+ --kdrive-token string OAuth Access Token as a JSON blob+ --kdrive-token-url string Token server url+ --kdrive-upload-concurrency int Concurrency for chunked uploads (default 4)--koofr-description string Description of the remote--koofr-encoding Encoding The encoding for the backend (default Slash,BackSlash,Del,Ctl,InvalidUtf8,Dot)--koofr-endpoint string The Koofr API endpoint to use@@ -31704,6 +33786,7 @@ Backend-only flags (these can be set in the config file also).--onedrive-root-folder-id string ID of the root folder--onedrive-server-side-across-configs Deprecated: use --server-side-across-configs instead--onedrive-tenant string ID of the service principal\[aq]s tenant. Also called its directory ID+ --onedrive-tenant-api-version string The SharePoint API version to use with tenant_url (default \[dq]v2.0\[dq])--onedrive-tenant-url string The tenant URL for non-admin OneDrive access--onedrive-token string OAuth Access Token as a JSON blob--onedrive-token-url string Token server url@@ -31718,7 +33801,7 @@ Backend-only flags (these can be set in the config file also).--oos-decompress If set this will decompress gzip encoded objects--oos-description string Description of the remote--oos-disable-checksum Don\[aq]t store MD5 checksum with object metadata- --oos-encoding Encoding The encoding for the backend (default Slash,InvalidUtf8,Dot)+ --oos-encoding Encoding The encoding for the backend (default Slash,CrLf,InvalidUtf8,Dot)--oos-endpoint string Endpoint for Object storage API--oos-leave-parts-on-error If true avoid calling abort upload on a failure, leaving all successfully uploaded parts for manual recovery--oos-max-upload-parts int Maximum number of parts in a multipart upload (default 10000)@@ -31913,7 +33996,7 @@ Backend-only flags (these can be set in the config file also).--sftp-disable-concurrent-reads If set don\[aq]t use concurrent reads--sftp-disable-concurrent-writes If set don\[aq]t use concurrent writes--sftp-disable-hashcheck Disable the execution of SSH commands to determine if remote file hashing is available- --sftp-encoding Encoding The encoding for the backend (default Slash,Del,Ctl,Dot)+ --sftp-encoding Encoding The encoding for the backend (default Slash,Dot)--sftp-hashes CommaSepList Comma separated list of supported checksum types--sftp-host string SSH host to connect to--sftp-host-key-algorithms SpaceSepList Space separated list of host key algorithms, ordered by preference@@ -31928,6 +34011,7 @@ Backend-only flags (these can be set in the config file also).--sftp-known-hosts-file string Optional path to known_hosts file--sftp-macs SpaceSepList Space separated list of MACs (message authentication code) algorithms, ordered by preference--sftp-md5sum-command string The command used to read MD5 hashes+ --sftp-multithread-upload Set this to use multi-thread (multi-connection) uploads--sftp-pass string SSH password, leave blank to use ssh-agent (obscured)--sftp-path-override string Override path used by SSH shell commands--sftp-pin-host-key Pin the server host key on first connection (Trust On First Use)@@ -31990,7 +34074,8 @@ Backend-only flags (these can be set in the config file also).--smb-port int SMB port number (default 445)--smb-spn string Service principal name--smb-use-kerberos Use Kerberos authentication- --smb-user string SMB username (default \[dq]$USER\[dq])+ --smb-user string SMB username+ --smb-workstation string Workstation name sent in NTLM authentication--storj-access-grant string Access grant--storj-api-key string API key--storj-description string Description of the remote@@ -32065,6 +34150,7 @@ Backend-only flags (these can be set in the config file also).--webdav-url string URL of http host to connect to--webdav-user string User name--webdav-vendor string Name of the WebDAV site/service/software you are using+ --yandex-app-folder Use the application folder as the root--yandex-auth-url string Auth server URL--yandex-client-credentials Use client credentials OAuth flow--yandex-client-id string OAuth Client Id@@ -32969,6 +35055,7 @@ Optional Flags:--force Bypass --max-delete safety check and run the sync. Consider using with --verbose-h, --help help for bisync--ignore-listing-checksum Do not use checksums for listings (add --ignore-checksum to additionally skip post-copy checksum checks)+ --max-delete-renames-aware Exclude tracked renames from the --max-delete safety check (requires --track-renames)--max-lock Duration Consider lock files older than this to be expired (default: 0 (never expire)) (minimum: 2m) (default 0s)--no-cleanup Retain working files (useful for troubleshooting and testing).--no-slow-hash Ignore listing checksums only on backends where they are slow@@ -33049,7 +35136,7 @@ symmetrical (https://github.com/rclone/rclone/issues/5681#issuecomment-938761815and more symmetrical options can be specified with the\f[V]--resync-mode\f[R] flag.)Carefully evaluate deltas using---dry-run (https://rclone.org/flags/#non-backend-flags).+--dry-run (https://rclone.org/docs/#n-dry-run)..PPFor a resync run, one of the paths may be empty (no files in the pathtree).@@ -33441,6 +35528,30 @@ To force the sync, either set a different delete percentage limit, e.g.\f[V]--force\f[R] to bypass the check..PPAlso see the all files changed check.+.SS --max-delete-renames-aware+.PP+When \f[V]--track-renames\f[R] is in use,+\f[V]--max-delete-renames-aware\f[R] can exclude files that are+guaranteed to match as tracked renames from the deletion count.+If the initial deletion count exceeds \f[V]--max-delete\f[R], bisync+performs a rename matching preflight before starting either sync+direction.+The run still aborts if the remaining unmatched deletions exceed the+limit.+.PP+The preflight uses the selected+\f[V]--track-renames-strategy\f[R] (https://rclone.org/docs/#track-renames-strategy)+and the attributes stored in the normal bisync listings.+A hash strategy forces both listings to use a common hash and is+incompatible with \f[V]--ignore-listing-checksum\f[R].+The option only exempts matches when the destination supports+server-side moves or copies and the selected strategy is supported by+both paths.+It requires \f[V]--track-renames\f[R] and is disabled by default because+hashing may add listing time.+\f[V]--force\f[R] bypasses both the safety check and this preflight.+The preflight is also disabled during \f[V]--resync\f[R], when rename+tracking is unavailable..SS --filters-file.PPBy using rclone filter features you can exclude file types or directory@@ -33697,8 +35808,8 @@ This is normal, and any differences will be detected and synced on thenext run..PPFor a robust integrity check of the current state of the remotes (as-opposed to just their listing snapshots), consider using \f[V]check\f[R]-(or+opposed to just their listing snapshots), consider using+\f[V]check\f[R] (https://rclone.org/commands/rclone_check/) (or\f[V]cryptcheck\f[R] (https://rclone.org/commands/rclone_cryptcheck/),if at least one path is a \f[V]crypt\f[R] remote) instead of\f[V]--check-sync\f[R], keeping in mind that differences are expected if@@ -34124,7 +36235,8 @@ resulted in unnecessary renames and duplicates.Now, when bisync comes to a file that it wants to rename (because it isnew/changed on both sides), it first checks whether the Path1 and Path2versions are currently \f[I]identical\f[R] (using the same underlying-function as \f[V]check\f[R].)+function as+\f[V]check\f[R] (https://rclone.org/commands/rclone_check/).)If bisync concludes that the files are identical, it will skip them andmove on.Otherwise, it will create renamed duplicates, as before.@@ -34270,24 +36382,38 @@ The following backends have known issues that need more investigation:.IP \[bu] 2\f[V]TestBisyncRemoteLocal/extended_filenames\f[R] (https://pub.rclone.org/integration-tests/current/huaweidrive-cmd.bisync-TestHuaweiDrive-1.txt).IP \[bu] 2-\f[V]TestBisyncRemoteLocal/normalization\f[R] (https://pub.rclone.org/integration-tests/current/huaweidrive-cmd.bisync-TestHuaweiDrive-1.txt)-.IP \[bu] 2\f[V]TestBisyncLocalRemote/ext_paths\f[R] (https://pub.rclone.org/integration-tests/current/huaweidrive-cmd.bisync-TestHuaweiDrive-1.txt).IP \[bu] 2\f[V]TestBisyncLocalRemote/extended_filenames\f[R] (https://pub.rclone.org/integration-tests/current/huaweidrive-cmd.bisync-TestHuaweiDrive-1.txt).IP \[bu] 2-4 more (https://pub.rclone.org/integration-tests/current/)+\f[V]TestBisyncLocalRemote/normalization\f[R] (https://pub.rclone.org/integration-tests/current/huaweidrive-cmd.bisync-TestHuaweiDrive-1.txt)+.IP \[bu] 2+3 more (https://pub.rclone.org/integration-tests/current/).RE.IP \[bu] 2\f[V]TestPcloud\f[R] (\f[V]pcloud\f[R]).RS 2.IP \[bu] 2-\f[V]TestBisyncRemoteRemote/check_access\f[R] (https://pub.rclone.org/integration-tests/current/pcloud-cmd.bisync-TestPcloud-1.txt)+\f[V]TestBisyncRemoteRemote/extended_filenames\f[R] (https://pub.rclone.org/integration-tests/current/pcloud-cmd.bisync-TestPcloud-1.txt)+.RE+.IP \[bu] 2+\f[V]TestYandex\f[R] (\f[V]yandex\f[R])+.RS 2+.IP \[bu] 2+\f[V]TestBisyncRemoteLocal/all_changed\f[R] (https://pub.rclone.org/integration-tests/current/yandex-cmd.bisync-TestYandex-1.txt)+.IP \[bu] 2+\f[V]TestBisyncRemoteLocal/backupdir\f[R] (https://pub.rclone.org/integration-tests/current/yandex-cmd.bisync-TestYandex-1.txt).IP \[bu] 2-\f[V]TestBisyncRemoteRemote/rmdirs\f[R] (https://pub.rclone.org/integration-tests/current/pcloud-cmd.bisync-TestPcloud-1.txt)+\f[V]TestBisyncRemoteLocal/basic\f[R] (https://pub.rclone.org/integration-tests/current/yandex-cmd.bisync-TestYandex-1.txt)+.IP \[bu] 2+\f[V]TestBisyncRemoteLocal/changes\f[R] (https://pub.rclone.org/integration-tests/current/yandex-cmd.bisync-TestYandex-1.txt)+.IP \[bu] 2+\f[V]TestBisyncRemoteLocal/check_access\f[R] (https://pub.rclone.org/integration-tests/current/yandex-cmd.bisync-TestYandex-1.txt)+.IP \[bu] 2+94 more (https://pub.rclone.org/integration-tests/current/).RE.IP \[bu] 2-Updated: 2026-07-31-010017+Updated: 2026-11-06-010024.PPThe following backends either have not been tested recently or haveknown issues that are deemed unfixable for the time being:@@ -34316,6 +36442,12 @@ known issues that are deemed unfixable for the time being:.IP \[bu] 2\f[V]TestJottacloud\f[R] (\f[V]jottacloud\f[R]).IP \[bu] 2+\f[V]TestKdrive\f[R] (\f[V]kdrive\f[R])+.RS 2+.IP \[bu] 2+\f[V]TestBisyncRemoteRemote/normalization\f[R]+.RE+.IP \[bu] 2\f[V]TestLinkbox\f[R] (\f[V]linkbox\f[R]).IP \[bu] 2\f[V]TestMailru\f[R] (\f[V]mailru\f[R])@@ -34340,9 +36472,11 @@ known issues that are deemed unfixable for the time being:.RS 2.IP \[bu] 2\f[V]TestBisyncRemoteRemote/extended_filenames\f[R]-.RE.IP \[bu] 2-\f[V]TestS3Rclone\f[R] (\f[V]s3\f[R])+\f[V]TestBisyncRemoteLocal/extended_filenames\f[R]+.IP \[bu] 2+\f[V]TestBisyncLocalRemote/extended_filenames\f[R]+.RE.IP \[bu] 2\f[V]TestSFTPRsyncNet\f[R] (\f[V]sftp\f[R]).IP \[bu] 2@@ -34443,11 +36577,12 @@ directory is to rename it to the same name on both sides.required after doing so, as bisync will automatically detect that Path1and Path2 are in agreement.).PP-Note that although the flag --track-renames ensures that renamed/moved-files won\[aq]t be deleted and uploaded again, they are still counted as-deleted files for purposes of the --max-delete flag (as this check-happens before the rename detection operation).-See this issue (https://github.com/rclone/rclone/issues/8685).+By default, renamed or moved files are still counted as deleted files+for purposes of \f[V]--max-delete\f[R], because the safety check happens+before rename detection.+Use \f[V]--max-delete-renames-aware\f[R] together with+\f[V]--track-renames\f[R] to run a rename preflight and exclude+guaranteed matches from this count..SS \f[V]--fast-list\f[R] used by default.PPUnlike most other rclone commands, bisync uses@@ -35420,151 +37555,11 @@ Also note a number of academic publications by BenjaminPierce (http://www.cis.upenn.edu/%7Ebcpierce/papers/index.shtml#File%20Synchronization)about \f[I]Unison\f[R] and synchronization in general..SS Changelog-.SS \f[V]v1.74.2\f[R]-.IP \[bu] 2-Fixed an issue causing \f[V]--conflict-loser pathname\f[R] to produce-unexpected behavior if using a non-default \f[V]--conflict-resolve\f[R]-value.-.SS \f[V]v1.74\f[R]-.IP \[bu] 2-Added several missing \f[V]rc\f[R] parameters.-.IP \[bu] 2-Optional \f[V]rc\f[R] parameters are now truly optional.-.IP \[bu] 2-\f[V]rc\f[R] output now provides more structured information.-.SS \f[V]v1.71\f[R]-.IP \[bu] 2-\f[V]bisync\f[R] is now officially released from beta.-.IP \[bu] 2-Fixed an issue causing hash type to be set incorrectly if-\f[V]path2\f[R] had slow hashes and \f[V]--no-slow-hash\f[R] or-\f[V]--slow-hash-sync-only\f[R] was in use.-.IP \[bu] 2-Fixed an issue causing errors when running concurrent bisync runs-through the \f[V]rc\f[R].-.SS \f[V]v1.69.1\f[R]-.IP \[bu] 2-Fixed an issue causing listings to not capture concurrent modifications-under certain conditions-.SS \f[V]v1.68\f[R]-.IP \[bu] 2-Fixed an issue affecting backends that round modtimes to a lower-precision.-.SS \f[V]v1.67\f[R]-.IP \[bu] 2-Added integration tests against all backends.-.SS \f[V]v1.66\f[R]-.IP \[bu] 2-Copies and deletes are now handled in one operation instead of two-.IP \[bu] 2-\f[V]--track-renames\f[R] and \f[V]--backup-dir\f[R] are now supported-.IP \[bu] 2-Partial uploads known issue on-\f[V]local\f[R]/\f[V]ftp\f[R]/\f[V]sftp\f[R] has been resolved (unless-using \f[V]--inplace\f[R])-.IP \[bu] 2-Final listings are now generated from sync results, to avoid needing to-re-list-.IP \[bu] 2-Bisync is now much more resilient to changes that happen during a bisync-run, and far less prone to critical errors / undetected changes-.IP \[bu] 2-Bisync is now capable of rolling a file listing back in cases of-uncertainty, essentially marking the file as needing to be rechecked-next time.-.IP \[bu] 2-A few basic terminal colors are now supported, controllable with-\f[V]--color\f[R] (https://rclone.org/docs/#color)-(\f[V]AUTO\f[R]|\f[V]NEVER\f[R]|\f[V]ALWAYS\f[R])-.IP \[bu] 2-Initial listing snapshots of Path1 and Path2 are now generated-concurrently, using the same \[dq]march\[dq] infrastructure as-\f[V]check\f[R] and \f[V]sync\f[R], for performance improvements and-less risk of-error (https://forum.rclone.org/t/bisync-bugs-and-feature-requests/37636#:~:text=4.%20Listings%20should%20alternate%20between%20paths%20to%20minimize%20errors).-.IP \[bu] 2-Fixed handling of unicode normalization and case insensitivity, support-for \f[V]--fix-case\f[R] (https://rclone.org/docs/#fix-case),-\f[V]--ignore-case-sync\f[R],-\f[V]--no-unicode-normalization\f[R] (https://rclone.org/docs/#no-unicode-normalization)-.IP \[bu] 2-\f[V]--resync\f[R] is now much more efficient (especially for users of-\f[V]--create-empty-src-dirs\f[R])-.IP \[bu] 2-Google Docs (and other files of unknown size) are now supported (with-the same options as in \f[V]sync\f[R])-.IP \[bu] 2-Equality checks before a sync conflict rename now fall back to-\f[V]cryptcheck\f[R] (when possible) or \f[V]--download\f[R], instead of-of \f[V]--size-only\f[R], when \f[V]check\f[R] is not available.-.IP \[bu] 2-Bisync no longer fails to find the correct listing file when configs are-overridden with backend-specific flags.-.IP \[bu] 2-Bisync now fully supports comparing based on any combination of size,-modtime, and checksum, lifting the prior restriction on backends without-modtime support.-.IP \[bu] 2-Bisync now supports a \[dq]Graceful Shutdown\[dq] mode to cleanly cancel-a run early without requiring \f[V]--resync\f[R].-.IP \[bu] 2-New \f[V]--recover\f[R] flag allows robust recovery in the event of-interruptions, without requiring \f[V]--resync\f[R].-.IP \[bu] 2-A new \f[V]--max-lock\f[R] setting allows lock files to automatically-renew and expire, for better automatic recovery when a run is-interrupted.-.IP \[bu] 2-Bisync now supports auto-resolving sync conflicts and customizing rename-behavior with new \f[V]--conflict-resolve\f[R],-\f[V]--conflict-loser\f[R], and \f[V]--conflict-suffix\f[R] flags.-.IP \[bu] 2-A new \f[V]--resync-mode\f[R] flag allows more control over which-version of a file gets kept during a \f[V]--resync\f[R].-.IP \[bu] 2-Bisync now supports-\f[V]--retries\f[R] (https://rclone.org/docs/#retries-int) and-\f[V]--retries-sleep\f[R] (when \f[V]--resilient\f[R] is set.)-.SS \f[V]v1.64\f[R]-.IP \[bu] 2-Fixed an-issue (https://forum.rclone.org/t/bisync-bugs-and-feature-requests/37636#:~:text=1.%20Dry%20runs%20are%20not%20completely%20dry)-causing dry runs to inadvertently commit filter changes-.IP \[bu] 2-Fixed an-issue (https://forum.rclone.org/t/bisync-bugs-and-feature-requests/37636#:~:text=2.%20%2D%2Dresync%20deletes%20data%2C%20contrary%20to%20docs)-causing \f[V]--resync\f[R] to erroneously delete empty folders and-duplicate files unique to Path2-.IP \[bu] 2-\f[V]--check-access\f[R] is now enforced during \f[V]--resync\f[R],-preventing data loss in certain user error-scenarios (https://forum.rclone.org/t/bisync-bugs-and-feature-requests/37636#:~:text=%2D%2Dcheck%2Daccess%20doesn%27t%20always%20fail%20when%20it%20should)-.IP \[bu] 2-Fixed an-issue (https://forum.rclone.org/t/bisync-bugs-and-feature-requests/37636#:~:text=5.%20Bisync%20reads%20files%20in%20excluded%20directories%20during%20delete%20operations)-causing bisync to consider more files than necessary due to overbroad-filters during delete operations-.IP \[bu] 2-Improved detection of false positive change-conflicts (https://forum.rclone.org/t/bisync-bugs-and-feature-requests/37636#:~:text=1.%20Identical%20files%20should%20be%20left%20alone%2C%20even%20if%20new/newer/changed%20on%20both%20sides)-(identical files are now left alone instead of renamed)-.IP \[bu] 2-Added support for-\f[V]--create-empty-src-dirs\f[R] (https://forum.rclone.org/t/bisync-bugs-and-feature-requests/37636#:~:text=3.%20Bisync%20should%20create/delete%20empty%20directories%20as%20sync%20does%2C%20when%20%2D%2Dcreate%2Dempty%2Dsrc%2Ddirs%20is%20passed)-.IP \[bu] 2-Added experimental \f[V]--resilient\f[R] mode to allow recovery from-self-correctable-errors (https://forum.rclone.org/t/bisync-bugs-and-feature-requests/37636#:~:text=2.%20Bisync%20should%20be%20more%20resilient%20to%20self%2Dcorrectable%20errors)-.IP \[bu] 2-Added new \f[V]--ignore-listing-checksum\f[R]-flag (https://forum.rclone.org/t/bisync-bugs-and-feature-requests/37636#:~:text=6.%20%2D%2Dignore%2Dchecksum%20should%20be%20split%20into%20two%20flags%20for%20separate%20purposes)-to distinguish from \f[V]--ignore-checksum\f[R]-.IP \[bu] 2-Performance-improvements (https://forum.rclone.org/t/bisync-bugs-and-feature-requests/37636#:~:text=6.%20Deletes%20take%20several%20times%20longer%20than%20copies)-for large remotes-.IP \[bu] 2-Documentation and testing improvements+.PP+Changes to bisync are listed in the rclone+changelog (https://rclone.org/changelog/).+The changelog this page kept until \f[V]v1.76\f[R] is archived on+GitHub (https://github.com/rclone/rclone/blob/0891ce30606c726990c98d70812d31b555782791/docs/content/bisync.md?plain=1#L1934-L2061)..SH Release signing.PPThe hashes of the binary artefacts of the rclone release are signed with@@ -35707,10 +37702,20 @@ Repeat for \f[V]MD5SUMS\f[R] and \f[V]SHA1SUMS\f[R] if desired..PPNow that we know the signatures on the hashes are OK we can verify thebinaries match the hashes, completing the verification.+.PP+Extract the signed data with:.IP.nf\f[C]-$ sha256sum -c SHA256SUMS 2>&1 | grep OK+$ gpg --output SHA256SUMS.signed --decrypt SHA256SUMS+\f[R]+.fi+.PP+Then check the binaries:+.IP+.nf+\f[C]+$ sha256sum -c --ignore-missing SHA256SUMS.signedrclone-v1.63.1-windows-amd64.zip: OK\f[R].fi@@ -35719,11 +37724,7 @@ Or do the check with rclone.IP.nf\f[C]-$ rclone hashsum sha256 -C SHA256SUMS rclone-v1.63.1-windows-amd64.zip-2023/09/11 10:53:58 NOTICE: SHA256SUMS: improperly formatted checksum line 0-2023/09/11 10:53:58 NOTICE: SHA256SUMS: improperly formatted checksum line 1-2023/09/11 10:53:58 NOTICE: SHA256SUMS: improperly formatted checksum line 49-2023/09/11 10:53:58 NOTICE: SHA256SUMS: 4 warning(s) suppressed...+$ rclone hashsum sha256 -C SHA256SUMS.signed rclone-v1.63.1-windows-amd64.zip= rclone-v1.63.1-windows-amd64.zip2023/09/11 10:53:58 NOTICE: Local file system at /tmp/check: 0 differences found2023/09/11 10:53:58 NOTICE: Local file system at /tmp/check: 1 matching files@@ -36225,10 +38226,14 @@ Ceph.IP \[bu] 2China Mobile Ecloud Elastic Object Storage (EOS).IP \[bu] 2+Clever Cloud Cellar+.IP \[bu] 2Cloudflare R2.IP \[bu] 2Arvan Cloud Object Storage (AOS).IP \[bu] 2+Autonomys Auto Drive+.IP \[bu] 2Bizfly Cloud Simple Storage.IP \[bu] 2Cubbit DS3@@ -36257,6 +38262,8 @@ IDrive e2.IP \[bu] 2Impossible Cloud.IP \[bu] 2+Infomaniak Swiss Backup+.IP \[bu] 2Intercolo Object Storage.IP \[bu] 2IONOS Cloud@@ -36273,6 +38280,8 @@ MEGA S4 Object Storage.IP \[bu] 2Minio.IP \[bu] 2+Neon Object Storage+.IP \[bu] 2Outscale.IP \[bu] 2OVHcloud@@ -36289,6 +38298,8 @@ RackCorp Object Storage.IP \[bu] 2Rclone Serve S3.IP \[bu] 2+RelAix.Storage+.IP \[bu] 2Scaleway.IP \[bu] 2Scality (RING / ARTESCA)@@ -37331,6 +39342,9 @@ error like below.\f[R].fi.PP+The message shows the object\[aq]s actual storage class, for example+\f[V]DEEP_ARCHIVE\f[R].+.PPIn this case you need torestore (http://docs.aws.amazon.com/AmazonS3/latest/user-guide/restore-archived-objects.html)the object(s) in question before accessing object contents.@@ -37415,14 +39429,15 @@ Object Lock headers during PUT).SS Standard options.PPHere are the Standard options specific to s3 (Amazon S3 Compliant-Storage Providers including AWS, Alibaba, ArvanCloud, BizflyCloud, Ceph,-ChinaMobile, Cloudflare, Cubbit, DigitalOcean, Dreamhost, Exaba, Fastly,-FileLu, FlashBlade, GCS, HCP, Hetzner, HuaweiOBS, IBMCOS, IDrive,-ImpossibleCloud, Intercolo, IONOS, Leviia, Liara, Linode, LyveCloud,-Magalu, Mega, Minio, Netease, Outscale, OVHcloud, Petabox, Qiniu,-Rabata, RackCorp, Rclone, Scaleway, Scality, SeaweedFS, Selectel,-Servercore, SpectraLogic, Storj, Synology, TencentCOS, US3, Wasabi,-Zadara, Zata, ZeroServices, Other).+Storage Providers including AWS, Alibaba, ArvanCloud, Autonomys,+BizflyCloud, Cellar, Ceph, ChinaMobile, Cloudflare, Cubbit,+DigitalOcean, Dreamhost, Exaba, Fastly, FileLu, FlashBlade, GCS, HCP,+Hetzner, HuaweiOBS, IBMCOS, IDrive, ImpossibleCloud, Infomaniak,+Intercolo, IONOS, Leviia, Liara, Linode, LyveCloud, Magalu, Mega, Minio,+Neon, Netease, Outscale, OVHcloud, Petabox, Qiniu, Rabata, RackCorp,+Rclone, RelAix, Scaleway, Scality, SeaweedFS, Selectel, Servercore,+SpectraLogic, Storj, Synology, TencentCOS, US3, Wasabi, Zadara, Zata,+ZeroServices, Other)..SS --s3-provider.PPChoose your S3 provider.@@ -37458,12 +39473,24 @@ Alibaba Cloud Object Storage System (OSS) formerly AliyunArvan Cloud Object Storage (AOS).RE.IP \[bu] 2+\[dq]Autonomys\[dq]+.RS 2+.IP \[bu] 2+Autonomys Auto Drive+.RE+.IP \[bu] 2\[dq]BizflyCloud\[dq].RS 2.IP \[bu] 2Bizfly Cloud Simple Storage.RE.IP \[bu] 2+\[dq]Cellar\[dq]+.RS 2+.IP \[bu] 2+Clever Cloud Cellar Object Storage+.RE+.IP \[bu] 2\[dq]Ceph\[dq].RS 2.IP \[bu] 2@@ -37566,6 +39593,12 @@ IDrive e2Impossible Cloud Object Storage.RE.IP \[bu] 2+\[dq]Infomaniak\[dq]+.RS 2+.IP \[bu] 2+Infomaniak Swiss Backup+.RE+.IP \[bu] 2\[dq]Intercolo\[dq].RS 2.IP \[bu] 2@@ -37620,6 +39653,12 @@ MEGA S4 Object StorageMinio Object Storage.RE.IP \[bu] 2+\[dq]Neon\[dq]+.RS 2+.IP \[bu] 2+Neon Object Storage+.RE+.IP \[bu] 2\[dq]Netease\[dq].RS 2.IP \[bu] 2@@ -37668,6 +39707,12 @@ RackCorp Object StorageRclone S3 Server.RE.IP \[bu] 2+\[dq]RelAix\[dq]+.RS 2+.IP \[bu] 2+RelAix.Storage+.RE+.IP \[bu] 2\[dq]Scaleway\[dq].RS 2.IP \[bu] 2@@ -37834,7 +39879,7 @@ Config: regionEnv Var: RCLONE_S3_REGION.IP \[bu] 2Provider:-AWS,BizflyCloud,Ceph,Cloudflare,Cubbit,DigitalOcean,Dreamhost,Exaba,Fastly,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IBMCOS,ImpossibleCloud,Intercolo,IONOS,Leviia,LyveCloud,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,Rabata,RackCorp,Scaleway,Scality,SeaweedFS,Selectel,Servercore,Synology,Wasabi,Zadara,Zata,ZeroServices,Other+AWS,BizflyCloud,Ceph,Cloudflare,Cubbit,DigitalOcean,Dreamhost,Exaba,Fastly,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IBMCOS,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,LyveCloud,Minio,Neon,Netease,Outscale,OVHcloud,Petabox,Qiniu,Rabata,RackCorp,Scaleway,Scality,SeaweedFS,Selectel,Servercore,Synology,Wasabi,Zadara,Zata,ZeroServices,Other.IP \[bu] 2Type: string.IP \[bu] 2@@ -38177,6 +40222,14 @@ EU CentralProvider: Fastly.RE.IP \[bu] 2+\[dq]eu-central-1\[dq]+.RS 2+.IP \[bu] 2+EU Central 1+.IP \[bu] 2+Provider: Fastly+.RE+.IP \[bu] 2\[dq]eu-south-1\[dq].RS 2.IP \[bu] 2@@ -38313,6 +40366,14 @@ NurembergProvider: Hetzner.RE.IP \[bu] 2+\[dq]af-north-1\[dq]+.RS 2+.IP \[bu] 2+AF-Cairo+.IP \[bu] 2+Provider: HuaweiOBS+.RE+.IP \[bu] 2\[dq]af-south-1\[dq].RS 2.IP \[bu] 2@@ -38321,6 +40382,14 @@ AF-JohannesburgProvider: HuaweiOBS.RE.IP \[bu] 2+\[dq]ap-southeast-1\[dq]+.RS 2+.IP \[bu] 2+CN-Hong Kong+.IP \[bu] 2+Provider: HuaweiOBS+.RE+.IP \[bu] 2\[dq]ap-southeast-2\[dq].RS 2.IP \[bu] 2@@ -38337,10 +40406,18 @@ AP-SingaporeProvider: HuaweiOBS.RE.IP \[bu] 2-\[dq]cn-east-3\[dq]+\[dq]ap-southeast-4\[dq].RS 2.IP \[bu] 2-CN East-Shanghai1+AP-Jakarta+.IP \[bu] 2+Provider: HuaweiOBS+.RE+.IP \[bu] 2+\[dq]ap-southeast-5\[dq]+.RS 2+.IP \[bu] 2+AP-Manila.IP \[bu] 2Provider: HuaweiOBS.RE@@ -38353,6 +40430,30 @@ CN East-Shanghai2Provider: HuaweiOBS.RE.IP \[bu] 2+\[dq]cn-east-3\[dq]+.RS 2+.IP \[bu] 2+CN East-Shanghai1+.IP \[bu] 2+Provider: HuaweiOBS+.RE+.IP \[bu] 2+\[dq]cn-east-4\[dq]+.RS 2+.IP \[bu] 2+CN East-Qingdao+.IP \[bu] 2+Provider: HuaweiOBS+.RE+.IP \[bu] 2+\[dq]cn-east-5\[dq]+.RS 2+.IP \[bu] 2+CN East-Suzhou+.IP \[bu] 2+Provider: HuaweiOBS+.RE+.IP \[bu] 2\[dq]cn-north-1\[dq].RS 2.IP \[bu] 2@@ -38361,6 +40462,14 @@ CN North-Beijing1Provider: HuaweiOBS.RE.IP \[bu] 2+\[dq]cn-north-2\[dq]+.RS 2+.IP \[bu] 2+CN North-Beijing2+.IP \[bu] 2+Provider: HuaweiOBS+.RE+.IP \[bu] 2\[dq]cn-north-4\[dq].RS 2.IP \[bu] 2@@ -38369,6 +40478,30 @@ CN North-Beijing4Provider: HuaweiOBS.RE.IP \[bu] 2+\[dq]cn-north-9\[dq]+.RS 2+.IP \[bu] 2+CN North-Ulanqab+.IP \[bu] 2+Provider: HuaweiOBS+.RE+.IP \[bu] 2+\[dq]cn-north-11\[dq]+.RS 2+.IP \[bu] 2+CN North-Hebei+.IP \[bu] 2+Provider: HuaweiOBS+.RE+.IP \[bu] 2+\[dq]cn-north-12\[dq]+.RS 2+.IP \[bu] 2+CN North-Hohhot+.IP \[bu] 2+Provider: HuaweiOBS+.RE+.IP \[bu] 2\[dq]cn-south-1\[dq].RS 2.IP \[bu] 2@@ -38377,26 +40510,66 @@ CN South-GuangzhouProvider: HuaweiOBS.RE.IP \[bu] 2-\[dq]ap-southeast-1\[dq]+\[dq]cn-south-2\[dq].RS 2.IP \[bu] 2-CN-Hong Kong+CN South-Shenzhen.IP \[bu] 2Provider: HuaweiOBS.RE.IP \[bu] 2-\[dq]sa-argentina-1\[dq]+\[dq]cn-south-4\[dq].RS 2.IP \[bu] 2-LA-Buenos Aires1+CN South-Guangzhou-InvitationOnly.IP \[bu] 2Provider: HuaweiOBS.RE.IP \[bu] 2-\[dq]sa-peru-1\[dq]+\[dq]cn-southwest-2\[dq].RS 2.IP \[bu] 2-LA-Lima1+CN Southwest-Guiyang1+.IP \[bu] 2+Provider: HuaweiOBS+.RE+.IP \[bu] 2+\[dq]cn-southwest-3\[dq]+.RS 2+.IP \[bu] 2+CN Southwest-Chongqing+.IP \[bu] 2+Provider: HuaweiOBS+.RE+.IP \[bu] 2+\[dq]eu-west-101\[dq]+.RS 2+.IP \[bu] 2+EU-Dublin+.IP \[bu] 2+Provider: HuaweiOBS+.RE+.IP \[bu] 2+\[dq]la-north-2\[dq]+.RS 2+.IP \[bu] 2+LA-Mexico City2+.IP \[bu] 2+Provider: HuaweiOBS+.RE+.IP \[bu] 2+\[dq]la-south-2\[dq]+.RS 2+.IP \[bu] 2+LA-Santiago+.IP \[bu] 2+Provider: HuaweiOBS+.RE+.IP \[bu] 2+\[dq]me-east-1\[dq]+.RS 2+.IP \[bu] 2+ME-Riyadh.IP \[bu] 2Provider: HuaweiOBS.RE@@ -38409,10 +40582,18 @@ LA-Mexico City1Provider: HuaweiOBS.RE.IP \[bu] 2-\[dq]sa-chile-1\[dq]+\[dq]ru-northwest-2\[dq].RS 2.IP \[bu] 2-LA-Santiago2+RU-Moscow2+.IP \[bu] 2+Provider: HuaweiOBS+.RE+.IP \[bu] 2+\[dq]sa-argentina-1\[dq]+.RS 2+.IP \[bu] 2+LA-Buenos Aires1.IP \[bu] 2Provider: HuaweiOBS.RE@@ -38425,10 +40606,26 @@ LA-Sao Paulo1Provider: HuaweiOBS.RE.IP \[bu] 2-\[dq]ru-northwest-2\[dq]+\[dq]sa-chile-1\[dq].RS 2.IP \[bu] 2-RU-Moscow2+LA-Santiago2+.IP \[bu] 2+Provider: HuaweiOBS+.RE+.IP \[bu] 2+\[dq]sa-peru-1\[dq]+.RS 2+.IP \[bu] 2+LA-Lima1+.IP \[bu] 2+Provider: HuaweiOBS+.RE+.IP \[bu] 2+\[dq]tr-west-1\[dq]+.RS 2+.IP \[bu] 2+TR-Istanbul.IP \[bu] 2Provider: HuaweiOBS.RE@@ -38489,6 +40686,16 @@ New York, USAProvider: ImpossibleCloud.RE.IP \[bu] 2+\[dq]us-east-1\[dq]+.RS 2+.IP \[bu] 2+Default region for compatibility.+.IP \[bu] 2+All data is stored in Switzerland.+.IP \[bu] 2+Provider: Infomaniak+.RE+.IP \[bu] 2\[dq]de-fra\[dq].RS 2.IP \[bu] 2@@ -38545,6 +40752,39 @@ Lenexa, USAProvider: IONOS.RE.IP \[bu] 2+\[dq]us-east-1\[dq]+.RS 2+.IP \[bu] 2+US East (N.+Virginia)+.IP \[bu] 2+Provider: Neon,Petabox,Rabata+.RE+.IP \[bu] 2+\[dq]us-east-2\[dq]+.RS 2+.IP \[bu] 2+US East (Ohio)+.IP \[bu] 2+Provider: Neon+.RE+.IP \[bu] 2+\[dq]eu-central-1\[dq]+.RS 2+.IP \[bu] 2+Europe (Frankfurt)+.IP \[bu] 2+Provider: Neon,Petabox+.RE+.IP \[bu] 2+\[dq]ap-southeast-1\[dq]+.RS 2+.IP \[bu] 2+Asia Pacific (Singapore)+.IP \[bu] 2+Provider: Neon,Petabox+.RE+.IP \[bu] 2\[dq]eu-west-2\[dq].RS 2.IP \[bu] 2@@ -38697,31 +40937,6 @@ Roubaix, France (Cold Archive)Provider: OVHcloud.RE.IP \[bu] 2-\[dq]us-east-1\[dq]-.RS 2-.IP \[bu] 2-US East (N.-Virginia)-.IP \[bu] 2-Provider: Petabox,Rabata-.RE-.IP \[bu] 2-\[dq]eu-central-1\[dq]-.RS 2-.IP \[bu] 2-Europe (Frankfurt)-.IP \[bu] 2-Provider: Petabox-.RE-.IP \[bu] 2-\[dq]ap-southeast-1\[dq]-.RS 2-.IP \[bu] 2-Asia Pacific (Singapore)-.IP \[bu] 2-Provider: Petabox-.RE-.IP \[bu] 2\[dq]me-south-1\[dq].RS 2.IP \[bu] 2@@ -39002,6 +41217,14 @@ Warsaw, PolandProvider: Scaleway.RE.IP \[bu] 2+\[dq]it-mil\[dq]+.RS 2+.IP \[bu] 2+Milan, Italy+.IP \[bu] 2+Provider: Scaleway+.RE+.IP \[bu] 2\[dq]ru-1\[dq].RS 2.IP \[bu] 2@@ -39163,7 +41386,7 @@ Config: endpointEnv Var: RCLONE_S3_ENDPOINT.IP \[bu] 2Provider:-AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,ChinaMobile,Cloudflare,Cubbit,DigitalOcean,Dreamhost,Exaba,Fastly,FileLu,FlashBlade,GCS,HCP,Hetzner,HuaweiOBS,IBMCOS,ImpossibleCloud,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Mega,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,Rabata,RackCorp,Rclone,Scaleway,Scality,SeaweedFS,Selectel,Servercore,SpectraLogic,Storj,Synology,TencentCOS,US3,Wasabi,Zadara,Zata,ZeroServices,Other+AWS,Alibaba,ArvanCloud,Autonomys,BizflyCloud,Cellar,Ceph,ChinaMobile,Cloudflare,Cubbit,DigitalOcean,Dreamhost,Exaba,Fastly,FileLu,FlashBlade,GCS,HCP,Hetzner,HuaweiOBS,IBMCOS,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Mega,Minio,Neon,Netease,Outscale,OVHcloud,Petabox,Qiniu,Rabata,RackCorp,Rclone,RelAix,Scaleway,Scality,SeaweedFS,Selectel,Servercore,SpectraLogic,Storj,Synology,TencentCOS,US3,Wasabi,Zadara,Zata,ZeroServices,Other.IP \[bu] 2Type: string.IP \[bu] 2@@ -39390,6 +41613,14 @@ Tabriz Iran (Shahriar)Provider: ArvanCloud.RE.IP \[bu] 2+\[dq]s3.auto-drive.autonomys.xyz\[dq]+.RS 2+.IP \[bu] 2+Auto Drive mainnet S3 endpoint+.IP \[bu] 2+Provider: Autonomys+.RE+.IP \[bu] 2\[dq]hn.ss.bfcplatform.vn\[dq].RS 2.IP \[bu] 2@@ -39406,6 +41637,14 @@ Ho Chi Minh endpointProvider: BizflyCloud.RE.IP \[bu] 2+\[dq]cellar-c2.services.clever-cloud.com\[dq]+.RS 2+.IP \[bu] 2+Paris, France+.IP \[bu] 2+Provider: Cellar+.RE+.IP \[bu] 2\[dq]eos-wuxi-1.cmecloud.cn\[dq].RS 2.IP \[bu] 2@@ -39770,6 +42009,14 @@ EU CentralProvider: Fastly.RE.IP \[bu] 2+\[dq]eu-central-1.object.fastlystorage.app\[dq]+.RS 2+.IP \[bu] 2+EU Central 1+.IP \[bu] 2+Provider: Fastly+.RE+.IP \[bu] 2\[dq]eu-south-1.object.fastlystorage.app\[dq].RS 2.IP \[bu] 2@@ -39914,6 +42161,14 @@ NurembergProvider: Hetzner.RE.IP \[bu] 2+\[dq]obs.af-north-1.myhuaweicloud.com\[dq]+.RS 2+.IP \[bu] 2+AF-Cairo+.IP \[bu] 2+Provider: HuaweiOBS+.RE+.IP \[bu] 2\[dq]obs.af-south-1.myhuaweicloud.com\[dq].RS 2.IP \[bu] 2@@ -39922,6 +42177,14 @@ AF-JohannesburgProvider: HuaweiOBS.RE.IP \[bu] 2+\[dq]obs.ap-southeast-1.myhuaweicloud.com\[dq]+.RS 2+.IP \[bu] 2+CN-Hong Kong+.IP \[bu] 2+Provider: HuaweiOBS+.RE+.IP \[bu] 2\[dq]obs.ap-southeast-2.myhuaweicloud.com\[dq].RS 2.IP \[bu] 2@@ -39938,10 +42201,18 @@ AP-SingaporeProvider: HuaweiOBS.RE.IP \[bu] 2-\[dq]obs.cn-east-3.myhuaweicloud.com\[dq]+\[dq]obs.ap-southeast-4.myhuaweicloud.com\[dq].RS 2.IP \[bu] 2-CN East-Shanghai1+AP-Jakarta+.IP \[bu] 2+Provider: HuaweiOBS+.RE+.IP \[bu] 2+\[dq]obs.ap-southeast-5.myhuaweicloud.com\[dq]+.RS 2+.IP \[bu] 2+AP-Manila.IP \[bu] 2Provider: HuaweiOBS.RE@@ -39954,6 +42225,30 @@ CN East-Shanghai2Provider: HuaweiOBS.RE.IP \[bu] 2+\[dq]obs.cn-east-3.myhuaweicloud.com\[dq]+.RS 2+.IP \[bu] 2+CN East-Shanghai1+.IP \[bu] 2+Provider: HuaweiOBS+.RE+.IP \[bu] 2+\[dq]obs.cn-east-4.myhuaweicloud.com\[dq]+.RS 2+.IP \[bu] 2+CN East-Qingdao+.IP \[bu] 2+Provider: HuaweiOBS+.RE+.IP \[bu] 2+\[dq]obs.cn-east-5.myhuaweicloud.com\[dq]+.RS 2+.IP \[bu] 2+CN East-Suzhou+.IP \[bu] 2+Provider: HuaweiOBS+.RE+.IP \[bu] 2\[dq]obs.cn-north-1.myhuaweicloud.com\[dq].RS 2.IP \[bu] 2@@ -39962,6 +42257,14 @@ CN North-Beijing1Provider: HuaweiOBS.RE.IP \[bu] 2+\[dq]obs.cn-north-2.myhuaweicloud.com\[dq]+.RS 2+.IP \[bu] 2+CN North-Beijing2+.IP \[bu] 2+Provider: HuaweiOBS+.RE+.IP \[bu] 2\[dq]obs.cn-north-4.myhuaweicloud.com\[dq].RS 2.IP \[bu] 2@@ -39970,6 +42273,30 @@ CN North-Beijing4Provider: HuaweiOBS.RE.IP \[bu] 2+\[dq]obs.cn-north-9.myhuaweicloud.com\[dq]+.RS 2+.IP \[bu] 2+CN North-Ulanqab+.IP \[bu] 2+Provider: HuaweiOBS+.RE+.IP \[bu] 2+\[dq]obs.cn-north-11.myhuaweicloud.com\[dq]+.RS 2+.IP \[bu] 2+CN North-Hebei+.IP \[bu] 2+Provider: HuaweiOBS+.RE+.IP \[bu] 2+\[dq]obs.cn-north-12.myhuaweicloud.com\[dq]+.RS 2+.IP \[bu] 2+CN North-Hohhot+.IP \[bu] 2+Provider: HuaweiOBS+.RE+.IP \[bu] 2\[dq]obs.cn-south-1.myhuaweicloud.com\[dq].RS 2.IP \[bu] 2@@ -39978,26 +42305,66 @@ CN South-GuangzhouProvider: HuaweiOBS.RE.IP \[bu] 2-\[dq]obs.ap-southeast-1.myhuaweicloud.com\[dq]+\[dq]obs.cn-south-2.myhuaweicloud.com\[dq].RS 2.IP \[bu] 2-CN-Hong Kong+CN South-Shenzhen.IP \[bu] 2Provider: HuaweiOBS.RE.IP \[bu] 2-\[dq]obs.sa-argentina-1.myhuaweicloud.com\[dq]+\[dq]obs.cn-south-4.myhuaweicloud.com\[dq].RS 2.IP \[bu] 2-LA-Buenos Aires1+CN South-Guangzhou-InvitationOnly.IP \[bu] 2Provider: HuaweiOBS.RE.IP \[bu] 2-\[dq]obs.sa-peru-1.myhuaweicloud.com\[dq]+\[dq]obs.cn-southwest-2.myhuaweicloud.com\[dq].RS 2.IP \[bu] 2-LA-Lima1+CN Southwest-Guiyang1+.IP \[bu] 2+Provider: HuaweiOBS+.RE+.IP \[bu] 2+\[dq]obs.cn-southwest-3.myhuaweicloud.com\[dq]+.RS 2+.IP \[bu] 2+CN Southwest-Chongqing+.IP \[bu] 2+Provider: HuaweiOBS+.RE+.IP \[bu] 2+\[dq]obs.eu-west-101.myhuaweicloud.eu\[dq]+.RS 2+.IP \[bu] 2+EU-Dublin+.IP \[bu] 2+Provider: HuaweiOBS+.RE+.IP \[bu] 2+\[dq]obs.la-north-2.myhuaweicloud.com\[dq]+.RS 2+.IP \[bu] 2+LA-Mexico City2+.IP \[bu] 2+Provider: HuaweiOBS+.RE+.IP \[bu] 2+\[dq]obs.la-south-2.myhuaweicloud.com\[dq]+.RS 2+.IP \[bu] 2+LA-Santiago+.IP \[bu] 2+Provider: HuaweiOBS+.RE+.IP \[bu] 2+\[dq]obs.me-east-1.myhuaweicloud.com\[dq]+.RS 2+.IP \[bu] 2+ME-Riyadh.IP \[bu] 2Provider: HuaweiOBS.RE@@ -40010,10 +42377,18 @@ LA-Mexico City1Provider: HuaweiOBS.RE.IP \[bu] 2-\[dq]obs.sa-chile-1.myhuaweicloud.com\[dq]+\[dq]obs.ru-northwest-2.myhuaweicloud.com\[dq].RS 2.IP \[bu] 2-LA-Santiago2+RU-Moscow2+.IP \[bu] 2+Provider: HuaweiOBS+.RE+.IP \[bu] 2+\[dq]obs.sa-argentina-1.myhuaweicloud.com\[dq]+.RS 2+.IP \[bu] 2+LA-Buenos Aires1.IP \[bu] 2Provider: HuaweiOBS.RE@@ -40026,10 +42401,26 @@ LA-Sao Paulo1Provider: HuaweiOBS.RE.IP \[bu] 2-\[dq]obs.ru-northwest-2.myhuaweicloud.com\[dq]+\[dq]obs.sa-chile-1.myhuaweicloud.com\[dq].RS 2.IP \[bu] 2-RU-Moscow2+LA-Santiago2+.IP \[bu] 2+Provider: HuaweiOBS+.RE+.IP \[bu] 2+\[dq]obs.sa-peru-1.myhuaweicloud.com\[dq]+.RS 2+.IP \[bu] 2+LA-Lima1+.IP \[bu] 2+Provider: HuaweiOBS+.RE+.IP \[bu] 2+\[dq]obs.tr-west-1.myhuaweicloud.com\[dq]+.RS 2+.IP \[bu] 2+TR-Istanbul.IP \[bu] 2Provider: HuaweiOBS.RE@@ -40586,6 +42977,38 @@ New York, USAProvider: ImpossibleCloud.RE.IP \[bu] 2+\[dq]s3.swiss-backup01.infomaniak.com\[dq]+.RS 2+.IP \[bu] 2+Swiss Backup Location 01+.IP \[bu] 2+Provider: Infomaniak+.RE+.IP \[bu] 2+\[dq]s3.swiss-backup02.infomaniak.com\[dq]+.RS 2+.IP \[bu] 2+Swiss Backup Location 02+.IP \[bu] 2+Provider: Infomaniak+.RE+.IP \[bu] 2+\[dq]s3.swiss-backup03.infomaniak.com\[dq]+.RS 2+.IP \[bu] 2+Swiss Backup Location 03+.IP \[bu] 2+Provider: Infomaniak+.RE+.IP \[bu] 2+\[dq]s3.swiss-backup04.infomaniak.com\[dq]+.RS 2+.IP \[bu] 2+Swiss Backup Location 04+.IP \[bu] 2+Provider: Infomaniak+.RE+.IP \[bu] 2\[dq]de-fra.i3storage.com\[dq].RS 2.IP \[bu] 2@@ -40862,90 +43285,114 @@ Fortaleza, CE (BR), br-ne1Provider: Magalu.RE.IP \[bu] 2-\[dq]s3.eu-amsterdam.megas4.com\[dq]+\[dq]s3.eu-luxembourg-1.megas4.com\[dq]+.RS 2+.IP \[bu] 2+Mega S4 Luxembourg 1+.IP \[bu] 2+Provider: Mega+.RE+.IP \[bu] 2+\[dq]s3.eu-luxembourg-2.megas4.com\[dq]+.RS 2+.IP \[bu] 2+Mega S4 Luxembourg 2+.IP \[bu] 2+Provider: Mega+.RE+.IP \[bu] 2+\[dq]s3.eu-amsterdam-1.megas4.com\[dq]+.RS 2+.IP \[bu] 2+Mega S4 Amsterdam 1+.IP \[bu] 2+Provider: Mega+.RE+.IP \[bu] 2+\[dq]s3.eu-amsterdam-2.megas4.com\[dq].RS 2.IP \[bu] 2-Mega S4 Amsterdam+Mega S4 Amsterdam 2.IP \[bu] 2Provider: Mega.RE.IP \[bu] 2-\[dq]s3.eu-luxembourg.megas4.com\[dq]+\[dq]s3.eu-paris-1.megas4.com\[dq].RS 2.IP \[bu] 2-Mega S4 Luxembourg+Mega S4 Paris 1.IP \[bu] 2Provider: Mega.RE.IP \[bu] 2-\[dq]s3.eu-paris.megas4.com\[dq]+\[dq]s3.eu-paris-2.megas4.com\[dq].RS 2.IP \[bu] 2-Mega S4 Paris+Mega S4 Paris 2.IP \[bu] 2Provider: Mega.RE.IP \[bu] 2-\[dq]s3.eu-barcelona.megas4.com\[dq]+\[dq]s3.eu-barcelona-1.megas4.com\[dq].RS 2.IP \[bu] 2-Mega S4 Barcelona+Mega S4 Barcelona 1.IP \[bu] 2Provider: Mega.RE.IP \[bu] 2-\[dq]s3.ca-montreal.megas4.com\[dq]+\[dq]s3.eu-barcelona-2.megas4.com\[dq].RS 2.IP \[bu] 2-Mega S4 Montreal+Mega S4 Barcelona 2.IP \[bu] 2Provider: Mega.RE.IP \[bu] 2-\[dq]s3.ca-vancouver.megas4.com\[dq]+\[dq]s3.ca-montreal-1.megas4.com\[dq].RS 2.IP \[bu] 2-Mega S4 Vancouver+Mega S4 Montreal 1.IP \[bu] 2Provider: Mega.RE.IP \[bu] 2-\[dq]s3.ap-tokyo.megas4.com\[dq]+\[dq]s3.ca-montreal-2.megas4.com\[dq].RS 2.IP \[bu] 2-Mega S4 Tokyo+Mega S4 Montreal 2.IP \[bu] 2Provider: Mega.RE.IP \[bu] 2-\[dq]s3.eu-central-1.s4.mega.io\[dq]+\[dq]s3.ca-vancouver-1.megas4.com\[dq].RS 2.IP \[bu] 2-Mega S4 eu-central-1 (Amsterdam, legacy)+Mega S4 Vancouver 1.IP \[bu] 2Provider: Mega.RE.IP \[bu] 2-\[dq]s3.eu-central-2.s4.mega.io\[dq]+\[dq]s3.ca-vancouver-2.megas4.com\[dq].RS 2.IP \[bu] 2-Mega S4 eu-central-2 (Bettembourg, legacy)+Mega S4 Vancouver 2.IP \[bu] 2Provider: Mega.RE.IP \[bu] 2-\[dq]s3.ca-central-1.s4.mega.io\[dq]+\[dq]s3.ap-tokyo-1.megas4.com\[dq].RS 2.IP \[bu] 2-Mega S4 ca-central-1 (Montreal, legacy)+Mega S4 Tokyo 1.IP \[bu] 2Provider: Mega.RE.IP \[bu] 2-\[dq]s3.ca-west-1.s4.mega.io\[dq]+\[dq]s3.ap-tokyo-2.megas4.com\[dq].RS 2.IP \[bu] 2-Mega S4 ca-west-1 (Vancouver, legacy)+Mega S4 Tokyo 2.IP \[bu] 2Provider: Mega.RE@@ -41393,6 +43840,22 @@ Auckland (New Zealand) EndpointProvider: RackCorp.RE.IP \[bu] 2+\[dq]s3-gw1.relaix.net\[dq]+.RS 2+.IP \[bu] 2+Aachen-1, Germany+.IP \[bu] 2+Provider: RelAix+.RE+.IP \[bu] 2+\[dq]s3-gw2.relaix.net\[dq]+.RS 2+.IP \[bu] 2+Aachen-2, Germany+.IP \[bu] 2+Provider: RelAix+.RE+.IP \[bu] 2\[dq]s3.nl-ams.scw.cloud\[dq].RS 2.IP \[bu] 2@@ -41417,6 +43880,14 @@ Warsaw EndpointProvider: Scaleway.RE.IP \[bu] 2+\[dq]s3.it-mil.scw.cloud\[dq]+.RS 2+.IP \[bu] 2+Milan Endpoint+.IP \[bu] 2+Provider: Scaleway+.RE+.IP \[bu] 2\[dq]localhost:8333\[dq].RS 2.IP \[bu] 2@@ -42035,7 +44506,7 @@ Config: location_constraintEnv Var: RCLONE_S3_LOCATION_CONSTRAINT.IP \[bu] 2Provider:-AWS,ArvanCloud,Ceph,ChinaMobile,DigitalOcean,Dreamhost,Exaba,GCS,HCP,Hetzner,IBMCOS,ImpossibleCloud,LyveCloud,Minio,Netease,Qiniu,Rabata,RackCorp,Scality,SeaweedFS,Synology,Wasabi,Zata,ZeroServices,Other+AWS,ArvanCloud,Ceph,ChinaMobile,DigitalOcean,Dreamhost,Exaba,GCS,HCP,Hetzner,IBMCOS,ImpossibleCloud,Infomaniak,LyveCloud,Minio,Netease,Qiniu,Rabata,RackCorp,Scality,SeaweedFS,Synology,Wasabi,Zata,ZeroServices,Other.IP \[bu] 2Type: string.IP \[bu] 2@@ -43020,7 +45491,7 @@ Config: aclEnv Var: RCLONE_S3_ACL.IP \[bu] 2Provider:-AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IBMCOS,IDrive,ImpossibleCloud,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,Scaleway,Scality,SeaweedFS,TencentCOS,US3,Wasabi,Zata,ZeroServices,Other+AWS,Alibaba,ArvanCloud,BizflyCloud,Cellar,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IBMCOS,IDrive,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,RelAix,Scaleway,Scality,SeaweedFS,TencentCOS,US3,Wasabi,Zata,ZeroServices,Other.IP \[bu] 2Type: string.IP \[bu] 2@@ -43037,7 +45508,7 @@ Owner gets FULL_CONTROL.No one else has access rights (default)..IP \[bu] 2Provider:-AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,Scaleway,Scality,SeaweedFS,US3,Wasabi,Zata,ZeroServices,Other+AWS,Alibaba,ArvanCloud,BizflyCloud,Cellar,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,RelAix,Scaleway,Scality,SeaweedFS,US3,Wasabi,Zata,ZeroServices,Other.RE.IP \[bu] 2\[dq]public-read\[dq]@@ -43048,7 +45519,7 @@ Owner gets FULL_CONTROL.The AllUsers group gets READ access..IP \[bu] 2Provider:-AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,Scaleway,Scality,SeaweedFS,TencentCOS,US3,Wasabi,Zata,ZeroServices,Other+AWS,Alibaba,ArvanCloud,BizflyCloud,Cellar,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,RelAix,Scaleway,Scality,SeaweedFS,TencentCOS,US3,Wasabi,Zata,ZeroServices,Other.RE.IP \[bu] 2\[dq]public-read-write\[dq]@@ -43061,7 +45532,7 @@ The AllUsers group gets READ and WRITE access.Granting this on a bucket is generally not recommended..IP \[bu] 2Provider:-AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,Scaleway,Scality,SeaweedFS,TencentCOS,Wasabi,Zata,ZeroServices,Other+AWS,Alibaba,ArvanCloud,BizflyCloud,Cellar,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,RelAix,Scaleway,Scality,SeaweedFS,TencentCOS,Wasabi,Zata,ZeroServices,Other.RE.IP \[bu] 2\[dq]authenticated-read\[dq]@@ -43072,7 +45543,7 @@ Owner gets FULL_CONTROL.The AuthenticatedUsers group gets READ access..IP \[bu] 2Provider:-AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,Scaleway,Scality,SeaweedFS,TencentCOS,Wasabi,Zata,ZeroServices,Other+AWS,Alibaba,ArvanCloud,BizflyCloud,Cellar,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,RelAix,Scaleway,Scality,SeaweedFS,TencentCOS,Wasabi,Zata,ZeroServices,Other.RE.IP \[bu] 2\[dq]bucket-owner-read\[dq]@@ -43086,7 +45557,7 @@ If you specify this canned ACL when creating a bucket, Amazon S3 ignoresit..IP \[bu] 2Provider:-AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,Scaleway,Scality,SeaweedFS,TencentCOS,Wasabi,Zata,ZeroServices,Other+AWS,Alibaba,ArvanCloud,BizflyCloud,Cellar,Ceph,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,RelAix,Scaleway,Scality,SeaweedFS,TencentCOS,Wasabi,Zata,ZeroServices,Other.RE.IP \[bu] 2\[dq]bucket-owner-full-control\[dq]@@ -43099,7 +45570,7 @@ If you specify this canned ACL when creating a bucket, Amazon S3 ignoresit..IP \[bu] 2Provider:-AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,Scaleway,Scality,SeaweedFS,TencentCOS,Wasabi,Zata,ZeroServices,Other+AWS,Alibaba,ArvanCloud,BizflyCloud,Cellar,Ceph,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IDrive,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,RelAix,Scaleway,Scality,SeaweedFS,TencentCOS,Wasabi,Zata,ZeroServices,Other.RE.IP \[bu] 2\[dq]private\[dq]@@ -43511,14 +45982,15 @@ Default: false.SS Advanced options.PPHere are the Advanced options specific to s3 (Amazon S3 Compliant-Storage Providers including AWS, Alibaba, ArvanCloud, BizflyCloud, Ceph,-ChinaMobile, Cloudflare, Cubbit, DigitalOcean, Dreamhost, Exaba, Fastly,-FileLu, FlashBlade, GCS, HCP, Hetzner, HuaweiOBS, IBMCOS, IDrive,-ImpossibleCloud, Intercolo, IONOS, Leviia, Liara, Linode, LyveCloud,-Magalu, Mega, Minio, Netease, Outscale, OVHcloud, Petabox, Qiniu,-Rabata, RackCorp, Rclone, Scaleway, Scality, SeaweedFS, Selectel,-Servercore, SpectraLogic, Storj, Synology, TencentCOS, US3, Wasabi,-Zadara, Zata, ZeroServices, Other).+Storage Providers including AWS, Alibaba, ArvanCloud, Autonomys,+BizflyCloud, Cellar, Ceph, ChinaMobile, Cloudflare, Cubbit,+DigitalOcean, Dreamhost, Exaba, Fastly, FileLu, FlashBlade, GCS, HCP,+Hetzner, HuaweiOBS, IBMCOS, IDrive, ImpossibleCloud, Infomaniak,+Intercolo, IONOS, Leviia, Liara, Linode, LyveCloud, Magalu, Mega, Minio,+Neon, Netease, Outscale, OVHcloud, Petabox, Qiniu, Rabata, RackCorp,+Rclone, RelAix, Scaleway, Scality, SeaweedFS, Selectel, Servercore,+SpectraLogic, Storj, Synology, TencentCOS, US3, Wasabi, Zadara, Zata,+ZeroServices, Other)..SS --s3-bucket-acl.PPCanned ACL used when creating buckets.@@ -43539,7 +46011,7 @@ Config: bucket_aclEnv Var: RCLONE_S3_BUCKET_ACL.IP \[bu] 2Provider:-AWS,Alibaba,ArvanCloud,BizflyCloud,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IBMCOS,IDrive,ImpossibleCloud,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Mega,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,Scaleway,SeaweedFS,Servercore,TencentCOS,US3,Wasabi,Zata,ZeroServices,Other+AWS,Alibaba,ArvanCloud,BizflyCloud,Cellar,Ceph,ChinaMobile,Cubbit,DigitalOcean,Dreamhost,Exaba,FileLu,GCS,HCP,Hetzner,HuaweiOBS,IBMCOS,IDrive,ImpossibleCloud,Infomaniak,Intercolo,IONOS,Leviia,Liara,Linode,LyveCloud,Magalu,Mega,Minio,Netease,Outscale,OVHcloud,Petabox,Qiniu,RackCorp,RelAix,Scaleway,SeaweedFS,Servercore,TencentCOS,US3,Wasabi,Zata,ZeroServices,Other.IP \[bu] 2Type: string.IP \[bu] 2@@ -43608,7 +46080,7 @@ Config: sse_customer_algorithm.IP \[bu] 2Env Var: RCLONE_S3_SSE_CUSTOMER_ALGORITHM.IP \[bu] 2-Provider: AWS,Ceph,ChinaMobile,Minio+Provider: AWS,Cellar,Ceph,ChinaMobile,Minio.IP \[bu] 2Type: string.IP \[bu] 2@@ -43634,7 +46106,7 @@ AES256To use SSE-C you may provide the secret encryption key used toencrypt/decrypt your data..PP-Alternatively you can provide --sse-customer-key-base64.+Alternatively you can provide --s3-sse-customer-key-base64..PPProperties:.IP \[bu] 2@@ -43642,7 +46114,7 @@ Config: sse_customer_key.IP \[bu] 2Env Var: RCLONE_S3_SSE_CUSTOMER_KEY.IP \[bu] 2-Provider: AWS,Ceph,ChinaMobile,Minio+Provider: AWS,Cellar,Ceph,ChinaMobile,Minio.IP \[bu] 2Type: string.IP \[bu] 2@@ -43662,7 +46134,7 @@ NoneIf using SSE-C you must provide the secret encryption key encoded inbase64 format to encrypt/decrypt your data..PP-Alternatively you can provide --sse-customer-key.+Alternatively you can provide --s3-sse-customer-key..PPProperties:.IP \[bu] 2@@ -43670,7 +46142,7 @@ Config: sse_customer_key_base64.IP \[bu] 2Env Var: RCLONE_S3_SSE_CUSTOMER_KEY_BASE64.IP \[bu] 2-Provider: AWS,Ceph,ChinaMobile,Minio+Provider: AWS,Cellar,Ceph,ChinaMobile,Minio.IP \[bu] 2Type: string.IP \[bu] 2@@ -43699,7 +46171,7 @@ Config: sse_customer_key_md5.IP \[bu] 2Env Var: RCLONE_S3_SSE_CUSTOMER_KEY_MD5.IP \[bu] 2-Provider: AWS,Ceph,ChinaMobile,Minio+Provider: AWS,Cellar,Ceph,ChinaMobile,Minio.IP \[bu] 2Type: string.IP \[bu] 2@@ -45136,6 +47608,64 @@ for more info on how to pass options and arguments..PPThese can be run on a running backend using the rc commandbackend/command (https://rclone.org/rc/#backend-command).+.SS link+.PP+Generate a signed link with response header overrides.+.IP+.nf+\f[C]+rclone backend link remote: [options] [<arguments>+]+\f[R]+.fi+.PP+This command generates a signed download link for one file, with+optional overrides for the HTTP response headers.+Pass the file path as a separate argument, relative to the remote path.+.PP+Usage examples:+.IP+.nf+\f[C]+rclone backend link s3:bucket path/to/file -o expire=1h+rclone backend link s3:bucket path/to/file -o response-expires=\[dq]Thu, 01 Jan 1970 00:00:00 GMT\[dq]+rclone backend link s3:bucket path/to/file -o response-content-disposition=\[aq]attachment; filename=\[dq]download.txt\[dq]\[aq]+\f[R]+.fi+.PP+The link expires after 7 days by default.+Use \f[V]-o expire=1h\f[R] to change its lifetime.+Durations must be at least 1 second; values above 7 days are limited to+7 days, as with \f[V]rclone link\f[R].+.PP+The \f[V]response-expires\f[R] option sets the HTTP Expires response+header, not the lifetime of the signed link.+It must be an HTTP date, such as+\f[V]Thu, 01 Jan 1970 00:00:00 GMT\f[R].+.PP+The overrides are included in the signature and must not be changed in+the returned URL.+They do not modify the object\[aq]s stored metadata.+.PP+Options:+.IP \[bu] 2+\[dq]expire\[dq]: How long the link will be valid (default 7d, maximum+7d).+.IP \[bu] 2+\[dq]response-cache-control\[dq]: Set the Cache-Control response header.+.IP \[bu] 2+\[dq]response-content-disposition\[dq]: Set the Content-Disposition+response header.+.IP \[bu] 2+\[dq]response-content-encoding\[dq]: Set the Content-Encoding response+header.+.IP \[bu] 2+\[dq]response-content-language\[dq]: Set the Content-Language response+header.+.IP \[bu] 2+\[dq]response-content-type\[dq]: Set the Content-Type response header.+.IP \[bu] 2+\[dq]response-expires\[dq]: Set the Expires response header to an HTTP+date..SS restore.PPRestore objects from GLACIER or INTELLIGENT-TIERING archive tier.@@ -45474,8 +48004,8 @@ From rclone v1.69 DirectoryBuckets (https://docs.aws.amazon.com/AmazonS3/latest/userguide/directory-buckets-overview.html)are supported..PP-You will need to set the \f[V]directory_buckets = true\f[R] config-parameter or use \f[V]--s3-directory-buckets\f[R].+You will need to set the \f[V]directory_bucket = true\f[R] config+parameter or use \f[V]--s3-directory-bucket\f[R]..PPNote that rclone cannot yet:.IP \[bu] 2@@ -45483,7 +48013,7 @@ Create directory buckets.IP \[bu] 2List directory buckets.PP-See the --s3-directory-buckets flag for more info+See the --s3-directory-bucket flag for more info.SS AWS Snowball Edge.PPAWS Snowball (https://aws.amazon.com/snowball/) is a hardware appliance@@ -45744,6 +48274,104 @@ server_side_encryption =storage_class =\f[R].fi+.SS Autonomys Auto Drive+.PP+Auto Drive (https://ai3.storage/) is permanent storage on the Autonomys+Network, with an S3-compatible API.+Deleting an object hides it, but its data and every earlier version are+kept.+Buckets are created by the first upload into them.+.PP+Use your Auto Drive API key as the \f[V]access_key_id\f[R], and any+value as the \f[V]secret_access_key\f[R].+.PP+Here is an example of making a configuration.+First run:+.IP+.nf+\f[C]+rclone config+\f[R]+.fi+.PP+This will guide you through an interactive setup process.+.IP+.nf+\f[C]+No remotes found, make a new one?+n) New remote+s) Set configuration password+q) Quit config+n/s/q> n++Enter name for new remote.+name> autodrive++Option Storage.+Type of storage to configure.+Storage> s3++Option provider.+Choose your S3 provider.+provider> Autonomys++Option env_auth.+Get AWS credentials from runtime (environment variables or EC2/ECS meta data if no env vars).+Only applies if access_key_id and secret_access_key is blank.+env_auth> false++Option access_key_id.+AWS Access Key ID.+access_key_id> YOUR_API_KEY++Option secret_access_key.+AWS Secret Access Key (password).+secret_access_key> ANY_VALUE++Option endpoint.+Endpoint for S3 API.+Choose a number from below, or type in your own value.+ 1 / Auto Drive mainnet S3 endpoint+ \[rs] (s3.auto-drive.autonomys.xyz)+endpoint> 1++Option bucket_object_lock_enabled.+Enable Object Lock when creating new buckets.+Enter a boolean value (true or false). Press Enter for the default (false).+bucket_object_lock_enabled>++Edit advanced config?+y) Yes+n) No (default)+y/n> n++Configuration complete.+Options:+- type: s3+- provider: Autonomys+- access_key_id: YOUR_API_KEY+- secret_access_key: ANY_VALUE+- endpoint: s3.auto-drive.autonomys.xyz+Keep this \[dq]autodrive\[dq] remote?+y) Yes this is OK (default)+e) Edit this remote+d) Delete this remote+y/e/d> y+\f[R]+.fi+.PP+The resulting configuration file should look like:+.IP+.nf+\f[C]+[autodrive]+type = s3+provider = Autonomys+access_key_id = YOUR_API_KEY+secret_access_key = ANY_VALUE+endpoint = s3.auto-drive.autonomys.xyz+\f[R]+.fi.SS BizflyCloud.PPBizfly Cloud Simple Storage (https://bizflycloud.vn/simple-storage) is@@ -45810,6 +48438,14 @@ supply the parameter \f[V]--s3-upload-cutoff 0\f[R] or put this in theconfig file as \f[V]upload_cutoff 0\f[R] to work around a bug whichcauses uploading of small files to fail..PP+Rclone defaults \f[V]sign_accept_encoding\f[R] to false for the Ceph+provider.+That avoids \f[V]SignatureDoesNotMatch\f[R] errors when a reverse proxy+(for example Cloudflare) rewrites the \f[V]Accept-Encoding\f[R] request+header.+You can override this with \f[V]--s3-sign-accept-encoding\f[R] /+\f[V]sign_accept_encoding\f[R] if needed.+.PPNote also that Ceph sometimes puts \f[V]/\f[R] in the passwords it givesusers.If you read the secret access key using the command line tools you will@@ -45841,7 +48477,7 @@ it will work fine..SS China Mobile Ecloud Elastic Object Storage (EOS).PPHere is an example of making an China Mobile Ecloud Elastic Object-Storage (EOS) (https:///ecloud.10086.cn/home/product-introduction/eos/)+Storage (EOS) (https://ecloud.10086.cn/home/product-introduction/eos/)configuration.First run:.IP@@ -46095,6 +48731,126 @@ d) Delete this remotey/e/d> y\f[R].fi+.SS Clever Cloud Cellar+.PP+Cellar (https://www.clever.cloud/product/cellar-object-storage/) is the+S3-compatible object storage service of Clever+Cloud (https://www.clever.cloud/), hosted in France.+Credentials and the endpoint are shown in the Cellar add-on dashboard,+or in the \f[V]CELLAR_ADDON_KEY_ID\f[R],+\f[V]CELLAR_ADDON_KEY_SECRET\f[R] and \f[V]CELLAR_ADDON_HOST\f[R]+environment variables of applications linked to it.+.PP+Cellar does not apply Object Lock headers sent when initiating a+multipart upload.+When setting Object Lock retention or legal hold options, also set+\f[V]--s3-object-lock-set-after-upload\f[R]: the lock is then applied+with separate API calls after the upload, so the object is not protected+by these settings until those calls succeed.+.PP+Cellar supports server-side encryption with customer-provided keys+(SSE-C), set with \f[V]sse_customer_algorithm\f[R] and+\f[V]sse_customer_key_base64\f[R].+It rejects server-side copies of SSE-C objects unless they are done as+multipart copies, so use \f[V]--no-update-modtime\f[R] to avoid+modification time updates, and expect server-side moves and renames of+encrypted objects smaller than \f[V]--s3-copy-cutoff\f[R] to fail.+See the Cellar SSE-C+documentation (https://www.clever.cloud/developers/doc/deploy/storage/cellar/#server-side-encryption-with-customer-keys-sse-c).+.PP+First run:+.IP+.nf+\f[C]+rclone config+\f[R]+.fi+.IP+.nf+\f[C]+This will guide you through an interactive setup process:++n) New remote+n/s/q> n++Enter name for new remote.+name> cellar++Option Storage.+Type of storage to configure.+Choose a number from below, or type in your own value.+[snip]+XX / Amazon S3 Compliant Storage Providers including AWS, ..., Cellar, ... and others+ \[rs] (s3)+Storage> s3++Option provider.+Choose your S3 provider.+[snip]+XX / Clever Cloud Cellar Object Storage+ \[rs] (Cellar)+provider> Cellar++Option env_auth.+Get AWS credentials from runtime (environment variables or EC2/ECS meta data if no env vars).+ 1 / Enter AWS credentials in the next step.+ \[rs] (false)+ 2 / Get AWS credentials from the environment (env vars or IAM).+ \[rs] (true)+env_auth> 1++Option access_key_id.+AWS Access Key ID.+access_key_id> ACCESS_KEY++Option secret_access_key.+AWS Secret Access Key (password).+secret_access_key> SECRET_KEY++Option endpoint.+Endpoint for S3 API.+ 1 / Paris, France+ \[rs] (cellar-c2.services.clever-cloud.com)+endpoint> 1++Option acl.+Canned ACL used when creating buckets and storing or copying objects.+[snip]+acl> 1++Option bucket_object_lock_enabled.+Enable Object Lock when creating new buckets.+bucket_object_lock_enabled>++Edit advanced config?+y/n> n++Configuration complete.+Options:+- type: s3+- provider: Cellar+- access_key_id: ACCESS_KEY+- secret_access_key: SECRET_KEY+- endpoint: cellar-c2.services.clever-cloud.com+- acl: private+Keep this \[dq]cellar\[dq] remote?+y/e/d> y+\f[R]+.fi+.PP+This will leave the config file looking like this.+.IP+.nf+\f[C]+[cellar]+type = s3+provider = Cellar+access_key_id = ACCESS_KEY+secret_access_key = SECRET_KEY+endpoint = cellar-c2.services.clever-cloud.com+acl = private+\f[R]+.fi.SS Cloudflare R2.PPCloudflare R2 (https://blog.cloudflare.com/r2-open-beta/) Storage allows@@ -47569,6 +50325,139 @@ endpoint = eu-central-2.storage.impossibleapi.netacl = private\f[R].fi+.SS Infomaniak Swiss Backup+.PP+Infomaniak Swiss+Backup (https://www.infomaniak.com/en/swiss-backup/cloud-storage) is a+cloud backup space in Switzerland that you can use however you want with+the script or software of your choice that is compatible with Swift, S3+or SFTP.+.SS Prerequisites+.PP+Create and setup your Swiss Backup product following this+documentation (https://www.infomaniak.com/en/support/faq/2546/create-an-s3-location-on-swiss-backup).+Memorize the generated \f[V]Endpoint\f[R], \f[V]Access Key\f[R], and+\f[V]Secret Key\f[R]:+.SS Configuration+.PP+To configure access to an S3 location on Swiss Backup, follow the steps+below:+.IP "1." 3+Run \f[V]rclone config\f[R] and select \f[V]n\f[R] for a new remote.+.IP+.nf+\f[C]+ rclone config+ No remotes found, make a new one?+ n) New remote+ s) Set configuration password+ q) Quit config+ n/s/q> n+\f[R]+.fi+.IP "2." 3+Give the name of the configuration.+For example, name it \[aq]swissbackup\[aq].+.IP+.nf+\f[C]+ name> swissbackup+\f[R]+.fi+.IP "3." 3+Select \f[V]s3\f[R] storage.+.IP+.nf+\f[C]+ Type of storage to configure.+ XX / Amazon S3 Compliant Storage Providers including AWS, ... - Tier 1+ \[rs] (s3)+ Storage> s3+\f[R]+.fi+.IP "4." 3+Select \f[V]Infomaniak\f[R] provider.+.IP+.nf+\f[C]+ Choose your S3 provider.+ xx / Infomaniak Swiss Backup+ \[rs] (Infomaniak)+ provider> Infomaniak+\f[R]+.fi+.IP "5." 3+Enter your Access Key and Secret Key.+\f[I]You can generate new ones by clicking \[aq]Generate new keys\[aq]+on your Swiss Backup product dashboard.\f[R]+.IP+.nf+\f[C]+ Option env_auth.+ 1 / Enter AWS credentials in the next step+ \[rs] (false)+ 2 / Get AWS credentials from the environment (env vars or IAM)+ \[rs] (true)+ env_auth> 1++ AWS Access Key ID.+ access_key_id> {YOUR-ACCESS-ID}++ AWS Secret Access Key (password)+ secret_access_key> {YOUR-ACCESS-SECRET}+\f[R]+.fi+.IP "6." 3+Select \f[V]us-east-1\f[R] region for compatibility.+.IP+.nf+\f[C]+ Region to connect to.+ / Default region for compatibility.+ 1 | All data is stored in Switzerland.+ \[rs] (us-east-1)+ region> 1+\f[R]+.fi+.IP "7." 3+Select the endpoint corresponding to the one generated with your Access+and Secret Keys.+.IP+.nf+\f[C]+ Endpoint for S3 API.+ 1 / Swiss Backup Location 01+ \[rs] (s3.swiss-backup01.infomaniak.com)+ 2 / Swiss Backup Location 02+ \[rs] (s3.swiss-backup02.infomaniak.com)+ 3 / Swiss Backup Location 03+ \[rs] (s3.swiss-backup03.infomaniak.com)+ 4 / Swiss Backup Location 04+ \[rs] (s3.swiss-backup04.infomaniak.com)+ endpoint> 1+\f[R]+.fi+.IP "8." 3+Adjust other parameters and advanced config at your convenience.+Leave blank if not sure.+.IP "9." 3+You should obtain something like this.+.IP+.nf+\f[C]+ Configuration complete.+ Options:+ - type: s3+ - provider: Infomaniak+ - access_key_id: xxxxxxxxxx+ - secret_access_key: xxxxxxxxxx+ - region: us-east-1+ - endpoint: s3.swiss-backup01.infomaniak.com+\f[R]+.fi+.IP "10." 4+Press \f[V]y\f[R] to validate.+Then \f[V]q\f[R] to quit..SS Intercolo Object Storage.PPIntercolo Object Storage (https://intercolo.de/object-storage) offers@@ -48361,6 +51250,11 @@ secret_access_key = SECRET_ACCESS_KEYendpoint = eu-central-1.linodeobjects.com\f[R].fi+.PP+Linode Object Storage is Ceph-backed.+Rclone defaults \f[V]sign_accept_encoding\f[R] to false for this+provider so SigV4 requests do not fail with+\f[V]SignatureDoesNotMatch\f[R] (see the Ceph section)..SS Magalu.PPHere is an example of making a Magalu Object@@ -48681,6 +51575,174 @@ So once set up, for example, to copy files into a bucketrclone copy /path/to/files minio:bucket\f[R].fi+.SS Neon Object Storage+.PP+Neon Object Storage (https://neon.com/docs/storage/overview) is S3+compatible object storage in a Neon project.+Each branch of the project has its own copy of the buckets and its own+S3 endpoint.+The endpoint looks like+\f[V]https://<branch_id>.storage.<cell>.<region>.aws.neon.tech\f[R].+.PP+In the Neon Console, click \f[B]Connect\f[R], open the \f[B]Storage\f[R]+tab and choose \f[B]Parameters only\f[R].+Use \f[V]AWS_ACCESS_KEY_ID\f[R], \f[V]AWS_SECRET_ACCESS_KEY\f[R],+\f[V]AWS_ENDPOINT_URL_S3\f[R] and \f[V]AWS_REGION\f[R] as+\f[V]access_key_id\f[R], \f[V]secret_access_key\f[R], \f[V]endpoint\f[R]+and \f[V]region\f[R].+To make a separate credential, use \f[B]Settings\f[R] >+\f[B]Credentials\f[R] or run+\f[V]neon credentials create --scope storage:read --scope storage:write\f[R].+.PP+A credential works on the branch it was made on and on that branch\[aq]s+child branches.+To use a child branch, change only \f[V]endpoint\f[R].+.PP+Neon also has these limits (see the S3 compatibility+docs (https://neon.com/docs/storage/s3-compatibility)):+.IP \[bu] 2+The maximum object size is 5 GiB for a single upload and 10 GiB for a+multipart upload.+.IP \[bu] 2+Neon rejects object keys containing control characters with+\f[V]400 InvalidURI\f[R].+To store files with these names, add \f[V]Ctl,Del\f[R] to the encoding,+for example \f[V]encoding = Slash,InvalidUtf8,Dot,Ctl,Del\f[R].+.IP \[bu] 2+Neon doesn\[aq]t support ACLs over the S3 API.+Leave \f[V]acl\f[R] and \f[V]bucket_acl\f[R] empty and set bucket access+(\f[V]private\f[R] or \f[V]public_read\f[R]) in Neon.+.IP \[bu] 2+Neon stores versioning and lifecycle configurations but doesn\[aq]t+enforce them.+.IP \[bu] 2+Neon may throttle heavy use with \f[V]503 SlowDown\f[R] errors.+rclone retries these.+.PP+Here is an example of making a Neon Object Storage configuration.+First run:+.IP+.nf+\f[C]+rclone config+\f[R]+.fi+.PP+This will guide you through an interactive setup process.+.IP+.nf+\f[C]+No remotes found, make a new one?+n) New remote+s) Set configuration password+q) Quit config+n/s/q> n++Enter name for new remote.+name> neon++Option Storage.+Type of storage to configure.+Choose a number from below, or type in your own value.+[snip]+XX / Amazon S3 Compliant Storage Providers including AWS, ... Neon, ...+ \[rs] (s3)+[snip]+Storage> s3++Option provider.+Choose your S3 provider.+Choose a number from below, or type in your own value.+Press Enter to leave empty.+[snip]+XX / Neon Object Storage+ \[rs] (Neon)+[snip]+provider> Neon++Option env_auth.+Get AWS credentials from runtime (environment variables or EC2/ECS meta data if no env vars).+Only applies if access_key_id and secret_access_key is blank.+Choose a number from below, or type in your own boolean value (true or false).+Press Enter for the default (false).+ 1 / Enter AWS credentials in the next step.+ \[rs] (false)+ 2 / Get AWS credentials from the environment (env vars or IAM).+ \[rs] (true)+env_auth>++Option access_key_id.+AWS Access Key ID.+Leave blank for anonymous access or runtime credentials.+Enter a value. Press Enter to leave empty.+access_key_id> NEON_ACCESS_KEY_ID++Option secret_access_key.+AWS Secret Access Key (password).+Leave blank for anonymous access or runtime credentials.+Enter a value. Press Enter to leave empty.+secret_access_key> NEON_SECRET_ACCESS_KEY++Option region.+Region to connect to.+Leave blank if you are using an S3 clone and you don\[aq]t have a region.+Choose a number from below, or type in your own value.+Press Enter to leave empty.+ 1 / US East (N. Virginia)+ \[rs] (us-east-1)+ 2 / US East (Ohio)+ \[rs] (us-east-2)+ 3 / Europe (Frankfurt)+ \[rs] (eu-central-1)+ 4 / Asia Pacific (Singapore)+ \[rs] (ap-southeast-1)+region> us-east-2++Option endpoint.+Endpoint for S3 API.+Required when using an S3 clone.+Enter a value. Press Enter to leave empty.+endpoint> https://<branch_id>.storage.<cell>.<region>.aws.neon.tech++Option bucket_object_lock_enabled.+Enable Object Lock when creating new buckets.+Enter a boolean value (true or false). Press Enter for the default (false).+bucket_object_lock_enabled>++Edit advanced config?+y) Yes+n) No (default)+y/n> n++Configuration complete.+Options:+- type: s3+- provider: Neon+- access_key_id: NEON_ACCESS_KEY_ID+- secret_access_key: NEON_SECRET_ACCESS_KEY+- region: us-east-2+- endpoint: https://<branch_id>.storage.<cell>.<region>.aws.neon.tech+Keep this \[dq]neon\[dq] remote?+y) Yes this is OK (default)+e) Edit this remote+d) Delete this remote+y/e/d> y+\f[R]+.fi+.PP+This will leave the config file looking like this.+.IP+.nf+\f[C]+[neon]+type = s3+provider = Neon+access_key_id = NEON_ACCESS_KEY_ID+secret_access_key = NEON_SECRET_ACCESS_KEY+region = us-east-2+endpoint = https://<branch_id>.storage.<cell>.<region>.aws.neon.tech+\f[R]+.fi.SS Netease NOS.PPFor Netease NOS configure as per the configurator@@ -49934,6 +52996,143 @@ use_multipart_uploads = falseNote that setting \f[V]use_multipart_uploads = false\f[R] is to workaround a bug (https://rclone.org/commands/rclone_serve_s3/#bugs) whichwill be fixed in due course.+.SS RelAix.Storage+.PP+RelAix.Storage is S3 compatible object storage from RelAix+Networks (https://www.relaix.net/), operated in its own data centres in+Aachen, Germany.+There are two locations, Aachen-1 and Aachen-2, each with its own+endpoint.+Buckets and data stay at the location of the endpoint you use.+.PP+You get an access key and a secret key from RelAix Networks.+Then run:+.IP+.nf+\f[C]+rclone config+\f[R]+.fi+.PP+This will guide you through an interactive setup process.+.IP+.nf+\f[C]+No remotes found, make a new one?+n) New remote+s) Set configuration password+q) Quit config+n/s/q> n++Enter name for new remote.+name> my-relaix++Option Storage.+Type of storage to configure.+Choose a number from below, or type in your own value.+[snip]+XX / Amazon S3 Compliant Storage Providers including AWS, ...+ \[rs] (s3)+[snip]+Storage> s3++Option provider.+Choose your S3 provider.+Choose a number from below, or type in your own value.+Press Enter to leave empty.+[snip]+XX / RelAix.Storage+ \[rs] (RelAix)+[snip]+provider> RelAix++Option env_auth.+Get AWS credentials from runtime (environment variables or EC2/ECS meta data if no env vars).+Only applies if access_key_id and secret_access_key is blank.+Choose a number from below, or type in your own boolean value (true or false).+Press Enter for the default (false).+ 1 / Enter AWS credentials in the next step.+ \[rs] (false)+ 2 / Get AWS credentials from the environment (env vars or IAM).+ \[rs] (true)+env_auth>++Option access_key_id.+AWS Access Key ID.+Leave blank for anonymous access or runtime credentials.+Enter a value. Press Enter to leave empty.+access_key_id> ACCESS_KEY++Option secret_access_key.+AWS Secret Access Key (password).+Leave blank for anonymous access or runtime credentials.+Enter a value. Press Enter to leave empty.+secret_access_key> SECRET_KEY++Option endpoint.+Endpoint for S3 API.+Required when using an S3 clone.+Choose a number from below, or type in your own value.+Press Enter to leave empty.+ 1 / Aachen-1, Germany+ \[rs] (s3-gw1.relaix.net)+ 2 / Aachen-2, Germany+ \[rs] (s3-gw2.relaix.net)+endpoint> 1++Option acl.+Canned ACL used when creating buckets and storing or copying objects.+[snip]+acl>++Option bucket_object_lock_enabled.+Enable Object Lock when creating new buckets.+Enter a boolean value (true or false). Press Enter for the default (false).+bucket_object_lock_enabled>++Edit advanced config?+y) Yes+n) No (default)+y/n> n++Configuration complete.+Options:+- type: s3+- provider: RelAix+- access_key_id: ACCESS_KEY+- secret_access_key: SECRET_KEY+- endpoint: s3-gw1.relaix.net+Keep this \[dq]my-relaix\[dq] remote?+y) Yes this is OK (default)+e) Edit this remote+d) Delete this remote+y/e/d> y+\f[R]+.fi+.PP+This will leave the config file looking like this.+.IP+.nf+\f[C]+[my-relaix]+type = s3+provider = RelAix+access_key_id = ACCESS_KEY+secret_access_key = SECRET_KEY+endpoint = s3-gw1.relaix.net+\f[R]+.fi+.PP+Object Lock can only be enabled when a bucket is created.+To create buckets with Object Lock, set+\f[V]bucket_object_lock_enabled = true\f[R] before running+\f[V]rclone mkdir\f[R].+.PP+To lock objects on upload with \f[V]object_lock_mode\f[R], also set+\f[V]object_lock_set_after_upload = true\f[R].+RelAix.Storage does not yet apply Object Lock headers sent with+multipart uploads, so rclone needs to set the lock in a separate call+after the upload..SS Scaleway.PPScaleway (https://www.scaleway.com/object-storage/) The Object Storage@@ -52027,6 +55226,28 @@ For example:mksquashfs 100files 100files.sqfs -comp zstd -b 1M\f[R].fi+.SS Symlinks+.PP+Symlinks stored in an archive are skipped unless the+\f[V]-l/--links\f[R] flag is supplied, as they are on the other+backends.+With the flag they are listed with a \f[V].rclonelink\f[R] suffix and+read back as the path they point at, the same convention the+local (https://rclone.org/local/#symlinks-junction-points) backend uses,+so they can be copied back as symlinks:+.IP+.nf+\f[C]+$ rclone lsl --links :archive:/tmp/test.zip+ 5 2026-09-28 14:16:59.000000000 src/file.txt+ 8 2026-09-28 14:16:59.000000000 src/link.txt.rclonelink+$ rclone cat --links :archive:/tmp/test.zip/src/link.txt.rclonelink+file.txt+$ rclone copy --links :archive:/tmp/test.zip /tmp/out+$ ls -l /tmp/out/src/link.txt+lrwxr-xr-x 1 user user 8 Sep 28 14:16 /tmp/out/src/link.txt -> file.txt+\f[R]+.fi.SS Limitations.PPFiles in the archive backend are read only.@@ -52885,6 +56106,11 @@ If the custom endpoint rewrites the requests for authentication, e.g.,in Cloudflare Workers, this header needs to be handled properly.Leave blank if you want to use the endpoint provided by Backblaze..PP+Rclone will not follow a redirect from this endpoint which downgrades+HTTPS to HTTP, and will not send the \[dq]Authorization\[dq] header or+the SSE-C key headers once a redirect has left the host and port of this+endpoint.+.PPThe URL provided here SHOULD have the protocol and SHOULD NOT have atrailing slash or specify the /file/bucket subpath as rclone willrequest files with \[dq]{download_url}/file/{bucket_name}/{path}\[dq].@@ -53009,7 +56235,7 @@ Advanced Encryption Standard (256 bits key length)To use SSE-C, you may provide the secret encryption key encoded in aUTF-8 compatible string to encrypt/decrypt your data.PP-Alternatively you can provide --sse-customer-key-base64.+Alternatively you can provide --b2-sse-customer-key-base64..PPProperties:.IP \[bu] 2@@ -53035,7 +56261,7 @@ NoneTo use SSE-C, you may provide the secret encryption key encoded inBase64 format to encrypt/decrypt your data.PP-Alternatively you can provide --sse-customer-key.+Alternatively you can provide --b2-sse-customer-key..PPProperties:.IP \[bu] 2@@ -53386,10 +56612,11 @@ just set..SS Invalid refresh token.PPAccording to the box-docs (https://developer.box.com/v2.0/docs/oauth-20#section-6-using-the-access-and-refresh-tokens):+docs (https://developer.box.com/guides/authentication/tokens/refresh/):.RS.PP-Each refresh_token is valid for one use in 60 days.+A Refresh Token is valid for 60 days and can be used to obtain a new+Access Token and Refresh Token only once..RE.PPThis means that if you@@ -57109,6 +60336,16 @@ filenames with the same name will encrypt the same.IP \[bu] 2filenames which start the same won\[aq]t have a common prefix.PP+A version string of the form \f[V]-vYYYY-MM-DD-HHMMSS-NNN\f[R] on the+end of a file name (as added by \f[V]--b2-versions\f[R] /+\f[V]--s3-versions\f[R]) is left in plain text so that versioned files+can be found.+Directory names are encrypted in full.+Rclone before v1.76 left such a suffix in plain text on directory names+too, so a directory named like this created by an older rclone will+appear in listings with a warning but can\[aq]t be opened or removed+until renamed on the underlying remote to the name given in the warning.+.PPThis uses a 32 byte key (256 bits) and a 16 byte (128 bits) IV both ofwhich are derived from the user password..PP@@ -57823,6 +61060,356 @@ Only new parameters need be passed as the values will default to thosecurrently in use..PPIt doesn\[aq]t return anything.+.SH dosya.dev+.PP+dosya.dev (https://dosya.dev) is a cloud file storage, sharing, and+synchronization platform with workspace-based multi-tenancy, file+versioning, and team collaboration features.+.SS Configuration+.PP+Here is an example of how to make a remote called \f[V]remote\f[R].+First run:+.IP+.nf+\f[C]+rclone config+\f[R]+.fi+.PP+This will guide you through an interactive setup process:+.IP+.nf+\f[C]+No remotes found, make a new one?+n) New remote+s) Set configuration password+q) Quit config+n/s/q> n++Enter name for new remote.+name> remote++Option Storage.+Type of storage to configure.+Choose a number from below, or type in your own value.+XX / dosya.dev+ \[rs] (dosya)+Storage> dosya++Option api_key.+Your API Key, get it from https://dosya.dev/profile under \[dq]API keys\[dq].+Enter a value.+api_key> dos_YOURAPIKEY++Option workspace_id.+Your workspace ID.++Leave this blank to use the one workspace the API key can reach, which+is what a key restricted to a single workspace does. If the key can+reach several, rclone will refuse and list their IDs for you to choose+from.+Enter a value. Press Enter to leave empty.+workspace_id>++Edit advanced config?+y) Yes+n) No (default)+y/n> n++Configuration complete.+Options:+- type: dosya+- api_key: dos_YOURAPIKEY+Keep this \[dq]remote\[dq] remote?+y) Yes this is OK (default)+e) Edit this remote+d) Delete this remote+y/e/d> y+\f[R]+.fi+.PP+Once configured you can then use \f[V]rclone\f[R] like this (replace+\f[V]remote\f[R] with the name you gave your remote):+.PP+List directories and files in the top level of your dosya workspace+.IP+.nf+\f[C]+rclone lsf remote:+\f[R]+.fi+.PP+To copy a local directory to a dosya directory called backup+.IP+.nf+\f[C]+rclone copy /home/source remote:backup+\f[R]+.fi+.PP+The dosya backend uploads large files in parts and stores them as single+files that remain accessible through the web app and other clients.+If you previously used a WebDAV remote with a chunker overlay for large+uploads, you can use this backend without the overlay.+.SS Getting your API key+.IP "1." 3+Log in to dosya.dev (https://dosya.dev)+.IP "2." 3+Go to Profile (https://dosya.dev/profile) > \f[B]API keys\f[R]+.IP "3." 3+Create a new API key+.IP "4." 3+Copy the key (starts with \f[V]dos_\f[R])+.PP+A key can be restricted to a single workspace when you create it, which+is the simplest thing to do for rclone: \f[V]workspace_id\f[R] can then+be left out of the config entirely.+.SS Workspaces+.PP+Leave \f[V]workspace_id\f[R] blank and rclone asks which workspaces the+key can reach.+A key restricted to one workspace, and an account that has only one,+both resolve without any configuration.+If the key can reach several, rclone refuses and lists their IDs so you+can set the one you want:+.IP+.nf+\f[C]+this API key can reach 2 workspaces, so workspace_id must be set to one of:+ ws_AAAAAAAA (Work)+ ws_BBBBBBBB (Personal)+\f[R]+.fi+.PP+To access several workspaces at once, create a separate remote for each:+.IP+.nf+\f[C]+rclone config create work dosya api_key=dos_XXX workspace_id=ws_WORK+rclone config create personal dosya api_key=dos_XXX workspace_id=ws_PERSONAL+\f[R]+.fi+.PP+Then use them independently:+.IP+.nf+\f[C]+rclone ls work:+rclone copy work:docs/ personal:backup/+\f[R]+.fi+.SS Modification times and hashes+.PP+dosya.dev supports modification times with a resolution of 1 second.+The modification time is sent with the upload, so a copied or synced+file keeps its original timestamp.+Modification times cannot be changed after upload without re-uploading+the file.+.PP+dosya.dev supports SHA-256 checksums.+The server computes the hash of a single-shot (small file) upload from+the bytes it receives, so \f[V]rclone check\f[R] and hash-based sync+work without downloading.+Large files uploaded in multiple parts have no server-side hash (the+parts are never re-read to hash the whole object), the same limitation+the S3 backend has for multipart objects; those objects report an empty+hash.+.SS Restricted filename characters+.PP+In addition to the default restricted characters+set (https://rclone.org/overview/#restricted-characters) the following+characters are also replaced:+.PP+.TS+tab(@);+l c c.+T{+Character+T}@T{+Value+T}@T{+Replacement+T}+_+T{+\[dq]+T}@T{+0x22+T}@T{+"+T}+T{+\[aq]+T}@T{+0x27+T}@T{+'+T}+T{+<+T}@T{+0x3C+T}@T{+<+T}+T{+>+T}@T{+0x3E+T}@T{+>+T}+T{+\[rs]+T}@T{+0x5C+T}@T{+\+T}+.TE+.PP+Invalid UTF-8 bytes will also be+replaced (https://rclone.org/overview/#invalid-utf8), as they can\[aq]t+be used in JSON strings.+.SS Public Links+.PP+dosya.dev supports \f[V]rclone link\f[R] to create public share links to+files.+You can use the \f[V]--expire\f[R] flag to specify how long the link+should be valid.+.IP+.nf+\f[C]+rclone link remote:path/to/file.txt+rclone link --expire 7d remote:path/to/file.txt+\f[R]+.fi+.SS Standard options+.PP+Here are the Standard options specific to dosya (dosya.dev).+.SS --dosya-api-key+.PP+Your API Key, get it from https://dosya.dev/profile under \[dq]API+keys\[dq].+.PP+Properties:+.IP \[bu] 2+Config: api_key+.IP \[bu] 2+Env Var: RCLONE_DOSYA_API_KEY+.IP \[bu] 2+Type: string+.IP \[bu] 2+Required: true+.SS --dosya-workspace-id+.PP+Your workspace ID.+.PP+Leave this blank to use the one workspace the API key can reach, which+is what a key restricted to a single workspace does.+If the key can reach several, rclone will refuse and list their IDs for+you to choose from.+.PP+Properties:+.IP \[bu] 2+Config: workspace_id+.IP \[bu] 2+Env Var: RCLONE_DOSYA_WORKSPACE_ID+.IP \[bu] 2+Type: string+.IP \[bu] 2+Required: false+.SS Advanced options+.PP+Here are the Advanced options specific to dosya (dosya.dev).+.SS --dosya-api-url+.PP+API base URL.+.PP+Properties:+.IP \[bu] 2+Config: api_url+.IP \[bu] 2+Env Var: RCLONE_DOSYA_API_URL+.IP \[bu] 2+Type: string+.IP \[bu] 2+Default: \[dq]https://api.dosya.dev\[dq]+.SS --dosya-upload-concurrency+.PP+Concurrency for multipart uploads.+.PP+This is the number of parts of the same file that are uploaded+concurrently.+The first part is always uploaded on its own, because the server creates+the upload when it receives it.+.PP+Each part in flight is held in memory, so memory use grows with this+value times the part size the server chooses.+.PP+If you are uploading large files over high-speed links and these uploads+do not fully utilize your bandwidth, then increasing this may help to+speed up the transfers.+.PP+Properties:+.IP \[bu] 2+Config: upload_concurrency+.IP \[bu] 2+Env Var: RCLONE_DOSYA_UPLOAD_CONCURRENCY+.IP \[bu] 2+Type: int+.IP \[bu] 2+Default: 4+.SS --dosya-encoding+.PP+The encoding for the backend.+.PP+See the encoding section in the+overview (https://rclone.org/overview/#encoding) for more info.+.PP+Properties:+.IP \[bu] 2+Config: encoding+.IP \[bu] 2+Env Var: RCLONE_DOSYA_ENCODING+.IP \[bu] 2+Type: Encoding+.IP \[bu] 2+Default:+Slash,LtGt,DoubleQuote,SingleQuote,BackSlash,Del,Ctl,LeftSpace,RightSpace,InvalidUtf8,Dot+.SS --dosya-description+.PP+Description of the remote.+.PP+Properties:+.IP \[bu] 2+Config: description+.IP \[bu] 2+Env Var: RCLONE_DOSYA_DESCRIPTION+.IP \[bu] 2+Type: string+.IP \[bu] 2+Required: false+.SS Limitations+.PP+dosya.dev only supports filenames up to 255 characters in length.+.PP+Empty files (0 bytes) cannot be uploaded to dosya.dev.+.PP+Upload requires knowing the file size in advance, so streaming uploads+of unknown-size files are not supported.+.PP+Each workspace limits the number of uploads a user can have in flight at+once (5 by default, configurable in the workspace settings).+When \f[V]--transfers\f[R] is set higher than that limit, rclone waits+for a free upload slot rather than failing the transfer, retrying every+5 seconds for up to 10 minutes.+.SS Deleting files and directories+.PP+rclone deletes files and directories permanently, in a single request+(\f[V]DELETE ?permanent=1\f[R]), bypassing the workspace trash.+Removing a directory also asks the server whether it is empty+(\f[V]?if_empty=1\f[R]) rather than listing it first..SH Drime.PPDrime (https://drime.cloud/) is a cloud storage and transfer service@@ -60126,6 +63713,93 @@ any excess files in the directory.rclone sync --interactive /home/local/directory remote:dir\f[R].fi+.SS Restricted filename characters+.PP+In addition to the default restricted characters+set (https://rclone.org/overview/#restricted-characters) the following+characters are also replaced:+.PP+.TS+tab(@);+l c c.+T{+Character+T}@T{+Value+T}@T{+Replacement+T}+_+T{+\[rs]+T}@T{+0x5C+T}@T{+\+T}+.TE+.PP+File names can also not end with the following characters.+These only get replaced if they are the last character in the name:+.PP+.TS+tab(@);+l c c.+T{+Character+T}@T{+Value+T}@T{+Replacement+T}+_+T{+SP+T}@T{+0x20+T}@T{+␠+T}+T{+LF+T}@T{+0x0A+T}@T{+␊+T}+T{+CR+T}@T{+0x0D+T}@T{+␍+T}+T{+HT+T}@T{+0x09+T}@T{+␉+T}+T{+VT+T}@T{+0x0B+T}@T{+␋+T}+.TE+.PP+Invalid UTF-8 bytes will also be+replaced (https://rclone.org/overview/#invalid-utf8).+.PP+Files.com treats the replacements \f[V]/\f[R] and \f[V]\\f[R] as if+they were \f[V]/\f[R] and \f[V]\[rs]\f[R], and the names \f[V].\f[R]+and \f[V]..\f[R] as if they were \f[V].\f[R] and \f[V]..\f[R], and+rejects them with \f[V]Invalid path\f[R].+This means that files with \f[V]/\f[R] or \f[V]\[rs]\f[R] in their+names, or which are called \f[V].\f[R] or \f[V]..\f[R], can\[aq]t be+stored on Files.com..SS Hashes.PPIn December 2024 \f[V]files.com\f[R] started supporting more@@ -60363,10 +64037,10 @@ although it is common to use either the password \[dq]anonymous\[dq] or\[dq]guest\[dq].Some servers require the use of a valid e-mail address as password..PP-Using on-the-fly or connection-string (https://rclone.org/docs/#connection-strings) remotes makes it-easy to access such servers, without requiring any configuration in-advance.+Using on-the-fly (https://rclone.org/docs/#backend-path-to-dir) or+connection string (https://rclone.org/docs/#connection-strings) remotes+makes it easy to access such servers, without requiring any+configuration in advance.The following are examples of that:.IP.nf@@ -64166,7 +67840,7 @@ usage is using the flag form --drive-size-as-quota when doing rclonels/lsl/lsf/lsjson/etc only..PPIf you do use this flag for syncing (not recommended) then you will need-to use --ignore size also.+to use --ignore-size also..PPProperties:.IP \[bu] 2@@ -65385,8 +69059,8 @@ You should now see the three scopes on your Data access page.Now press save at the bottom!.RE.IP " 6." 4-After adding scopes, click Audience Scroll down and click \[dq]+ Add-users\[dq].+After adding scopes, click Audience.+Scroll down and click \[dq]+ Add users\[dq].Add yourself as a test user and press save..IP " 7." 4Go to Overview on the left panel, click \[dq]Create OAuth client\[dq].@@ -65404,6 +69078,19 @@ same Google Workspace.)Go to \[dq]Audience\[dq] and then click \[dq]PUBLISH APP\[dq] button andconfirm.Add yourself as a test user if you haven\[aq]t already.+.RS 4+.PP+(If the \[dq]PUBLISH APP\[dq] button is greyed out, Google now requires+a homepage URL and a privacy policy URL before publishing, even for a+personal single-user app.+Go to \[dq]Branding\[dq] in the left panel, fill in \[dq]Application+home page\[dq] and \[dq]Application privacy policy link\[dq] - a free+GitHub Pages site is fine for these if you don\[aq]t have a domain of+your own - then \[dq]Authorized domains\[dq] with the domain you used,+and click Save.+Return to \[dq]Audience\[dq] and the \[dq]PUBLISH APP\[dq] button should+now be available.)+.RE.IP "10." 4Provide the noted client ID and client secret to rclone..IP "11." 4@@ -66814,15 +70501,24 @@ Preparation section in the \[dq]Huawei Account Service DevelopmentGuide\[dq]..SS Enable Drive Application Switch.IP "1." 3-Log in to the AppGallery Connect website and click \[dq]My-Projects\[dq].+Log in to AppGallery Connect with your Huawei account..IP "2." 3-Find your project in the project list, select the application that needs-to access Drive in the project, and click \[dq]API Management\[dq].+Create a project or select an existing project..IP "3." 3-In the \[dq]API Management\[dq] area of the \[dq]Project Settings\[dq]-page, click the switch on the right side of Drive Kit to ensure the-switch is in the open state.+Add a web application to the project.+.IP "4." 3+Open \[dq]Project settings\[dq] for the application and copy the OAuth+2.0 Client ID and Client Secret from the \[dq]General\[dq] tab.+.IP "5." 3+Open \[dq]API Management\[dq] for the project and enable \[dq]Cloud+Drive\[dq].+.IP "6." 3+Add \f[V]http://127.0.0.1:53682/\f[R] as a redirect URI for the OAuth+2.0 Client ID.+.PP+The older Huawei Developer Console may show some of the same credential+information, but AppGallery Connect is the place to create the project,+application, API permission, and redirect URI used by rclone..SS Sign Huawei Cloud Drive Privacy Notice and User Agreement.PPGuide users to visit <https://cloud.huawei.com/>.@@ -66838,27 +70534,17 @@ Development Guide\[dq]..PPKey steps summary:.IP "1." 3-Log into the Huawei Developer-Console (https://developer.huawei.com/consumer/en/console) with your-Huawei account.+Create or select a project in AppGallery Connect..IP "2." 3-Create a new project or select an existing project.+Add a web application to the project..IP "3." 3-Go to \[dq]Manage APIs\[dq] and enable the \[dq]Drive Kit\[dq] API.+Enable Cloud Drive in API Management..IP "4." 3-Click on \[dq]Credentials\[dq] in the left side panel.+Configure \f[V]http://127.0.0.1:53682/\f[R] as the OAuth redirect URI..IP "5." 3-Click \[dq]Create Credentials\[dq] and choose \[dq]OAuth client ID\[dq].+Copy the OAuth 2.0 Client ID and Client Secret from the application..IP "6." 3-Choose \[dq]Web application\[dq] as the application type.-.IP "7." 3-Add \f[V]http://localhost:53682/\f[R] to the \[dq]Authorized redirect-URIs\[dq].-.IP "8." 3-Click \[dq]Create\[dq] and note down the Client ID and Client Secret.-.IP "9." 3-Provide the noted client ID and client secret to rclone during-configuration.+Provide the client ID and client secret to rclone during configuration..SS Scopes.PPRclone uses the following OAuth2 scopes when accessing Huawei Drive:@@ -68335,11 +72021,13 @@ path \f[V]fix\f[R], the resolved URL will be\f[V]/fix\f[R] the resolved URL will be\f[V]https://beta.rclone.org/fix\f[R] as the absolute path is resolvedfrom the root of the domain.+The path can\[aq]t change the scheme, host or user of the configured+url, so a path such as \f[V]//other.example.com/fix\f[R] is rejected..PPIf the path following the \f[V]remote:\f[R] ends with \f[V]/\f[R] itwill be assumed to point to a directory.If the path does not end with \f[V]/\f[R], then a HEAD request is sent-and the response used to decide if it it is treated as a file or a+and the response used to decide if it is treated as a file or adirectory (run with \f[V]-vv\f[R] to see details).When --http-no-head is specified, a path without ending \f[V]/\f[R] isalways assumed to be a file.@@ -68547,6 +72235,13 @@ For example, to set a Cookie use \[aq]Cookie,name=value\[aq], orYou can set multiple headers, e.g.\[aq]\[dq]Cookie\[dq],\[dq]name=value\[dq],\[dq]Authorization\[dq],\[dq]xxx\[dq]\[aq]..PP+The headers are only sent to the host in the configured URL.+If the server redirects to another host (including a subdomain or a+different port) the headers are not sent to it, or to any further hop in+that redirect chain.+When headers are set, a redirect from https to http is refused as it+would send them in cleartext, even with --allow-https-downgrade.+.PPProperties:.IP \[bu] 2Config: headers@@ -68610,6 +72305,51 @@ Env Var: RCLONE_HTTP_NO_HEADType: bool.IP \[bu] 2Default: false+.SS --http-no-head-size+.PP+Set the reported file size for --http-no-head.+.PP+When HEAD requests are disabled, rclone normally reports an unknown file+size.+Set this option to a non-negative value to report that size for every+file, including 0.+Leave it unset (the default, shown as \[dq]off\[dq]) to keep reporting+an unknown size.+.PP+The configured size is authoritative.+If it does not match the actual file size, transfers may fail their size+check and retry unless --ignore-size is used.+It can also make multi-thread transfers request incorrect ranges and+make mounted files unreadable past the reported size.+.PP+Properties:+.IP \[bu] 2+Config: no_head_size+.IP \[bu] 2+Env Var: RCLONE_HTTP_NO_HEAD_SIZE+.IP \[bu] 2+Type: SizeSuffix+.IP \[bu] 2+Default: off+.SS --http-no-head-time+.PP+Set a default modification time for --http-no-head.+.PP+When HEAD requests are disabled, rclone normally reports an unset+modification time.+Set this option to report the specified time for every file instead.+Changing the configured time later makes every file appear modified to a+sync.+.PP+Properties:+.IP \[bu] 2+Config: no_head_time+.IP \[bu] 2+Env Var: RCLONE_HTTP_NO_HEAD_TIME+.IP \[bu] 2+Type: Time+.IP \[bu] 2+Default: off.SS --http-description.PPDescription of the remote.@@ -71245,6 +74985,396 @@ Jottacloud exhibits some inconsistent behaviours regarding deleted filesand folders which may cause Copy, Move and DirMove operations topreviously deleted paths to fail.Emptying the trash should help in such cases.+.SH kDrive+.PP+Paths are specified as \f[V]remote:path\f[R]+.PP+Paths may be as deep as required, e.g.+\f[V]remote:directory/subdirectory\f[R].+.SS Configuration+.PP+The initial setup for kDrive involves getting an OAuth2 token which you+need to do in your browser.+\f[V]rclone config\f[R] walks you through it.+.PP+kDrive uses the PKCE OAuth2 flow (RFC 7636), which means no+\f[V]client_secret\f[R] is required.+The \f[V]client_id\f[R] is optional and defaults to rclone\[aq]s+built-in application credentials.+.PP+Here is an example of how to make a remote called \f[V]remote\f[R].+First run:+.IP+.nf+\f[C]+rclone config+\f[R]+.fi+.PP+This will guide you through an interactive setup process:+.IP+.nf+\f[C]+No remotes found, make a new one?+n) New remote+s) Set configuration password+q) Quit config+n/s/q> n+name> remote+Type of storage to configure.+Choose a number from below, or type in your own value+[snip]+XX / Infomaniak kDrive+ \[rs] \[dq]kdrive\[dq]+[snip]+Storage> kdrive++Option client_id.+OAuth Client Id.+Leave blank normally.+Enter a value. Press Enter to leave empty.+client_id>++Option drive_id.+ID of the kDrive to use.+When showing a folder on kdrive, you can find the drive_id here:+https://ksuite.infomaniak.com/{account_id}/kdrive/app/drive/{drive_id}/files/...+Enter a value. Press Enter to leave empty.+drive_id> 0654321++Use web browser to automatically authenticate rclone with remote?+ * Say Y if the machine running rclone has a web browser you can use+ * Say N if running rclone on a (remote) machine without web browser access+If not sure try Y. If Y failed, try N.+y/n> y++Waiting for code...+Got code++Configuration complete.+Options:+- type: kdrive+- drive_id: 0654321++Keep this \[dq]remote\[dq] remote?+y) Yes this is OK+e) Edit this remote+d) Delete this remote+y/e/d> y+\f[R]+.fi+.PP+Once configured you can then use \f[V]rclone\f[R] like this (replace+\f[V]remote\f[R] with the name you gave your remote):+.PP+List directories in top level of your rclone+.IP+.nf+\f[C]+rclone lsd remote:+\f[R]+.fi+.PP+\f[B]Note\f[R] that by default the top level directory of a private+kDrive storage is your \[dq]Private\[dq] folder.+When used with a Pro account, a second folder \[dq]Common documents\[dq]+is also available, which can be used as the root by setting+\f[V]root_folder_id\f[R] to \f[V]common\f[R].+\f[I]These top-level folders cannot be modified\f[R].+.PP+List all the files in your kDrive+.IP+.nf+\f[C]+rclone ls remote:+\f[R]+.fi+.PP+To copy a local directory to a kDrive directory called backup+.IP+.nf+\f[C]+rclone copy /home/source remote:backup+\f[R]+.fi+.SS Modification times and hashes+.PP+kDrive allows modification times to be set on objects accurate to 1+second.+These will be used to detect whether objects need syncing or not.+.PP+kDrive supports the XXH3 checksum hash, so you can use the+\f[V]--checksum\f[R] flag.+XXH3 is only available for files uploaded as a single chunk: files+uploaded in chunks (by rclone when larger than+\f[V]--kdrive-chunk-size\f[R], or by the kDrive apps which chunk files+at their own sizes) have a \[dq]nested\[dq] hash which can\[aq]t be+compared with a plain XXH3, and rclone reports no hash for those files.+.SS Restricted filename characters+.PP+In addition to the default restricted characters+set (https://rclone.org/overview/#restricted-characters) the following+characters are also replaced:+.PP+.TS+tab(@);+l c c.+T{+Character+T}@T{+Value+T}@T{+Replacement+T}+_+T{+LeftSpace+T}@T{+0x20+T}@T{+␠+T}+T{+RightSpace+T}@T{+0x20+T}@T{+␠+T}+.TE+.PP+Invalid UTF-8 bytes will also be+replaced (https://rclone.org/overview/#invalid-utf8), as they can\[aq]t+be used in JSON strings.+.SS Overwriting files+.PP+When rclone overwrites an existing file, kDrive keeps the replaced+content as a new version of the file.+Old versions count against the drive quota, so repeatedly overwriting+large files grows the storage usage.+Manage old versions from the kDrive web app.+.SS Uploading files of unknown size+.PP+kDrive can\[aq]t upload files of unknown size, so \f[V]rclone rcat\f[R]+and streaming uploads are not supported.+.SS Deleting files+.PP+Deleted files will be moved to the trash.+Your subscription level will determine how long items stay in the trash.+\f[V]rclone cleanup\f[R] can be used to empty the trash.+.SS Root folder ID+.PP+You can set the \f[V]root_folder_id\f[R] for rclone.+This is the directory (identified by its numeric \f[V]Folder ID\f[R])+that rclone considers to be the root of your kDrive drive.+.PP+Normally you will leave this blank and rclone will use your private+directory.+You can also use one of the convenience shortcuts:+.IP \[bu] 2+\f[V]private\f[R]: your user private directory (default)+.IP \[bu] 2+\f[V]common\f[R]: the kDrive common directory, shared among users+.PP+Or the numeric \f[V]Folder ID\f[R] of any directory in your drive.+The root of the drive only contains the \f[V]Private\f[R] and+\f[V]Common documents\f[R] folders, so a numeric ID is needed to use any+other directory as root.+In order to find the \f[V]Folder ID\f[R], list the parent directory of+the folder you wish rclone to use with \f[V]rclone lsf\f[R]; the first+column of the output is the folder ID:+.PP+Example:+.IP+.nf+\f[C]+$ rclone lsf --dirs-only -Fip --csv remote:Documents+456,Sub folder/+789,Photos/+\f[R]+.fi+.PP+So if the folder you want rclone to use is \[dq]Photos/\[dq], then use+the returned id from \f[V]rclone lsf\f[R] command (ex.+\f[V]789\f[R]) as the \f[V]root_folder_id\f[R] variable value in the+config file.+.SS Standard options+.PP+Here are the Standard options specific to kdrive (Infomaniak kDrive).+.SS --kdrive-client-id+.PP+OAuth Client Id.+.PP+Leave blank normally.+.PP+Properties:+.IP \[bu] 2+Config: client_id+.IP \[bu] 2+Env Var: RCLONE_KDRIVE_CLIENT_ID+.IP \[bu] 2+Type: string+.IP \[bu] 2+Required: false+.SS --kdrive-drive-id+.PP+ID of the kDrive to use.+.PP+When showing a folder on kdrive, you can find the drive_id here:+https://ksuite.infomaniak.com/{account_id}/kdrive/app/drive/{drive_id}/files/...+.PP+Properties:+.IP \[bu] 2+Config: drive_id+.IP \[bu] 2+Env Var: RCLONE_KDRIVE_DRIVE_ID+.IP \[bu] 2+Type: string+.IP \[bu] 2+Required: false+.SS Advanced options+.PP+Here are the Advanced options specific to kdrive (Infomaniak kDrive).+.SS --kdrive-token+.PP+OAuth Access Token as a JSON blob.+.PP+Properties:+.IP \[bu] 2+Config: token+.IP \[bu] 2+Env Var: RCLONE_KDRIVE_TOKEN+.IP \[bu] 2+Type: string+.IP \[bu] 2+Required: false+.SS --kdrive-auth-url+.PP+Auth server URL.+.PP+Leave blank to use the provider defaults.+.PP+Properties:+.IP \[bu] 2+Config: auth_url+.IP \[bu] 2+Env Var: RCLONE_KDRIVE_AUTH_URL+.IP \[bu] 2+Type: string+.IP \[bu] 2+Required: false+.SS --kdrive-token-url+.PP+Token server url.+.PP+Leave blank to use the provider defaults.+.PP+Properties:+.IP \[bu] 2+Config: token_url+.IP \[bu] 2+Env Var: RCLONE_KDRIVE_TOKEN_URL+.IP \[bu] 2+Type: string+.IP \[bu] 2+Required: false+.SS --kdrive-encoding+.PP+The encoding for the backend.+.PP+See the encoding section in the+overview (https://rclone.org/overview/#encoding) for more info.+.PP+Properties:+.IP \[bu] 2+Config: encoding+.IP \[bu] 2+Env Var: RCLONE_KDRIVE_ENCODING+.IP \[bu] 2+Type: Encoding+.IP \[bu] 2+Default: Slash,Del,Ctl,LeftSpace,RightSpace,InvalidUtf8,Dot+.SS --kdrive-root-folder-id+.PP+The folder to use as root.+.PP+You may use either one of the convenience shortcuts [private or common]+or an explicit folder ID.+.IP \[bu] 2+private: your user private directory+.IP \[bu] 2+common: the kDrive common directory, shared among users+.PP+Properties:+.IP \[bu] 2+Config: root_folder_id+.IP \[bu] 2+Env Var: RCLONE_KDRIVE_ROOT_FOLDER_ID+.IP \[bu] 2+Type: string+.IP \[bu] 2+Required: false+.SS --kdrive-chunk-size+.PP+Files above this size are uploaded in chunks of this size.+.PP+The kDrive API accepts chunks between 1 MB and 1 GB, up to 10,000 chunks+per file and a total file size of 1 TiB.+The chunk size is raised automatically for very large files to stay+within the 10,000 chunks limit.+.PP+Properties:+.IP \[bu] 2+Config: chunk_size+.IP \[bu] 2+Env Var: RCLONE_KDRIVE_CHUNK_SIZE+.IP \[bu] 2+Type: SizeSuffix+.IP \[bu] 2+Default: 20Mi+.SS --kdrive-upload-concurrency+.PP+Concurrency for chunked uploads.+.PP+Properties:+.IP \[bu] 2+Config: upload_concurrency+.IP \[bu] 2+Env Var: RCLONE_KDRIVE_UPLOAD_CONCURRENCY+.IP \[bu] 2+Type: int+.IP \[bu] 2+Default: 4+.SS --kdrive-endpoint+.PP+The API endpoint to use.+.PP+Leave blank normally.+There is no reason to change the endpoint except for internal use or+beta-testing.+.PP+Properties:+.IP \[bu] 2+Config: endpoint+.IP \[bu] 2+Env Var: RCLONE_KDRIVE_ENDPOINT+.IP \[bu] 2+Type: string+.IP \[bu] 2+Required: false+.SS --kdrive-description+.PP+Description of the remote.+.PP+Properties:+.IP \[bu] 2+Config: description+.IP \[bu] 2+Env Var: RCLONE_KDRIVE_DESCRIPTION+.IP \[bu] 2+Type: string+.IP \[bu] 2+Required: false.SH Koofr.PPPaths are specified as \f[V]remote:path\f[R]@@ -71843,6 +75973,16 @@ Required: falseInvalid UTF-8 bytes will also bereplaced (https://rclone.org/overview/#invalid-utf8), as they can\[aq]tbe used in JSON strings.+.PP+Linkbox converts file extensions to lower case on upload, so a file+uploaded as \f[V]file.TXT\f[R] is stored as \f[V]file.txt\f[R].+.PP+Linkbox renames an uploaded file to \f[V]name(1)\f[R] if its name is the+start of the name of a file already in the same directory, for example+when uploading \f[V]apple\f[R] to a directory containing+\f[V]applebanana\f[R].+rclone then can\[aq]t find the file it uploaded and the upload fails+with \f[V]object not found\f[R]..SH Mail.ru Cloud.PPMail.ru Cloud (https://cloud.mail.ru/) is a cloud storage provided by a@@ -73575,6 +77715,26 @@ ignores the override..PPNotes: - Rclone ignores reserved \f[V]x-ms-*\f[R] keys (except\f[V]x-ms-tags\f[R]) for user metadata.+.SS Hierarchical namespace+.PP+Storage accounts with hierarchical namespace enabled (Azure Data Lake+Storage Gen2) have real directories.+On these accounts rclone supports empty directories, creating them with+\f[V]rclone mkdir\f[R] and removing them with \f[V]rclone rmdir\f[R],+whatever the \f[V]--azureblob-directory-markers\f[R] setting.+.PP+Note that deleting the last file in a directory leaves the directory+behind.+Use \f[V]rclone rmdirs\f[R] or \f[V]--delete-empty-src-dirs\f[R] to+remove empty directories.+.PP+Rclone reads whether the storage account has a hierarchical namespace+when it first uses the remote and saves it as the \f[V]hns\f[R] option+in the config file.+Set \f[V]--azureblob-hns\f[R] to override this.+.PP+Blob tags (\f[V]x-ms-tags\f[R] metadata) aren\[aq]t supported on these+accounts..SS Performance.PPWhen uploading large files, increasing the value of@@ -73911,14 +78071,14 @@ be explicitly specified using exactly one of the\f[V]msi_mi_res_id\f[R] parameters..PPIf none of \f[V]msi_object_id\f[R], \f[V]msi_client_id\f[R], or-\f[V]msi_mi_res_id\f[R] is set, this is is equivalent to using+\f[V]msi_mi_res_id\f[R] is set, this is equivalent to using\f[V]env_auth\f[R]..SS Federated Identity Credentials.PPIf these variables are set, rclone will authenticate with federatedidentity..IP \[bu] 2-\f[V]tenant\f[R]: tenant ID of of the storage+\f[V]tenant\f[R]: tenant ID of the storage.IP \[bu] 2\f[V]client_id\f[R]: client ID of the application the user willauthenticate to storage@@ -74521,6 +78681,57 @@ Env Var: RCLONE_AZUREBLOB_LIST_CHUNKType: int.IP \[bu] 2Default: 5000+.SS --azureblob-use-arrow-list+.PP+Use the Apache Arrow listing format.+.PP+If set, directory listings are fetched using the ListBlobs Apache Arrow+response format instead of XML.+Arrow responses are smaller and much cheaper to parse, making listings+of large containers several times faster.+Combine with \[dq]list_parallelism\[dq] for the biggest gains.+.PP+\[dq]Blob Listing with Apache Arrow\[dq] is in public preview at+Microsoft and is only supported on flat namespace accounts.+On accounts with a hierarchical namespace (ADLS Gen2), or where the+feature is otherwise unavailable, the server returns XML and the listing+transparently falls back to the normal XML path (logged at debug level).+.PP+Properties:+.IP \[bu] 2+Config: use_arrow_list+.IP \[bu] 2+Env Var: RCLONE_AZUREBLOB_USE_ARROW_LIST+.IP \[bu] 2+Type: bool+.IP \[bu] 2+Default: false+.SS --azureblob-list-parallelism+.PP+Number of parallel shards to list a directory with.+.PP+If set greater than 1, the blob name keyspace of each directory is split+into ranges which are listed concurrently using the Arrow+startFrom/endBefore range parameters.+This can dramatically speed up listing containers with millions of+objects, for both recursive (ListR) and single directory listings.+Speed keeps improving up to a parallelism of around 30.+.PP+This has no effect unless \[dq]use_arrow_list\[dq] is also set, as Arrow+is the only listing path that supports server-side name ranges.+If the account does not support range listing (e.g.+it has a hierarchical namespace) the listing falls back to sequential.+The default of 0 (or 1) lists sequentially.+.PP+Properties:+.IP \[bu] 2+Config: list_parallelism+.IP \[bu] 2+Env Var: RCLONE_AZUREBLOB_LIST_PARALLELISM+.IP \[bu] 2+Type: int+.IP \[bu] 2+Default: 0.SS --azureblob-access-tier.PPAccess tier of blob: hot, cool, cold or archive.@@ -74662,6 +78873,9 @@ creates an empty object ending with \[dq]/\[dq], to persist the folder.This object also has the metadata \[dq]hdi_isfolder = true\[dq] toconform to the Microsoft standard..PP+This is not needed on storage accounts with a hierarchical namespace+(see the hns option) as these always support empty directories.+.PPProperties:.IP \[bu] 2Config: directory_markers@@ -74671,6 +78885,25 @@ Env Var: RCLONE_AZUREBLOB_DIRECTORY_MARKERSType: bool.IP \[bu] 2Default: false+.SS --azureblob-hns+.PP+Set if the storage account has a hierarchical namespace.+.PP+Storage accounts with hierarchical namespace enabled (Azure Data Lake+Storage Gen2) have real directories which can be empty.+.PP+If this is unset then rclone will read it from the storage account and+save it in the config file.+.PP+Properties:+.IP \[bu] 2+Config: hns+.IP \[bu] 2+Env Var: RCLONE_AZUREBLOB_HNS+.IP \[bu] 2+Type: Tristate+.IP \[bu] 2+Default: unset.SS --azureblob-no-check-container.PPIf set, don\[aq]t attempt to check the container exists or create it.@@ -75394,14 +79627,14 @@ be explicitly specified using exactly one of the\f[V]msi_mi_res_id\f[R] parameters..PPIf none of \f[V]msi_object_id\f[R], \f[V]msi_client_id\f[R], or-\f[V]msi_mi_res_id\f[R] is set, this is is equivalent to using+\f[V]msi_mi_res_id\f[R] is set, this is equivalent to using\f[V]env_auth\f[R]..SS Federated Identity Credentials.PPIf these variables are set, rclone will authenticate with federatedidentity..IP \[bu] 2-\f[V]tenant\f[R]: tenant ID of of the storage+\f[V]tenant\f[R]: tenant ID of the storage.IP \[bu] 2\f[V]client_id\f[R]: client ID of the application the user willauthenticate to storage@@ -75848,7 +80081,7 @@ Required: falseUpload chunk size..PPNote that this is stored in memory and there may be up to-\[dq]--transfers\[dq] * \[dq]--azurefile-upload-concurrency\[dq] chunks+\[dq]--transfers\[dq] * \[dq]--azurefiles-upload-concurrency\[dq] chunksstored at once in memory..PPProperties:@@ -75872,7 +80105,7 @@ and these uploads do not fully utilize your bandwidth, then increasingthis may help to speed up the transfers..PPNote that chunks are stored in memory and there may be up to-\[dq]--transfers\[dq] * \[dq]--azurefile-upload-concurrency\[dq] chunks+\[dq]--transfers\[dq] * \[dq]--azurefiles-upload-concurrency\[dq] chunksstored at once in memory..PPProperties:@@ -75970,6 +80203,9 @@ Paths are specified as \f[V]remote:path\f[R].PPPaths may be as deep as required, e.g.\f[V]remote:directory/subdirectory\f[R].+.PP+\f[B]Note\f[R]: OneDrive for Business and SharePoint users need rclone+v1.76 or later from April 1, 2027 - see Pre-authenticated URLs..SS Configuration.PPThe initial setup for OneDrive involves getting a token from Microsoft@@ -76250,7 +80486,7 @@ If you do not have admin access to your organization\[aq]s OneDrive forBusiness, you can still connect by manually providing the SharePointtenant URL and drive ID.This works by overriding the base API URL from the standard Microsoft-Graph endpoint to the SharePoint v2.0 endpoint.+Graph endpoint to the SharePoint API endpoint..SS Steps to manually obtain credentials.IP "1." 3Open your browser and navigate to@@ -76269,6 +80505,16 @@ Extract the following information from the response:\[dq].driveAccessToken\[dq]: \[dq]access_token={access_token}\[dq]\f[R].fi+.PP+In some guest access scenarios SharePoint may also expose a matching+v2.1 URL and token pair:+.IP+.nf+\f[C]+\[dq].driveUrlV21\[dq]: \[dq]{tenant_url}/v2.1/drives/{drive_id}\[dq],+\[dq].driveAccessTokenV21\[dq]: \[dq]access_token={access_token}\[dq]+\f[R]+.fi.RE.SS Rclone configuration.PP@@ -76284,6 +80530,17 @@ drive_type = business\f[R].fi.PP+If you use a v2.1 token, set the matching SharePoint API version:+.IP+.nf+\f[C]+tenant_api_version = v2.1+\f[R]+.fi+.PP+If \f[V]tenant_api_version\f[R] is omitted, rclone defaults to+\f[V]v2.0\f[R].+.PPSince the exact expiry time cannot be determined from web traffic, setthe expiry to a future date.Note that the token will eventually expire and you will need to repeat@@ -76330,11 +80587,11 @@ This can be useful with \f[V]rclone mount\f[R] and rclone rc vfs/refreshrecursive=true (https://rclone.org/rc/#vfs-refresh)) to very quicklyfill the mount with information about all the files..PP-The API used for the recursive listing (\f[V]ListR\f[R]) only supports-listing from the root of the drive.-This will become increasingly inefficient the further away you get from-the root as rclone will have to discard files outside of the directory-you are using.+Rclone asks for the delta listing of the directory being listed.+Some drives only support delta listings from the root of the drive, in+which case rclone lists the whole drive and discards the files outside+of the directory you are using, which is increasingly inefficient the+further away from the root you get..PPSome commands (like \f[V]rclone lsf -R\f[R]) will use \f[V]ListR\f[R] bydefault - you can turn this off with \f[V]--disable ListR\f[R] if you@@ -76663,7 +80920,7 @@ Default: off.PPThe tenant URL for non-admin OneDrive access..PP-Set this to your SharePoint tenant URL to use the SharePoint v2.0 API+Set this to your SharePoint tenant URL to use the SharePoint APIendpoint instead of the standard Microsoft Graph API.This allows accessing business OneDrive without admin consent..PP@@ -76683,6 +80940,25 @@ Env Var: RCLONE_ONEDRIVE_TENANT_URLType: string.IP \[bu] 2Required: false+.SS --onedrive-tenant-api-version+.PP+The SharePoint API version to use with tenant_url.+.PP+Set this to the SharePoint API version matching the browser-extracted+access token.+For example, use v2.1 with a driveAccessTokenV21 token.+.PP+This only applies when tenant_url is set.+.PP+Properties:+.IP \[bu] 2+Config: tenant_api_version+.IP \[bu] 2+Env Var: RCLONE_ONEDRIVE_TENANT_API_VERSION+.IP \[bu] 2+Type: string+.IP \[bu] 2+Default: \[dq]v2.0\[dq].SS --onedrive-chunk-size.PPChunk size to upload files with - must be multiple of 320k (327,680@@ -76855,17 +81131,14 @@ Default: 1000.PPRemove all versions on modifying operations..PP-Onedrive for business creates versions when rclone uploads new files-overwriting an existing one and when it sets the modification time.+Onedrive creates versions when rclone uploads new files overwriting an+existing one and when it sets the modification time..PPThese versions take up space out of the quota..PPThis flag checks for versions after file upload and setting modificationtime and removes all but the last version..PP-\f[B]NB\f[R] Onedrive personal can\[aq]t currently delete versions so-don\[aq]t use this flag there.-.PPProperties:.IP \[bu] 2Config: no_versions@@ -76883,8 +81156,8 @@ Normally files will get sent to the recycle bin on deletion.Setting this flag causes them to be permanently deleted.Use with care..PP-OneDrive personal accounts do not support the permanentDelete API, it-only applies to OneDrive for Business and SharePoint document libraries.+This works with OneDrive for Business, SharePoint document libraries,+and OneDrive personal accounts, including free accounts..PPProperties:.IP \[bu] 2@@ -76964,14 +81237,19 @@ Creates a read-write link to the item..RS 2.IP \[bu] 2Creates an embeddable link to the item.+.IP \[bu] 2+Only available in OneDrive personal..RE.RE.SS --onedrive-link-password.PPSet the password for links created by the link command..PP-At the time of writing this only works with OneDrive personal paid-accounts.+At the time of writing this works with OneDrive for Business and+OneDrive personal paid accounts.+.PP+OneDrive personal free accounts can\[aq]t set a password or an expiry+time (with --expire) on links..PPProperties:.IP \[bu] 2@@ -77110,24 +81388,13 @@ rclone rc vfs/refresh recursive=true\f[R].fi.PP-\f[B]However\f[R] the delta listing API \f[B]only\f[R] works at the root-of the drive.-If you use it not at the root then it recurses from the root and-discards all the data that is not under the directory you asked for.+Rclone asks for the delta listing of the directory being listed.+If the drive only supports delta listings at the root of the drive (as+some older OneDrive for Business and SharePoint drives do) then rclone+lists from the root and discards all the data that is not under the+directory you asked for.So it will be correct but may not be very efficient..PP-This is why this flag is not set as the default.-.PP-As a rule of thumb if nearly all of your data is under rclone\[aq]s root-directory (the \f[V]root/directory\f[R] in-\f[V]onedrive:root/directory\f[R]) then using this flag will be a big-performance win.-If your data is mostly not under the root then using this flag will be a-big performance loss.-.PP-It is recommended if you are mounting your onedrive at the root (or near-the root when using crypt) and using rclone \f[V]rc vfs/refresh\f[R].-.PPProperties:.IP \[bu] 2Config: delta@@ -77396,8 +81663,7 @@ _T{btimeT}@T{-Time of file birth (creation) with S accuracy (mS for OneDrive-Personal).+Time of file birth (creation) with S accuracy.T}@T{RFC 3339T}@T{@@ -77498,7 +81764,7 @@ T}T{mtimeT}@T{-Time of last modification with S accuracy (mS for OneDrive Personal).+Time of last modification with S accuracy.T}@T{RFC 3339T}@T{@@ -77570,8 +81836,7 @@ T}T{shared-timeT}@T{-Time when the item was shared, with S accuracy (mS for OneDrive-Personal).+Time when the item was shared, with S accuracy.T}@T{RFC 3339T}@T{@@ -77582,7 +81847,7 @@ T}T{utimeT}@T{-Time of upload with S accuracy (mS for OneDrive Personal).+Time of upload with S accuracy.T}@T{RFC 3339T}@T{@@ -77692,11 +81957,6 @@ all old versions.Or you can set the \f[V]no_versions\f[R] parameter to \f[V]true\f[R] andrclone will remove versions after operations which create new versions.This takes extra transactions so only enable it if you need it.-.PP-\f[B]Note\f[R] At the time of writing Onedrive Personal creates versions-(but not for setting the modification time) but the API for removing-them returns \[dq]API not found\[dq] so cleanup and-\f[V]no_versions\f[R] should not be used on Onedrive Personal..SS Disabling versioning.PPStarting October 2018, users will no longer be able to disable@@ -77770,9 +82030,38 @@ rclone cleanup --interactive remote:path/subdir # interactively remove all old vrclone cleanup remote:path/subdir # unconditionally remove all old version for path/subdir\f[R].fi-.PP-\f[B]NB\f[R] Onedrive personal can\[aq]t currently delete versions.SS Troubleshooting+.SS Pre-authenticated URLs+.PP+Microsoft is retiring pre-authenticated (tempauth)+URLs (https://devblogs.microsoft.com/microsoft365dev/pre-authenticated-tempauth-urls-are-retiring-move-to-microsoft-entra-id-tokens-before-april-1-2027/)+in OneDrive for Business and SharePoint from April 1, 2027.+.PP+\f[B]Upgrade to rclone v1.76 or later before April 1, 2027.\f[R] Earlier+versions rely on pre-authenticated URLs, so after that date their+downloads, chunked uploads and server-side copies will fail with+OneDrive for Business and SharePoint.+Rclone v1.76 and later keep working with the existing configuration,+apart from the client credentials flow with a client secret (see below).+.PP+Rclone uses these URLs when downloading files, when uploading files+bigger than \f[V]--onedrive-upload-cutoff\f[R] in chunks and when doing+server-side copies.+When SharePoint returns them without pre-authentication rclone+authenticates them with a SharePoint token which it gets using the+existing login, so no reconfiguration is needed.+OneDrive Personal is not affected.+.PP+If rclone can\[aq]t get a SharePoint token, or SharePoint refuses it,+you will see the error+\f[V]SharePoint refused to authenticate a request\f[R].+In this case rclone uploads files smaller than 250 MiB in a single+request instead and copies files by downloading and uploading them, but+bigger files can\[aq]t be uploaded and downloads may fail.+.PP+SharePoint does not accept tokens from the client credentials flow which+were got with a client secret, so this affects+\f[V]client_credentials\f[R] users..SS Excessive throttling or blocked on SharePoint.PPIf you experience excessive throttling or is being blocked on SharePoint@@ -77850,7 +82139,7 @@ You can\[aq]t do much about it, maybe write an email to your admins..PPHowever, there are other ways to interact with your OneDrive account.Have a look at the WebDAV backend:-<https://rclone.org/webdav/#sharepoint>+<https://rclone.org/webdav/#sharepoint-online>.SS invalid_grant (AADSTS50076).IP.nf@@ -77895,6 +82184,32 @@ The shortcut will appear in \f[V]My files\f[R], you can access it withrclone, it behaves like a normal folder/file.[IMAGE: in_my_files (https://i.imgur.com/0S8H3li.png)][IMAGE: rclone_mount (https://i.imgur.com/2Iq66sW.png)]+.PP+Listing such a shortcut may still fail with+.IP+.nf+\f[C]+invalidRequest: The provided drive id appears to be malformed, or does not represent a valid drive.+\f[R]+.fi+.PP+as the drive it points to can\[aq]t always be read with the drive ID the+API supplies.+\f[V]Personal Vault\f[R] is a shortcut of the same kind and fails to+list in the same way, though it reports+\f[V]invalidResourceId: ObjectHandle is Invalid\f[R].+.PP+A normal listing reports the error for that folder and carries on, so+the rest of the drive is still listed, and excluding the folder (for+example with \f[V]--exclude \[dq]/Personal Vault/**\[dq]\f[R]) avoids+the error.+.PP+A recursive listing with \f[V]--fast-list\f[R] stops at the first error,+so listing the root of a drive which has one of these shortcuts in it+will fail.+Filters don\[aq]t help here as the folder is listed by the backend+before the filters are applied - use \f[V]--disable ListR\f[R] (or+don\[aq]t use \f[V]--fast-list\f[R]) on such a drive..SS Live Photos uploaded from iOS (small video clips in .heic files).PPThe iOS OneDrive app introduced upload and@@ -79077,7 +83392,7 @@ Env Var: RCLONE_OOS_ENCODING.IP \[bu] 2Type: Encoding.IP \[bu] 2-Default: Slash,InvalidUtf8,Dot+Default: Slash,CrLf,InvalidUtf8,Dot.SS --oos-leave-parts-on-error.PPIf true avoid calling abort upload on a failure, leaving all@@ -80351,7 +84666,7 @@ Please follow the Get started (https://sia.tech/get-started) guide andinstall one..PPrclone interacts with Sia network by talking to the Sia daemon via HTTP-API (https://sia.tech/docs/) which is usually available on port+API (https://docs.sia.tech/) which is usually available on port\f[I]9980\f[R].By default you will run the daemon locally on the same computer soit\[aq]s safe to leave the API password blank (the API URL will be@@ -80614,6 +84929,9 @@ Blomp Cloud Storage (https://www.blomp.com/cloud-storage/).IP \[bu] 2IBM Bluemix Cloud ObjectStorageSwift (https://console.bluemix.net/docs/infrastructure/objectstorage-swift/index.html)+.IP \[bu] 2+Infomaniak Swiss+Backup (https://www.infomaniak.com/en/swiss-backup/cloud-storage).PPPaths are specified as \f[V]remote:container\f[R] (or \f[V]remote:\f[R]for the \f[V]lsd\f[R] command.)@@ -80763,8 +85081,8 @@ rclone sync --interactive /home/local/directory remote:container.fi.SS Configuration from an OpenStack credentials file.PP-An OpenStack credentials file typically looks something something like-this (without the comments)+An OpenStack credentials file typically looks something like this+(without the comments).IP.nf\f[C]@@ -81017,6 +85335,30 @@ OVH.IP \[bu] 2Blomp Cloud Storage.RE+.IP \[bu] 2+\[dq]https://swiss-backup01.infomaniak.com/identity/v3\[dq]+.RS 2+.IP \[bu] 2+Infomaniak Swiss Backup Location 01+.RE+.IP \[bu] 2+\[dq]https://swiss-backup02.infomaniak.com/identity/v3\[dq]+.RS 2+.IP \[bu] 2+Infomaniak Swiss Backup Location 02+.RE+.IP \[bu] 2+\[dq]https://swiss-backup03.infomaniak.com/identity/v3\[dq]+.RS 2+.IP \[bu] 2+Infomaniak Swiss Backup Location 03+.RE+.IP \[bu] 2+\[dq]https://swiss-backup04.infomaniak.com/identity/v3\[dq]+.RS 2+.IP \[bu] 2+Infomaniak Swiss Backup Location 04+.RE.RE.SS --swift-user-id.PP@@ -81480,7 +85822,7 @@ Request\[dq] error rather than a more sensible error when theauthentication fails for Swift..PPSo this most likely means your username / password is wrong.-You can investigate further with the \f[V]--dump-bodies\f[R] flag.+You can investigate further with the \f[V]--dump bodies\f[R] flag..PPThis may also be caused by specifying the region when you shouldn\[aq]thave (e.g.@@ -82169,6 +86511,8 @@ Default: falseFiles bigger than this will be cached on disk to calculate hash ifrequired..PP+Files of unknown size are always cached on disk.+.PPProperties:.IP \[bu] 2Config: hash_memory_limit@@ -84266,6 +88610,14 @@ T}Invalid UTF-8 bytes will also bereplaced (https://rclone.org/overview/#invalid-utf8), as they can\[aq]tbe used in JSON strings.+.PP+Seafile stores file and directory names in Unicode NFC form whatever+form they are uploaded in, so rclone normalizes names to NFC before+sending them.+Listings therefore always show NFC names, even for files created with+decomposed (NFD) names as macOS produces them.+Library names are not normalized as Seafile keeps them exactly as+entered..SS Seafile and rclone link.PPRclone supports generating share links for non-encrypted libraries only.@@ -84977,7 +89329,7 @@ The shell type auto-detection logic, described above, means that bydefault rclone will try to run a shell command the first time a new sftpremote is accessed.If you configure a sftp remote without a config file, e.g.-an on the fly (https://rclone.org/docs/#backend-path-to-dir%5D) remote,+an on the fly (https://rclone.org/docs/#backend-path-to-dir) remote,rclone will have nowhere to store the result, and it will re-run thecommand on every access.To avoid this you should explicitly set the \f[V]shell_type\f[R] option@@ -85072,6 +89424,48 @@ mod_sftp).If you are using one of these servers, you can set the option\f[V]set_modtime = false\f[R] in your RClone backend configuration todisable this behaviour.+.SS Restricted filename characters+.PP+SFTP servers on Unix-like systems can store any character in a file name+except NUL and \f[V]/\f[R], so by default only these are replaced:+.PP+.TS+tab(@);+l c c.+T{+Character+T}@T{+Value+T}@T{+Replacement+T}+_+T{+NUL+T}@T{+0x00+T}@T{+␀+T}+T{+/+T}@T{+0x2F+T}@T{+/+T}+.TE+.PP+If the server runs on Windows, which can\[aq]t store many more+characters, add the Windows restricted characters with+\f[V]--sftp-encoding\f[R], as in the FTP+example (https://rclone.org/overview/#encoding-example-ftp):+.IP+.nf+\f[C]+--sftp-encoding \[dq]Slash,LtGt,DoubleQuote,Colon,Question,Asterisk,Pipe,BackSlash,Ctl,RightSpace,RightPeriod,InvalidUtf8,Dot\[dq]+\f[R]+.fi.SS About command.PPThe \f[V]about\f[R] command returns the total space, free space, and@@ -85573,7 +89967,7 @@ Env Var: RCLONE_SFTP_ENCODING.IP \[bu] 2Type: Encoding.IP \[bu] 2-Default: Slash,Del,Ctl,Dot+Default: Slash,Dot.SS --sftp-set-modtime.PPSet the modified time on the remote if set.@@ -85879,6 +90273,35 @@ Env Var: RCLONE_SFTP_DISABLE_CONCURRENT_WRITESType: bool.IP \[bu] 2Default: false+.SS --sftp-multithread-upload+.PP+Set this to use multi-thread (multi-connection) uploads.+.PP+Normally rclone uploads a single large file over one SFTP connection.+With this set rclone uploads large files over several SFTP connections+at once (one per chunk), which greatly improves throughput on high+latency links, the same way multi-thread downloads already work.+Each connection writes a different, non-overlapping byte range of the+file, so this requires a server that allows several handles to the same+file with writes at arbitrary offsets, which OpenSSH does.+.PP+This is off by default because many SFTP servers only accept sequential+writes (for example some object storage backed SFTP gateways) and would+fail large uploads with \[dq]truncate failed\[dq] or \[dq]invalid+offset\[dq] style errors.+.PP+This is ignored (uploads stay single-threaded) when+--sftp-disable-concurrent-writes or --sftp-connections is in use.+.PP+Properties:+.IP \[bu] 2+Config: multithread_upload+.IP \[bu] 2+Env Var: RCLONE_SFTP_MULTITHREAD_UPLOAD+.IP \[bu] 2+Type: bool+.IP \[bu] 2+Default: false.SS --sftp-idle-timeout.PPMax time before closing idle connections.@@ -86225,14 +90648,14 @@ This can be re-enabled on a per-connection basis by setting the\f[V]use_insecure_cipher\f[R] setting in the configuration file to\f[V]true\f[R].Further details on the insecurity of this cipher can be found in this-paper (http://www.isg.rhul.ac.uk/~kp/SandPfinal.pdf).+paper (https://conferences.computer.org/sp/pdfs/sp/2009/oakland2009-01.pdf)..PPSFTP isn\[aq]t supported under plan9 until thisissue (https://github.com/pkg/sftp/issues/156) is fixed..PPNote that since SFTP isn\[aq]t HTTP based the following flags don\[aq]t-work with it: \f[V]--dump-headers\f[R], \f[V]--dump-bodies\f[R],-\f[V]--dump-auth\f[R].+work with it: \f[V]--dump headers\f[R], \f[V]--dump bodies\f[R],+\f[V]--dump auth\f[R]..PPNote that \f[V]--timeout\f[R] and \f[V]--contimeout\f[R] are bothsupported.@@ -86682,6 +91105,8 @@ Required: true.PPSMB username..PP+Leave blank to use the current user name.+.PPProperties:.IP \[bu] 2Config: user@@ -86690,7 +91115,7 @@ Env Var: RCLONE_SMB_USER.IP \[bu] 2Type: string.IP \[bu] 2-Default: \[dq]$USER\[dq]+Required: false.SS --smb-port.PPSMB port number.@@ -86780,6 +91205,26 @@ Default: false.SS Advanced options.PPHere are the Advanced options specific to smb (SMB / CIFS).+.SS --smb-workstation+.PP+Workstation name sent in NTLM authentication.+.PP+The NetBIOS name sent as the client workstation in the NTLM AUTHENTICATE+message.+Some servers (e.g.+domain-joined Windows servers with per-account workstation restrictions+or NTLM allow-lists) require this to match a specific name.+Leave blank to send no workstation name (the default).+.PP+Properties:+.IP \[bu] 2+Config: workstation+.IP \[bu] 2+Env Var: RCLONE_SMB_WORKSTATION+.IP \[bu] 2+Type: string+.IP \[bu] 2+Required: false.SS --smb-idle-timeout.PPMax time before closing idle connections.@@ -87045,8 +91490,7 @@ To make a new Storj configuration you need one of the following:.IP \[bu] 2Access Grant that someone else shared with you..IP \[bu] 2-API-Key (https://documentation.storj.io/getting-started/uploading-your-first-object/create-an-api-key)+API Key (https://storj.dev/learn/concepts/access/access-grants/api-key)of a Storj project you are a member of..PPHere is an example of how to make a remote called \f[V]remote\f[R].@@ -88818,6 +93262,9 @@ hashes.Depending on the exact version of ownCloud or Nextcloud hashes mayappear on all objects, or only on objects which had a hash uploaded withthem.+With Nextcloud, rclone asks the server to calculate the SHA1 of uploads+which had no hash to send, such as streamed uploads, and after setting+the modification time, which discards the stored hash..SS Standard options.PPHere are the Standard options specific to webdav (WebDAV).@@ -88999,6 +93446,13 @@ For example, to set a Cookie use \[aq]Cookie,name=value\[aq], orYou can set multiple headers, e.g.\[aq]\[dq]Cookie\[dq],\[dq]name=value\[dq],\[dq]Authorization\[dq],\[dq]xxx\[dq]\[aq]..PP+The headers are only sent to the host in the configured URL.+If the server redirects to another host (including a subdomain or a+different port) the headers are not sent to it, or to any further hop in+that redirect chain.+The exception is the directory listing and file download requests when+auth_redirect is set, as that keeps all credentials across redirects.+.PPProperties:.IP \[bu] 2Config: headers@@ -89150,8 +93604,8 @@ ownCloud supports modified times using the \f[V]X-OC-Mtime\f[R] header.This is configured in an identical way to ownCloud.Note that Nextcloud initially did not support streaming of files(\f[V]rcat\f[R]) whereas ownCloud did, but-this (https://github.com/nextcloud/nextcloud-snap/issues/365) seems to-be fixed as of 2020-11-27 (tested with rclone v1.53.1 and Nextcloud+this (https://github.com/nextcloud-snap/nextcloud-snap/issues/365) seems+to be fixed as of 2020-11-27 (tested with rclone v1.53.1 and NextcloudServer v19)..SS ownCloud Infinite Scale.PP@@ -89261,7 +93715,9 @@ property to compare your documents:.PPUse this option if you are hosting remotes over WebDAV provided byrclone.-Read rclone serve webdav for more details.+Read rclone serve+webdav (https://rclone.org/commands/rclone_serve_webdav/) for more+details..PPrclone serve supports modified times using the \f[V]X-OC-Mtime\f[R]header.@@ -89489,6 +93945,16 @@ set (https://rclone.org/overview/#restricted-characters) are replaced.Invalid UTF-8 bytes will also bereplaced (https://rclone.org/overview/#invalid-utf8), as they can\[aq]tbe used in JSON strings.+.SS Application folder+.PP+If you registered your own OAuth application with Yandex and only+granted it the \[dq]Application Folder\[dq] permission+(\f[V]cloud_api:disk.app_folder\f[R]) instead of full disk access, then+you must set the \f[V]app_folder\f[R] option to \f[V]true\f[R].+Yandex Disk requires paths to be addressed with an \f[V]app:/\f[R]+prefix instead of the usual \f[V]disk:/\f[R] prefix when using this+restricted scope, and rclone will return a 403 Forbidden error on every+request if this option isn\[aq]t set correctly for the token in use..SS Standard options.PPHere are the Standard options specific to yandex (Yandex Disk).@@ -89653,6 +94119,28 @@ Env Var: RCLONE_YANDEX_UPLOAD_WAITType: Duration.IP \[bu] 2Default: 0s+.SS --yandex-app-folder+.PP+Use the application folder as the root.+.PP+If you registered your own OAuth application with Yandex with only the+\[dq]Application Folder\[dq] permission (cloud_api:disk.app_folder)+instead of full disk access, then rclone needs to address paths with the+\[dq]app:/\[dq] prefix instead of the usual \[dq]disk:/\[dq] prefix,+otherwise all requests fail with a 403 Forbidden error.+.PP+Set this to true if your OAuth token only has access to the application+folder.+.PP+Properties:+.IP \[bu] 2+Config: app_folder+.IP \[bu] 2+Env Var: RCLONE_YANDEX_APP_FOLDER+.IP \[bu] 2+Type: bool+.IP \[bu] 2+Default: false.SS --yandex-description.PPDescription of the remote.@@ -90741,6 +95229,9 @@ $ rclone -L ls /tmp/a6 b/one\f[R].fi+.PP+Symlinks which point to a parent directory would loop forever, so rclone+reports them as errors and skips them..SS --local-links, --links, -l.PPNormally rclone will ignore symlinks or junction points (which behave@@ -90809,6 +95300,9 @@ $ tree /tmp/b\f[R].fi.PP+A \f[V].rclonelink\f[R] file whose contents are too long to be a path+(more than 128 KiB) is refused rather than turned into a symlink.+.PPHowever, if copied back without \[aq]-l\[aq].IP.nf@@ -91174,6 +95668,14 @@ However, for use cases where data redundancy is preferable,--local-no-clone can be used to disable cloning and force \[dq]deep\[dq]copies..PP+Cloning also preserves macOS-specific metadata such as extended+attributes, which a normal copy does not, so this metadata is lost when+cloning is disabled.+.PP+If the destination file already exists, cloning replaces it with a new+file, even with --inplace, so hard links to it are broken.+Use --local-no-clone to update existing files in place.+.PPCurrently, cloning is only supported when using APFS on macOS (supportfor other platforms may be added in the future.).PP@@ -91198,362 +95700,2204 @@ Use this flag to disable preallocation..PPProperties:.IP \[bu] 2-Config: no_preallocate+Config: no_preallocate+.IP \[bu] 2+Env Var: RCLONE_LOCAL_NO_PREALLOCATE+.IP \[bu] 2+Type: bool+.IP \[bu] 2+Default: false+.SS --local-no-sparse+.PP+Disable sparse files for multi-thread downloads.+.PP+On Windows platforms rclone will make sparse files when doing+multi-thread downloads.+This avoids long pauses on large files where the OS zeros the file.+However sparse files may be undesirable as they cause disk fragmentation+and can be slow to work with.+.PP+Properties:+.IP \[bu] 2+Config: no_sparse+.IP \[bu] 2+Env Var: RCLONE_LOCAL_NO_SPARSE+.IP \[bu] 2+Type: bool+.IP \[bu] 2+Default: false+.SS --local-no-set-modtime+.PP+Disable setting modtime.+.PP+Normally rclone updates modification time of files after they are done+uploading.+This can cause permissions issues on Linux platforms when the user+rclone is running as does not own the file uploaded, such as when+copying to a CIFS mount owned by another user.+If this option is enabled, rclone will no longer update the modtime+after copying a file.+.PP+Properties:+.IP \[bu] 2+Config: no_set_modtime+.IP \[bu] 2+Env Var: RCLONE_LOCAL_NO_SET_MODTIME+.IP \[bu] 2+Type: bool+.IP \[bu] 2+Default: false+.SS --local-metadata-restore-special-bits+.PP+Restore the setuid, setgid and sticky bits from metadata.+.PP+When restoring metadata with --metadata rclone applies the+\[dq]mode\[dq] from the source.+By default rclone applies only the permission bits and strips the+setuid, setgid and sticky bits.+.PP+The \[dq]mode\[dq] comes from the source remote which may not be+trusted.+Restoring a setuid or setgid bit onto freshly written, source-controlled+content can plant a setuid binary, which is dangerous in particular when+restoring from an untrusted source while running as root.+For this reason these bits are not restored by default.+.PP+If you trust the source and want the setuid, setgid and sticky bits+restored - for example when restoring a system backup made by rclone -+set this flag.+.PP+Properties:+.IP \[bu] 2+Config: metadata_restore_special_bits+.IP \[bu] 2+Env Var: RCLONE_LOCAL_METADATA_RESTORE_SPECIAL_BITS+.IP \[bu] 2+Type: bool+.IP \[bu] 2+Default: false+.SS --local-fatal-if-no-space+.PP+Make out-of-space errors fatal during transfers.+.PP+When enabled, an out of space error while writing, creating a directory,+or creating a file returns a fatal error so that rclone aborts rather+than retrying the operation.+Useful for backup scripts that should halt loudly on a full disk rather+than spin retrying.+.PP+Properties:+.IP \[bu] 2+Config: fatal_if_no_space+.IP \[bu] 2+Env Var: RCLONE_LOCAL_FATAL_IF_NO_SPACE+.IP \[bu] 2+Type: bool+.IP \[bu] 2+Default: false+.SS --local-time-type+.PP+Set what kind of time is returned.+.PP+Normally rclone does all operations on the mtime or Modification time.+.PP+If you set this flag then rclone will return the Modified time as+whatever you set here.+So if you use \[dq]rclone lsl --local-time-type ctime\[dq] then you will+see ctimes in the listing.+.PP+If the OS doesn\[aq]t support returning the time_type specified then+rclone will silently replace it with the modification time which all+OSes support.+.IP \[bu] 2+mtime is supported by all OSes+.IP \[bu] 2+atime is supported on all OSes except: plan9, js+.IP \[bu] 2+btime is only supported on: Windows, macOS, freebsd, netbsd+.IP \[bu] 2+ctime is supported on all Oses except: Windows, plan9, js+.PP+Note that setting the time will still set the modified time so this is+only useful for reading.+.PP+Properties:+.IP \[bu] 2+Config: time_type+.IP \[bu] 2+Env Var: RCLONE_LOCAL_TIME_TYPE+.IP \[bu] 2+Type: mtime|atime|btime|ctime+.IP \[bu] 2+Default: mtime+.IP \[bu] 2+Examples:+.RS 2+.IP \[bu] 2+\[dq]mtime\[dq]+.RS 2+.IP \[bu] 2+The last modification time.+.RE+.IP \[bu] 2+\[dq]atime\[dq]+.RS 2+.IP \[bu] 2+The last access time.+.RE+.IP \[bu] 2+\[dq]btime\[dq]+.RS 2+.IP \[bu] 2+The creation time.+.RE+.IP \[bu] 2+\[dq]ctime\[dq]+.RS 2+.IP \[bu] 2+The last status change time.+.RE+.RE+.SS --local-hashes+.PP+Comma separated list of supported checksum types.+.PP+Properties:+.IP \[bu] 2+Config: hashes+.IP \[bu] 2+Env Var: RCLONE_LOCAL_HASHES+.IP \[bu] 2+Type: CommaSepList+.IP \[bu] 2+Default:+.SS --local-encoding+.PP+The encoding for the backend.+.PP+See the encoding section in the+overview (https://rclone.org/overview/#encoding) for more info.+.PP+Properties:+.IP \[bu] 2+Config: encoding+.IP \[bu] 2+Env Var: RCLONE_LOCAL_ENCODING+.IP \[bu] 2+Type: Encoding+.IP \[bu] 2+Default: Slash,Dot+.SS --local-description+.PP+Description of the remote.+.PP+Properties:+.IP \[bu] 2+Config: description+.IP \[bu] 2+Env Var: RCLONE_LOCAL_DESCRIPTION+.IP \[bu] 2+Type: string+.IP \[bu] 2+Required: false+.SS Metadata+.PP+Depending on which OS is in use the local backend may return only some+of the system metadata.+Setting system metadata is supported on all OSes but setting user+metadata is only supported on linux, freebsd, netbsd, macOS and Solaris.+It is \f[B]not\f[R] supported on Windows yet (see+pkg/attrs#47 (https://github.com/pkg/xattr/issues/47)).+.PP+User metadata is stored as extended attributes (which may not be+supported by all file systems) under the \[dq]user.*\[dq] prefix.+.PP+Metadata is supported on files and directories.+.PP+When restoring metadata with \f[V]--metadata\f[R] rclone applies the+\[dq]mode\[dq], \[dq]uid\[dq] and \[dq]gid\[dq] from the source.+These come from the source remote which may not be trusted, so restoring+metadata as root from an untrusted source can change file ownership and+is not recommended.+A user other than root keeps its own uid and sets the gid only to a+group it is a member of, as rsync does.+The setuid, setgid and sticky bits are not restored by default - see the+\f[V]--local-metadata-restore-special-bits\f[R] flag.+.PP+Here are the possible system metadata items for the local backend.+.PP+.TS+tab(@);+lw(11.1n) lw(11.1n) lw(11.1n) lw(16.6n) lw(20.3n).+T{+Name+T}@T{+Help+T}@T{+Type+T}@T{+Example+T}@T{+Read Only+T}+_+T{+atime+T}@T{+Time of last access+T}@T{+RFC 3339+T}@T{+2006-01-02T15:04:05.999999999Z07:00+T}@T{+N+T}+T{+btime+T}@T{+Time of file birth (creation)+T}@T{+RFC 3339+T}@T{+2006-01-02T15:04:05.999999999Z07:00+T}@T{+N+T}+T{+gid+T}@T{+Group ID of owner+T}@T{+decimal number+T}@T{+500+T}@T{+N+T}+T{+mode+T}@T{+File type and mode+T}@T{+octal, unix style+T}@T{+0100664+T}@T{+N+T}+T{+mtime+T}@T{+Time of last modification+T}@T{+RFC 3339+T}@T{+2006-01-02T15:04:05.999999999Z07:00+T}@T{+N+T}+T{+rdev+T}@T{+Device ID (if special file)+T}@T{+hexadecimal+T}@T{+1abc+T}@T{+N+T}+T{+uid+T}@T{+User ID of owner+T}@T{+decimal number+T}@T{+500+T}@T{+N+T}+.TE+.PP+See the metadata (https://rclone.org/docs/#metadata) docs for more info.+.SS Backend commands+.PP+Here are the commands specific to the local backend.+.PP+Run them with:+.IP+.nf+\f[C]+rclone backend COMMAND remote:+\f[R]+.fi+.PP+The help below will explain what arguments each command takes.+.PP+See the backend (https://rclone.org/commands/rclone_backend/) command+for more info on how to pass options and arguments.+.PP+These can be run on a running backend using the rc command+backend/command (https://rclone.org/rc/#backend-command).+.SS noop+.PP+A null operation for testing backend commands.+.IP+.nf+\f[C]+rclone backend noop remote: [options] [<arguments>+]+\f[R]+.fi+.PP+This is a test command which has some options you can try to change the+output.+.PP+Options:+.IP \[bu] 2+\[dq]echo\[dq]: Echo the input arguments.+.IP \[bu] 2+\[dq]error\[dq]: Return an error based on option value.+.SH Changelog+.SS v1.76.0 - 2026-11-06+.PP+See commits (https://github.com/rclone/rclone/compare/v1.75.0...v1.76.0)+.IP \[bu] 2+New backends+.RS 2+.IP \[bu] 2+Dosya (https://rclone.org/dosya/) (Kaya)+.IP \[bu] 2+Infomaniak kDrive (https://rclone.org/kdrive/) (Guillaume MALLERET)+.RE+.IP \[bu] 2+New S3 Providers+.RS 2+.IP \[bu] 2+Autonomys Auto Drive (https://rclone.org/s3/#autonomys) (Jim Counter)+.IP \[bu] 2+Clever Cloud Cellar (https://rclone.org/s3/#clever-cloud-cellar) (David+Legrand)+.IP \[bu] 2+Infomaniak Swiss Backup (https://rclone.org/s3/#infomaniak) (Louis)+.IP \[bu] 2+Neon Object Storage (https://rclone.org/s3/#neon) (Rishi Raj Jain)+.IP \[bu] 2+RelAix.Storage (https://rclone.org/s3/#relaix) (mygrexit)+.RE+.IP \[bu] 2+New commands+.RS 2+.IP \[bu] 2+index (https://rclone.org/commands/rclone_index/): Write static+directory listings into a remote (Nick Craig-Wood)+.RE+.IP \[bu] 2+Security+.RS 2+.IP \[bu] 2+build+.RS 2+.IP \[bu] 2+Disable Multipath TCP listeners by default to mitigate CVE-2026-98432+(Nick Craig-Wood)+.IP \[bu] 2+Fix multiple CVEs including CVE-2026-111943 by upgrading to go1.27.3+(Nick Craig-Wood)+.IP \[bu] 2+Update \f[V]golang.org/x/net\f[R] to v0.62.0 to fix multiple CVEs (Nick+Craig-Wood)+.IP \[bu] 2+Update \f[V]github.com/andybalholm/brotli\f[R] to v1.2.7 to fix+CVE-2026-111497 (Nick Craig-Wood)+.RE+.IP \[bu] 2+fshttp: Refuse HTTPS to HTTP redirects unless+\f[V]--allow-https-downgrade\f[R] GHSA-m97j-7x65-wh93 CVE-PENDING (Nick+Craig-Wood)+.IP \[bu] 2+lib/http: Stop \f[V]--htpasswd\f[R] and \f[V]--user\f[R] revealing which+usernames exist through timing GHSA-hwj7-wp9g-wrv9 CVE-PENDING (Nick+Craig-Wood)+.IP \[bu] 2+local+.RS 2+.IP \[bu] 2+Fix user metadata escaping the root through a planted symlink+GHSA-r5wv-vqqh-2r9c CVE-PENDING (Nick Craig-Wood)+.IP \[bu] 2+Fix MkdirMetadata finding directories outside the root through a planted+symlink (Nick Craig-Wood)+.IP \[bu] 2+Fix Remove and Rmdir deleting outside the root through a planted symlink+(Nick Craig-Wood)+.IP \[bu] 2+Fix DirMove creating the destination parent through a planted symlink+(Nick Craig-Wood)+.RE+.IP \[bu] 2+netstorage: Fix API auth headers leaking to other hosts on redirect+GHSA-m97j-7x65-wh93 CVE-PENDING (Nick Craig-Wood)+.IP \[bu] 2+rc+.RS 2+.IP \[bu] 2+Limit request size when the rc server has no authentication+GHSA-f8p4-8g4j-q3m2 CVE-PENDING (Nick Craig-Wood)+.IP \[bu] 2+Reject unauthorised and unknown calls before reading the request+GHSA-f8p4-8g4j-q3m2 CVE-PENDING (Nick Craig-Wood)+.RE+.IP \[bu] 2+seafile: Fix HTTPS to HTTP redirects leaking credentials+GHSA-m97j-7x65-wh93 CVE-PENDING (Nick Craig-Wood)+.IP \[bu] 2+serve: Make auth proxy revocation take effect within 5 minutes+GHSA-vmgx-g9c5-669p CVE-PENDING (Nick Craig-Wood)+.IP \[bu] 2+serve sftp+.RS 2+.IP \[bu] 2+Fix auth proxy letting a public key be used as a password+GHSA-66qv-66mg-9h74 CVE-PENDING (Nick Craig-Wood)+.IP \[bu] 2+Fix auth proxy running for public keys the client can\[aq]t use+GHSA-66qv-66mg-9h74 CVE-PENDING (Nick Craig-Wood)+.RE+.IP \[bu] 2+swift+.RS 2+.IP \[bu] 2+Fix auth token leaking to other hosts or over plaintext on redirect+GHSA-m97j-7x65-wh93 CVE-PENDING (Nick Craig-Wood)+.IP \[bu] 2+Fix password being resent to another host on auth redirect+GHSA-m97j-7x65-wh93 CVE-PENDING (Nick Craig-Wood)+.RE+.IP \[bu] 2+ulozto: Fix auth tokens leaking on redirects from download links+GHSA-3xpx-hjg2-vrjh CVE-PENDING (Nick Craig-Wood)+.RE+.IP \[bu] 2+New Features+.RS 2+.IP \[bu] 2+accounting+.RS 2+.IP \[bu] 2+Stop \f[V]--max-transfer\f[R] overshooting the limit (Nick Craig-Wood)+.IP \[bu] 2+Add \[dq]Updated dirs\[dq] count of directory modtime/metadata updates+to the stats (Nick Craig-Wood)+.RE+.IP \[bu] 2+batcher: Log when committing uploads has finished (phatlc)+.IP \[bu] 2+bisync+.RS 2+.IP \[bu] 2+Make \f[V]--max-delete\f[R] aware of \f[V]--track-renames\f[R] (Morax)+.IP \[bu] 2+Fall back to \f[V]--download\f[R] when a checksum is missing (nielash)+.IP \[bu] 2+Reject \f[V]--max-delete\f[R] above 100 (nielash)+.RE+.IP \[bu] 2+build+.RS 2+.IP \[bu] 2+Update to go1.27 and make go1.26 the minimum required version (Nick+Craig-Wood)+.IP \[bu] 2+Update all dependencies (Nick Craig-Wood)+.RE+.IP \[bu] 2+cmd: Complete enum flag values from their choices (Demiao Huang)+.IP \[bu] 2+config: Only run \f[V]--password-command\f[R] once when using+\f[V]--daemon\f[R] (phatlc)+.IP \[bu] 2+doc fixes (ashwinibailangadi, ferrumclaudepilgrim, frodleTian, FTCHD,+gomitrah, Nick Craig-Wood, qodelicious, Satvik Mishra, ZainnQureshii)+.IP \[bu] 2+docker: Make container user UID and GID configurable via build ARGs+(Vaibhav Mashal)+.IP \[bu] 2+filter+.RS 2+.IP \[bu] 2+Speed up large \f[V]--include-from\f[R] lists of directories (kaii9)+.IP \[bu] 2+Make \f[V][!...]\f[R] in globs a negated character class (kevin9327)+.IP \[bu] 2+Make \f[V][\[ha]...]\f[R] in globs exclude \f[V]/\f[R] and document it+(kevin9327)+.RE+.IP \[bu] 2+fs+.RS 2+.IP \[bu] 2+Reject sizes too large for SizeSuffix instead of silently disabling the+limit (KBS)+.IP \[bu] 2+Stop writing config for on the fly and environment variable remotes into+the config file (Nick Craig-Wood)+.IP \[bu] 2+Reject durations like 300y that overflow instead of wrapping (kevin9327)+.IP \[bu] 2+Accept multi-unit relative times like 1d12h in Time options (kevin)+.IP \[bu] 2+Reject NaN sizes and CountSuffix values that overflow (kevin9327)+.IP \[bu] 2+Reject \f[V]--bwlimit\f[R] times without a colon or with signed digits+(kevin9327)+.IP \[bu] 2+Reject NaN durations instead of wrapping (Soujas K)+.IP \[bu] 2+Add MIME type for JPEG XL (.jxl) files (Seth Westphal)+.RE+.IP \[bu] 2+gui+.RS 2+.IP \[bu] 2+Take the GUI out of beta (Nick Craig-Wood)+.IP \[bu] 2+Update embedded release to 1.2.1 (Nick Craig-Wood)+.RE+.IP \[bu] 2+hash: Add CRC-32C (Castagnoli) support (Dirk Petersen)+.IP \[bu] 2+index: Add partial runs which only re-index changed directories (Nick+Craig-Wood)+.IP \[bu] 2+lib/http+.RS 2+.IP \[bu] 2+Return 416 for ranges \f[V]--rc-serve\f[R] and serve restic can\[aq]t+satisfy (Rohit Behera)+.IP \[bu] 2+Replace \f[V]github.com/abbot/go-http-auth\f[R] with built in htpasswd+support (Nick Craig-Wood)+.IP \[bu] 2+Accept the Basic auth scheme in any case (kevin)+.RE+.IP \[bu] 2+lib/proxy: Send the decoded user and password in Proxy-Authorization+(kevin)+.IP \[bu] 2+lib/transform: Make titlecase capitalise words instead of upper-casing+them (kevin)+.IP \[bu] 2+ncdu: Show progress and the item count when deleting a selection (Lev+Devaev)+.IP \[bu] 2+operations+.RS 2+.IP \[bu] 2+Check checksums in rcat with known size (phatlc)+.IP \[bu] 2+Speed up checksum by hashing files in parallel (Teddy Tennant)+.IP \[bu] 2+Fall back to a single stream when a multi-thread source ignores ranges+(Morax)+.IP \[bu] 2+Show directory operations in the \f[V]--progress\f[R] display (Nick+Craig-Wood)+.IP \[bu] 2+Add \f[V]operations/getfile\f[R] remote control API endpoint (Hakan+İSMAİL)+.IP \[bu] 2+Make \f[V]--immutable\f[R] work with copyto, moveto and single file copy+(phatlc)+.IP \[bu] 2+Respect \f[V]--backup-dir\f[R] in deletefile (Cao Yuhang)+.RE+.IP \[bu] 2+rc+.RS 2+.IP \[bu] 2+Deprecate \f[V]--rc-web-gui\f[R] flag (Nick Craig-Wood)+.IP \[bu] 2+Add sync report parameters to \f[V]sync/sync\f[R], \f[V]sync/copy\f[R]+and \f[V]sync/move\f[R] (Nick Craig-Wood)+.RE+.IP \[bu] 2+rest: Reject negative sizes in HTTP response headers (Yousef Issa)+.IP \[bu] 2+serve http+.RS 2+.IP \[bu] 2+Sort directory listings in the browser (Nick Craig-Wood)+.IP \[bu] 2+Redesign the directory listing page (Nick Craig-Wood)+.RE+.IP \[bu] 2+serve s3+.RS 2+.IP \[bu] 2+Report the object\[aq]s modtime as Last-Modified after a PUT (kevin9327)+.IP \[bu] 2+Support changing object tags (Nick Craig-Wood)+.IP \[bu] 2+Quote the ETags of objects in listings (Nick Craig-Wood)+.IP \[bu] 2+Support listing with delimiters other than \[dq]/\[dq] (Nick Craig-Wood)+.IP \[bu] 2+Support directory markers and refuse file/directory conflicts (Nick+Craig-Wood)+.IP \[bu] 2+Make the ETag of multipart uploads agree with HEAD and support+partNumber (Nick Craig-Wood)+.IP \[bu] 2+Make conditional writes atomic against other writes through the server+(Nick Craig-Wood)+.IP \[bu] 2+Make the ETag returned on upload agree with HEAD for any+\f[V]--etag-hash\f[R] (Nick Craig-Wood)+.IP \[bu] 2+List empty directories as directory markers (Nick Craig-Wood)+.IP \[bu] 2+Allow copying directory markers (Nick Craig-Wood)+.IP \[bu] 2+Remove the directories made for a failed upload (Nick Craig-Wood)+.IP \[bu] 2+Keep the MD5 ETag of objects on remotes without MD5 (Nick Craig-Wood)+.IP \[bu] 2+Don\[aq]t read directories needlessly when listing with a delimiter+(Nick Craig-Wood)+.IP \[bu] 2+Don\[aq]t compute the MD5 of multipart uploads unless the ETag is MD5+(Nick Craig-Wood)+.IP \[bu] 2+List the directory markers put for directories holding other keys (Nick+Craig-Wood)+.IP \[bu] 2+Let a failed CompleteMultipartUpload be retried (Nick Craig-Wood)+.IP \[bu] 2+List objects with the storage class they were put with (Nick Craig-Wood)+.IP \[bu] 2+Add \f[V]--directory-markers\f[R] to list every directory as a marker+(Nick Craig-Wood)+.IP \[bu] 2+Remove a directory with its last key after its marker is deleted (Nick+Craig-Wood)+.IP \[bu] 2+Make \f[V]--no-cleanup\f[R] a deprecated alias of+\f[V]--directory-markers\f[R] (Nick Craig-Wood)+.IP \[bu] 2+Document which S3 features are supported and the design goals (Nick+Craig-Wood)+.RE+.IP \[bu] 2+serve webdav+.RS 2+.IP \[bu] 2+Sort directory listings in the browser (Nick Craig-Wood)+.IP \[bu] 2+Redesign the directory listing page (Nick Craig-Wood)+.RE+.IP \[bu] 2+size: Include source in JSON output (Jiacheng Xu)+.IP \[bu] 2+sync: Add pattern-based transfer ordering (Shurong Cao)+.IP \[bu] 2+touch: Parse documented timestamp formats (Nicholas Velten)+.RE+.IP \[bu] 2+Bug Fixes+.RS 2+.IP \[bu] 2+accounting: Fix totalBytes counting server-side copies and moves twice+(ilya lavrenov)+.IP \[bu] 2+bisync+.RS 2+.IP \[bu] 2+Fix conflict winner missing from listings when Path2 wins (nielash)+.IP \[bu] 2+Fix \f[V]--conflict-resolve\f[R] comparing a zero size or modtime+(nielash)+.IP \[bu] 2+Fix endless retry loop for skipped files (nielash)+.IP \[bu] 2+Fix listings ignoring a modtime less than 1s older (nielash)+.IP \[bu] 2+Fix rc ignoring global \f[V]--dry-run\f[R] and \f[V]--max-delete\f[R]+settings (nielash)+.IP \[bu] 2+Fix rc ignoring legacy backupdir1/2 and invalid values (nielash)+.IP \[bu] 2+Fix internal error when a conflict rename reuses deleted names+(TastyHeadphones)+.RE+.IP \[bu] 2+fs: Fix durations like 1d12h being rejected (Rohit Behera)+.IP \[bu] 2+lib/encoder: Fix decoding a trailing character after an escaped quote+(kevin9327)+.IP \[bu] 2+list: Fix ext sort error sometimes being ignored at the end of a listing+(Nick Craig-Wood)+.IP \[bu] 2+ncdu: Update to tcell v3 which fixes duplicated keys on Windows (Nick+Craig-Wood)+.IP \[bu] 2+operations+.RS 2+.IP \[bu] 2+Fix making a directory with modtime asking twice with+\f[V]--interactive\f[R] (Nick Craig-Wood)+.IP \[bu] 2+Fix case- or Unicode-only renames deleting the file (nielash)+.IP \[bu] 2+Fix moving a file onto its own name deleting it (nielash)+.IP \[bu] 2+Fix cat \f[V]--offset\f[R] before the start of a file printing its end+(b84330808)+.IP \[bu] 2+Fix oversized getfile tail requests (唐僧洗发用飘柔)+.RE+.IP \[bu] 2+rc: Fix LogLevel in \f[V]_config\f[R] being ignored (Nick Craig-Wood)+.IP \[bu] 2+rest: Fix retries reading the request body after the request has+finished (Nick Craig-Wood)+.IP \[bu] 2+serve dlna: Fix crash on Browse with a negative StartingIndex or+RequestedCount (phatlc)+.IP \[bu] 2+serve restic: Fix temporary errors confusing restic into leaving locks+behind (Azhar)+.IP \[bu] 2+serve s3+.RS 2+.IP \[bu] 2+Update to gofakes3 v0.0.8 which fixes 500 errors with go1.27 (Nick+Craig-Wood)+.IP \[bu] 2+Upgrade gofakes3 to v0.0.11 to fix IsTruncated and Docker Distribution+Registry (DEVN)+.IP \[bu] 2+Fix buckets being undeletable after an aborted multipart upload (Nick+Craig-Wood)+.IP \[bu] 2+Fix CopyObject metadata including the source\[aq]s and an mtime (Nick+Craig-Wood)+.RE+.IP \[bu] 2+serve sftp: Fix rmdir deleting a file instead of returning an error+(phatlc)+.IP \[bu] 2+sync+.RS 2+.IP \[bu] 2+Fix directory modtimes being set twice (Nick Craig-Wood)+.IP \[bu] 2+Fix \f[V]--delete-empty-src-dirs\f[R] disabled by a prior error in a+reused stats group (ghwany)+.RE+.IP \[bu] 2+walk: Fix duplicate directories with \f[V]--fast-list\f[R] when+directories are listed after their contents (Nick Craig-Wood)+.RE+.IP \[bu] 2+Mount+.RS 2+.IP \[bu] 2+Fix VFS instance leaks on mount failures and unmounts (Hakan İSMAİL)+.RE+.IP \[bu] 2+VFS+.RS 2+.IP \[bu] 2+Preserve shrinking truncate on uncached files (47th)+.IP \[bu] 2+Fix crash renaming a directory while the dir cache expires (Shreyansh+Jain)+.IP \[bu] 2+Fix virtual entries leak in removed directory (Biagio Peccerillo)+.IP \[bu] 2+Fix renamed items listed twice after a case or Unicode-only match+(nielash)+.IP \[bu] 2+Fix uploads never starting while a dirty file keeps being read (Neil+Cawse)+.RE+.IP \[bu] 2+Local+.RS 2+.IP \[bu] 2+Speed up default checksummed copies by writing in larger blocks (Dominik+Sander)+.IP \[bu] 2+Make out of space errors fatal during multi-thread transfers+(ferrumclaudepilgrim)+.IP \[bu] 2+Stop \f[V]--copy-links\f[R] following symlink loops (phatlc)+.IP \[bu] 2+Limit the size of symlink targets read from .rclonelink files (Nick+Craig-Wood)+.IP \[bu] 2+Fix server-side copy on macOS when the destination exists (Ayush Sharma)+.IP \[bu] 2+Set only the ownership a non-root user may set (Dirk Petersen)+.RE+.IP \[bu] 2+Crypt+.RS 2+.IP \[bu] 2+Fix nil pointer panic in ObjectInfo.Hash (fixfon)+.RE+.IP \[bu] 2+Archive+.RS 2+.IP \[bu] 2+Read symlinks stored in zip and squashfs archives (Lev Devaev)+.IP \[bu] 2+Fix reading an archive in the root of a remote (Oshino Penguin)+.RE+.IP \[bu] 2+Azure Blob+.RS 2+.IP \[bu] 2+Enable on Solaris (acoeur)+.IP \[bu] 2+Fix directories on storage accounts with hierarchical namespace (Nick+Craig-Wood)+.IP \[bu] 2+Fix NewObject on a directory returning \[dq]is not a regular file\[dq]+(Nick Craig-Wood)+.IP \[bu] 2+Fix listing a file as a directory on hierarchical namespace accounts+(Nick Craig-Wood)+.IP \[bu] 2+Fix sync at the root failing on soft deleted containers (Ludovic Bondon)+.RE+.IP \[bu] 2+Box+.RS 2+.IP \[bu] 2+Reuse multipart upload part buffers via the global pool (Nick+Craig-Wood)+.IP \[bu] 2+Fix \[ti] and environment variables not being expanded in+\f[V]box_config_file\f[R] (Nick Craig-Wood)+.RE+.IP \[bu] 2+Chunker+.RS 2+.IP \[bu] 2+Clean stale chunks after server-side overwrite (hahahahahayesyeseys)+.RE+.IP \[bu] 2+Compress+.RS 2+.IP \[bu] 2+Buffer small uploads in the global memory pool (Nick Craig-Wood)+.IP \[bu] 2+Stop buffering the compressibility check output in memory (Nick+Craig-Wood)+.RE+.IP \[bu] 2+Drime+.RS 2+.IP \[bu] 2+Fix endless listing of directories with more than 20,000 entries (Maksym+Tymoshyk)+.RE+.IP \[bu] 2+Drive+.RS 2+.IP \[bu] 2+Fetch metadata permissions concurrently (Loi Nguyen)+.IP \[bu] 2+Reuse resumable-upload chunk buffers via the global memory pool+(SillyZir)+.IP \[bu] 2+Stop a fast-list error from hanging the sync (Sash)+.RE+.IP \[bu] 2+Dropbox+.RS 2+.IP \[bu] 2+Reuse upload chunk buffers via the global memory pool (Nick Craig-Wood)+.RE+.IP \[bu] 2+Filelu+.RS 2+.IP \[bu] 2+Reuse multipart upload buffers via the global pool and retry failed+parts (Nick Craig-Wood)+.IP \[bu] 2+Encode multipart init paths (wjc)+.RE+.IP \[bu] 2+FTP+.RS 2+.IP \[bu] 2+Discard connections that fail with non-FTP errors (Daniel Bae)+.IP \[bu] 2+Fix ranged reads failing when the server answers 425 to an early close+(Robert Kubiś)+.RE+.IP \[bu] 2+Hasher+.RS 2+.IP \[bu] 2+Keep cached hash when only the modtime changes (Aniket)+.RE+.IP \[bu] 2+Hidrive+.RS 2+.IP \[bu] 2+Stop copying every upload chunk a second time (Nick Craig-Wood)+.IP \[bu] 2+Buffer the start of each upload from the global memory pool (Nick+Craig-Wood)+.IP \[bu] 2+Reuse chunked upload buffers via the global memory pool (Nick+Craig-Wood)+.IP \[bu] 2+Buffer small file updates from the global memory pool (Nick Craig-Wood)+.IP \[bu] 2+Fix truncated file left behind when an upload hits+\f[V]--max-transfer\f[R] (Nick Craig-Wood)+.RE+.IP \[bu] 2+HTTP+.RS 2+.IP \[bu] 2+Add metadata defaults for \f[V]--http-no-head\f[R] (Giridhar)+.RE+.IP \[bu] 2+Huaweidrive+.RS 2+.IP \[bu] 2+Build small uploads in pooled memory instead of a growing buffer (Nick+Craig-Wood)+.IP \[bu] 2+Reuse resumable-upload chunk buffers via the global pool (Nick+Craig-Wood)+.IP \[bu] 2+Fix creating, copying and renaming files with non-ASCII names (Nick+Craig-Wood)+.RE+.IP \[bu] 2+Iclouddrive+.RS 2+.IP \[bu] 2+Mark the backend as case insensitive (berce)+.RE+.IP \[bu] 2+Imagekit+.RS 2+.IP \[bu] 2+Stop multipart writers after failed uploads (Florian Josef Altendorfer)+.RE+.IP \[bu] 2+Jottacloud+.RS 2+.IP \[bu] 2+Buffer MD5 pre-reads in the global memory pool (Nick Craig-Wood)+.IP \[bu] 2+Fix server-side copy onto a trashed file failing as corrupted (Nick+Craig-Wood)+.IP \[bu] 2+Fix server-side copy onto a trashed file keeping the old name case (Nick+Craig-Wood)+.IP \[bu] 2+Fix authentication with whitelabel services elkjop and elko (albertony)+.IP \[bu] 2+Fix traditional authentication failing with invalid_client (albertony)+.RE+.IP \[bu] 2+Linkbox+.RS 2+.IP \[bu] 2+Buffer the hashed 10 MiB file prefix in the global memory pool (Nick+Craig-Wood)+.RE+.IP \[bu] 2+Mailru+.RS 2+.IP \[bu] 2+Buffer speedup hashing in the global memory pool and fix upload retries+(Nick Craig-Wood)+.RE+.IP \[bu] 2+Onedrive+.RS 2+.IP \[bu] 2+OneDrive for Business and SharePoint users must upgrade to v1.76+ before+April 1, 2027 (Nick Craig-Wood)+.IP \[bu] 2+Authenticate SharePoint downloads, uploads and copies after tempauth+retirement (Nick Craig-Wood)+.IP \[bu] 2+Return an error if a ranged download doesn\[aq]t return the requested+range (Nick Craig-Wood)+.IP \[bu] 2+Reuse upload chunk buffers via the global memory pool (Nick Craig-Wood)+.IP \[bu] 2+Allow \f[V]--onedrive-upload-cutoff\f[R] up to the documented limit of+250 MiB (Can Arslan)+.IP \[bu] 2+Add configurable SharePoint tenant API version (Mattias Michaux)+.IP \[bu] 2+Make \f[V]--onedrive-delta\f[R] quicker when not listing from the drive+root (Nick Craig-Wood)+.IP \[bu] 2+Keep the delta token when a ChangeNotify poll fails (xiaozou-wine)+.IP \[bu] 2+Show the size of directories in listings (GingerAdonis)+.IP \[bu] 2+Poll async copy status with the configured HTTP client (Soujas K)+.RE+.IP \[bu] 2+Opendrive+.RS 2+.IP \[bu] 2+Reuse upload chunk buffers via the global memory pool (Nick Craig-Wood)+.RE+.IP \[bu] 2+Oracleobjectstorage+.RS 2+.IP \[bu] 2+Add server-side object moves (tomaszni)+.IP \[bu] 2+Encode carriage returns and line feeds (tomaszni)+.IP \[bu] 2+Purge objects in batches (tomaszni)+.IP \[bu] 2+Retry 409 ConcurrentObjectUpdate errors (berce)+.RE+.IP \[bu] 2+Pcloud+.RS 2+.IP \[bu] 2+Stop multipart writers after failed uploads (Florian Josef Altendorfer)+.RE+.IP \[bu] 2+Pikpak+.RS 2+.IP \[bu] 2+Share upload chunk buffers with the global memory pool (Nick Craig-Wood)+.IP \[bu] 2+Use pooled memory for the gcid hash buffer and spool unknown-size+uploads to disk (Nick Craig-Wood)+.IP \[bu] 2+Fix downloads that return an HTTP error page as the file (Cestercian)+.IP \[bu] 2+Stop multipart writers after failed uploads (Florian Josef Altendorfer)+.IP \[bu] 2+Fix uploads failing when the server is slow to complete them (Nick+Craig-Wood)+.RE+.IP \[bu] 2+Putio+.RS 2+.IP \[bu] 2+Reuse upload chunk buffers via the global memory pool (Nick Craig-Wood)+.RE+.IP \[bu] 2+S3+.RS 2+.IP \[bu] 2+Avoid buffering Object Lock uploads when the source MD5 is known (Nick+Craig-Wood)+.IP \[bu] 2+Add backend link command with signed response header overrides (eliotee)+.IP \[bu] 2+Add Fastly EU Central 1 region (Leon Brocard)+.IP \[bu] 2+Add the MD5 of multipart chunks to the debug log (interested.tortoise)+.IP \[bu] 2+Report the real storage class when an object needs restoring (Dirk+Petersen)+.IP \[bu] 2+Speed up multi-thread uploads by checksumming chunks in 1 MiB reads+(Roman Butor)+.IP \[bu] 2+Add IT-MIL region to Scaleway (Alessio Margelli)+.IP \[bu] 2+Add missing Huawei OBS regions (Enbiya Göral)+.IP \[bu] 2+Fix rmdir of the root directory with \f[V]--s3-directory-markers\f[R]+(Nick Craig-Wood)+.IP \[bu] 2+Fix presigned URLs for GCS, Linode and Ceph (Leon Brocard)+.RE+.IP \[bu] 2+Seafile+.RS 2+.IP \[bu] 2+Fix operations on files whose names are not in Unicode NFC form (Fred)+.IP \[bu] 2+Stop multipart writers after failed uploads (Florian Josef Altendorfer)+.IP \[bu] 2+Fix double slash in the URL when getting a 2FA token (Nick Craig-Wood)+.RE+.IP \[bu] 2+SFTP+.RS 2+.IP \[bu] 2+Implement multi-thread uploads (Splainte)+.IP \[bu] 2+Fix data race when using \f[V]--sftp-ssh\f[R] (Dirk Petersen)+.IP \[bu] 2+Fix hash check for names starting with \[dq]-\[dq] with relative+\f[V]--sftp-path-override\f[R] (Dirk Einecke)+.IP \[bu] 2+Fix sync deleting files with control characters in their names (nielash)+.IP \[bu] 2+Fix hashing files with a newline in their name on rsync.net (Nick+Craig-Wood)+.RE+.IP \[bu] 2+Shade+.RS 2+.IP \[bu] 2+Stop copying each upload chunk into a second heap buffer (Nick+Craig-Wood)+.RE+.IP \[bu] 2+Smb+.RS 2+.IP \[bu] 2+Add workstation field to smb backend (Kalin Stoyanov)+.RE+.IP \[bu] 2+Swift+.RS 2+.IP \[bu] 2+Add Infomaniak Swiss Backup provider (Louis)+.RE+.IP \[bu] 2+WebDAV+.RS 2+.IP \[bu] 2+Reject ranged responses which ignore the range (Morax)+.IP \[bu] 2+Retry Nextcloud chunk uploads individually (Stochastic)+.IP \[bu] 2+Fix Nextcloud public share access by sending X-Requested-With+(drakeo338)+.IP \[bu] 2+Fix memory leak uploading to Infinite Scale (jxj)+.RE+.IP \[bu] 2+Yandex+.RS 2+.IP \[bu] 2+Add \f[V]app_folder\f[R] option to support+\f[V]cloud_api:disk.app_folder\f[R] OAuth scope (Sanjay Kanth A)+.RE+.IP \[bu] 2+Zoho+.RS 2+.IP \[bu] 2+Stop multipart writers after failed uploads (Florian Josef Altendorfer)+.RE+.SS v1.75.2 - 2026-10-09+.PP+See commits (https://github.com/rclone/rclone/compare/v1.75.1...v1.75.2)+.IP \[bu] 2+Security+.RS 2+.IP \[bu] 2+b2: Fix account token and SSE-C key leaking on redirects from+\f[V]--b2-download-url\f[R] GHSA-gpw7-77cq-9f72 CVE-2026-116196 (Nick+Craig-Wood)+.IP \[bu] 2+build+.RS 2+.IP \[bu] 2+Fix multiple CVEs by upgrading to go1.26.9 (Nick Craig-Wood)+.RS 2+.IP \[bu] 2+CVE-2026-97032: net/http: HTTP/2 server crash due to HPACK encoder race+.IP \[bu] 2+CVE-2026-78659: net/http: HTTP/2 server memory exhaustion due to Trailer+headers+.IP \[bu] 2+CVE-2026-97031: crypto/tls: reject malformed ECH outer extension+references+.IP \[bu] 2+CVE-2026-94444: cmd/go: checksum bypass for golang.org/fips140+.IP \[bu] 2+CVE-2026-94447: cmd/go: checksum database bypass for+golang.org/toolchain+.IP \[bu] 2+CVE-2026-94448: html/template: reset context tracking on consecutive+template expressions+.IP \[bu] 2+CVE-2026-97030: html/template: recognize |yield| as regexp preceder+keyword+.IP \[bu] 2+CVE-2026-94440: net/textproto, mime/multipart: memory limit bypass when+parsing MIME headers+.IP \[bu] 2+CVE-2026-56866: net/http: HTTP/1 client connection desynchronization+after CONNECT rejection+.IP \[bu] 2+CVE-2026-94439: net/http: HTTP/1 server connection desynchronization+after 2xx CONNECT response+.IP \[bu] 2+CVE-2026-78669: net/http: excessive CPU consumption from repeated+initial window changes+.IP \[bu] 2+CVE-2026-78660: net/http: HTTP/2 transport accepts malformed+framing-related headers+.IP \[bu] 2+CVE-2026-56857: os: Root.Mkdir(All) can follow junctions out of the root+on Windows+.IP \[bu] 2+CVE-2026-78667: net/http: lack of limit on size of parsed Range headers+.IP \[bu] 2+CVE-2026-78663: net/http: double flow control refund on HTTP/2 server+streams+.RE+.IP \[bu] 2+Update golang.org/x/net to v0.60.0 to fix multiple CVEs (Nick+Craig-Wood)+.RS 2+.IP \[bu] 2+CVE-2026-78659: net/http: HTTP/2 server memory exhaustion due to Trailer+headers+.IP \[bu] 2+CVE-2026-78669: net/http: excessive CPU consumption from repeated+initial window changes+.IP \[bu] 2+CVE-2026-78663: net/http: double flow control refund on HTTP/2 server+streams+.IP \[bu] 2+CVE-2026-97032: net/http: HTTP/2 server crash due to HPACK encoder race+.RE+.IP \[bu] 2+Update google.golang.org/grpc to fix CVE-2026-84445 (dependabot[bot])+.RS 2+.IP \[bu] 2+CVE-2026-84445: xds: server panic on a request missing both :authority+and Host headers+.RE+.RE+.IP \[bu] 2+fshttp: Drop \f[V]--header\f[R] values over HTTP after HTTPS downgrade+GHSA-rrwf-2qr6-p832 CVE-PENDING (Nick Craig-Wood)+.IP \[bu] 2+http: Stop a path starting with // moving the remote to another host+GHSA-xqmq-v4wj-rw5m CVE-2026-116192 (Nick Craig-Wood)+.IP \[bu] 2+lib/http: Require authentication for OPTIONS requests+GHSA-gj73-fh6v-92fj CVE-2026-116193 (Nick Craig-Wood)+.IP \[bu] 2+local: Fix .rclonelink names escaping the root with \f[V]--links\f[R]+GHSA-55qp-jrwr-x827 CVE-2026-116195 (Nick Craig-Wood)+.IP \[bu] 2+netstorage+.RS 2+.IP \[bu] 2+Stop a path starting with // moving the remote to another host+GHSA-xqmq-v4wj-rw5m CVE-2026-116192 (Nick Craig-Wood)+.IP \[bu] 2+Stop a listing resume token moving signed requests to another host+GHSA-xqmq-v4wj-rw5m CVE-2026-116192 (Nick Craig-Wood)+.RE+.IP \[bu] 2+pcloud+.RS 2+.IP \[bu] 2+Fix OAuth config accepting callbacks without state GHSA-hvpr-p4pv-4f46+CVE-2026-116194 (Nick Craig-Wood)+.IP \[bu] 2+Only accept pCloud hosts from the hostname in the OAuth callback+GHSA-hvpr-p4pv-4f46 (Nick Craig-Wood)+.RE+.IP \[bu] 2+selfupdate+.RS 2+.IP \[bu] 2+Fix hash check trusting unsigned data in the signed SHA256SUMS+GHSA-w33h-3qgq-8mv8 CVE-2026-116189 (Nick Craig-Wood)+.IP \[bu] 2+Fix release verification instructions trusting unsigned hashes+GHSA-w33h-3qgq-8mv8 (Nick Craig-Wood)+.RE+.IP \[bu] 2+serve nfs: Fix path traversal by refusing unsafe file names in the VFS+GHSA-44pm-2q5p-fm63 CVE-2026-116191 (Nick Craig-Wood)+.IP \[bu] 2+serve s3+.RS 2+.IP \[bu] 2+Fix presigned URL being turned into a copy of any object+GHSA-w3h7-jxxq-vfrp CVE-PENDING (Nick Craig-Wood)+.IP \[bu] 2+Stop auth proxy users listing or using each other\[aq]s multipart+uploads and metadata GHSA-35g6-9fjv-rmx5 CVE-2026-116187 (Nick+Craig-Wood)+.IP \[bu] 2+Fix multipart buffer limits being bypassed by overflowing part sizes+GHSA-qjw8-8hmp-c3m6 CVE-2026-116188 (Nick Craig-Wood)+.IP \[bu] 2+Add \f[V]--multipart-max-uploads\f[R] to limit uploads in progress (Nick+Craig-Wood)+.IP \[bu] 2+Add \f[V]--multipart-streaming-buffer-total\f[R] to limit memory across+uploads (Nick Craig-Wood)+.IP \[bu] 2+Add \f[V]--metadata-max-objects\f[R] to limit the metadata kept in+memory (Nick Craig-Wood)+.RE+.IP \[bu] 2+serve webdav: Fix path traversal above the served directory+GHSA-44pm-2q5p-fm63 CVE-2026-116191 (Nick Craig-Wood)+.IP \[bu] 2+webdav: Fix headers option leaking to other hosts on redirect+GHSA-3rqf-7h69-gg5v CVE-2026-116190 (Nick Craig-Wood)+.RE+.IP \[bu] 2+Bug Fixes+.RS 2+.IP \[bu] 2+accounting+.RS 2+.IP \[bu] 2+Stop averaging when the last check finishes (Kunpeng Xie)+.IP \[bu] 2+Keep speed averaging active when transfers overlap (kaii9)+.RE+.IP \[bu] 2+dedupe: Fix rename mode giving up after 100 names and make it faster+(Nick Craig-Wood)+.IP \[bu] 2+doc fixes (Dhevenddra K G, enkvadrat, GhostCoder6969, James Truitt,+maximilize, n4n5, Nick Craig-Wood, PSR94, Rohit Behera, shaurya, subomi)+.IP \[bu] 2+docker: Fix files uploaded with the wrong mime type (hsdfat)+.IP \[bu] 2+fs+.RS 2+.IP \[bu] 2+Fix about showing a negative total when a quota reaches the int64+maximum (youdie006)+.IP \[bu] 2+Fix bandwidth limits below 1 KiB being multiplied by 1024 (Roland)+.IP \[bu] 2+Fix \f[V]--bwlimit\f[R] timetables written out of order using the wrong+limit (Rohit Behera)+.IP \[bu] 2+Make BwTimetable.Set replace the timetable instead of appending to it+(youdie006)+.IP \[bu] 2+Retry \[dq]network is unreachable\[dq] and \[dq]network is down\[dq]+errors (nielash)+.RE+.IP \[bu] 2+fserrors: Fix out of space detection on Windows (ferrumclaudepilgrim)+.IP \[bu] 2+lib/rest: Limit the size of HTTP response bodies read into memory (Nick+Craig-Wood)+.IP \[bu] 2+log: Fix race when adding a log output while logging (Nick Craig-Wood)+.IP \[bu] 2+oauthutil: Fix Renew.Shutdown nil timer panic and race (TastyHeadphones)+.IP \[bu] 2+operations+.RS 2+.IP \[bu] 2+Stop \f[V]--copy-dest\f[R] replacing files with \f[V]--immutable\f[R]+(phatlc)+.IP \[bu] 2+Fix ignored error in rcat probe reads (Wang Chencheng)+.IP \[bu] 2+Fix hang when deleting files and a fatal error occurs (Nick Craig-Wood)+.IP \[bu] 2+Fix hang and lost context when moving a directory file by file (Nick+Craig-Wood)+.IP \[bu] 2+Don\[aq]t sleep for a Retry-After error when the transfer is cancelled+(Nick Craig-Wood)+.RE+.IP \[bu] 2+rc+.RS 2+.IP \[bu] 2+Make job/status and job/list require authentication (Nick Craig-Wood)+.IP \[bu] 2+Fix rc API accepting an out of range number and overflowing 64 bits+(KBS)+.IP \[bu] 2+Reject out of range integer parameters instead of truncating them (Nick+Craig-Wood)+.IP \[bu] 2+Fix large numeric parameters being rejected on 32 bit builds (Jeremy+Schoemaker)+.RE+.IP \[bu] 2+selfupdate: Fix \f[V]--version X.Y\f[R] depending on ./ in the download+site links (Nick Craig-Wood)+.IP \[bu] 2+serve+.RS 2+.IP \[bu] 2+Fix \f[V]--auth-proxy\f[R] stopping VFS cache uploads after 5 minutes+(Nick Craig-Wood)+.IP \[bu] 2+Fix \f[V]--auth-proxy\f[R] backends never being shut down after+simultaneous logins (Nick Craig-Wood)+.RE+.IP \[bu] 2+serve dlna: Log unescaped paths (phatlc)+.IP \[bu] 2+serve docker: Fix volume path being lost when the plugin restarts+(phatlc)+.IP \[bu] 2+serve ftp+.RS 2+.IP \[bu] 2+Fix transfers failing after 5 minutes with \f[V]--auth-proxy\f[R] (Nick+Craig-Wood)+.IP \[bu] 2+Fix VFS leak when the server fails to start (Nick Craig-Wood)+.RE+.IP \[bu] 2+serve http+.RS 2+.IP \[bu] 2+Fix downloads failing after 5 minutes with \f[V]--auth-proxy\f[R] (Nick+Craig-Wood)+.IP \[bu] 2+Escape filename in zip download Content-Disposition header (Harsh Raj+Singhania)+.RE+.IP \[bu] 2+serve restic+.RS 2+.IP \[bu] 2+Fix repositories named data (kaii9)+.IP \[bu] 2+Prevent concurrent append-only overwrites (Vasek Sraier)+.RE+.IP \[bu] 2+serve s3+.RS 2+.IP \[bu] 2+Apply the multipart upload limits and \f[V]--metadata-max-objects\f[R]+to each user separately (Nick Craig-Wood)+.IP \[bu] 2+Stream in-order multipart parts without buffering them (Nick Craig-Wood)+.IP \[bu] 2+Start the upload to the backend on the first multipart part (Nick+Craig-Wood)+.IP \[bu] 2+Don\[aq]t buffer re-uploads of multipart parts already streamed (Nick+Craig-Wood)+.IP \[bu] 2+Count buffered multipart parts in whole memory pages (Nick Craig-Wood)+.IP \[bu] 2+Time out multipart parts waiting for buffer space (Nick Craig-Wood)+.IP \[bu] 2+Limit the memory requests can use (Nick Craig-Wood)+.IP \[bu] 2+Forget expired multipart uploads completely (Nick Craig-Wood)+.IP \[bu] 2+Forget the metadata of deleted objects (Nick Craig-Wood)+.IP \[bu] 2+List objects lazily so paging a deep hierarchy is fast (Nick Craig-Wood)+.IP \[bu] 2+Fix multipart upload corruption if the backend write of an early part+fails (Nick Craig-Wood)+.IP \[bu] 2+Fix uploads failing when the auth proxy expires the user\[aq]s backend+(Nick Craig-Wood)+.IP \[bu] 2+Fix stale metadata being returned for objects changed elsewhere (Nick+Craig-Wood)+.IP \[bu] 2+Fix \f[V]--etag-hash auto\f[R] crashing or using the wrong hash with+\f[V]--auth-proxy\f[R] (Nick Craig-Wood)+.IP \[bu] 2+Fix CopyObject of a missing object not returning NoSuchKey (Nick+Craig-Wood)+.IP \[bu] 2+Fix log messages with bad format strings (Nick Craig-Wood)+.IP \[bu] 2+Document memory use and bring the docs up to date (Nick Craig-Wood)+.RE+.IP \[bu] 2+serve webdav+.RS 2+.IP \[bu] 2+Fix transfers failing after 5 minutes with \f[V]--auth-proxy\f[R] (Nick+Craig-Wood)+.IP \[bu] 2+Fix crash when the server fails to start (Nick Craig-Wood)+.IP \[bu] 2+Escape filename in zip download Content-Disposition header (Harsh Raj+Singhania)+.RE+.IP \[bu] 2+vfscache: Fix hang when the cache cleaner is disabled+(ferrumclaudepilgrim)+.RE+.IP \[bu] 2+Mount+.RS 2+.IP \[bu] 2+Fix \f[V]--daemon\f[R] failing when \f[V]--rc\f[R] is enabled (somaz)+.IP \[bu] 2+Ignore com.apple.* xattrs on the macOS FSKit backend (Eugene)+.RE+.IP \[bu] 2+VFS+.RS 2+.IP \[bu] 2+Fix reads returning zeros after a file changes during handle caching+(Neil Cawse)+.IP \[bu] 2+Stop setting a file\[aq]s modtime from discarding its cached data (Neil+Cawse)+.IP \[bu] 2+Fix crash reading the metadata of a file which is being written (Nick+Craig-Wood)+.IP \[bu] 2+Fix a VFS being reused while it is being shut down (Nick Craig-Wood)+.IP \[bu] 2+Fix AddVirtual ignoring isDir (phatlc)+.IP \[bu] 2+Stop \f[V]--links\f[R] objects called \[dq]...rclonelink\[dq] appearing+as \[dq]..\[dq] in directory listings (Nick Craig-Wood)+.RE+.IP \[bu] 2+Local+.RS 2+.IP \[bu] 2+Fix directories named *.rclonelink being treated as links with+\f[V]--links\f[R] (Nick Craig-Wood)+.RE+.IP \[bu] 2+Archive+.RS 2+.IP \[bu] 2+Fix corrupt listings when listing a zip directory more than once (Nick+Craig-Wood)+.IP \[bu] 2+Fix zip file entries named for a directory causing confusion (Nick+Craig-Wood)+.IP \[bu] 2+Fix listing entries with a leading slash as if they were in the root+(Nick Craig-Wood)+.IP \[bu] 2+Fix crash when creating archive to stdout (Vladimir Babin)+.IP \[bu] 2+Fix a squashfs path to a single file listing its whole directory (Nick+Craig-Wood)+.RE+.IP \[bu] 2+Azure Blob+.RS 2+.IP \[bu] 2+Use the released Azure SDK for Apache Arrow listing (Nick Craig-Wood)+.RE+.IP \[bu] 2+Box+.RS 2+.IP \[bu] 2+Fix log messages with bad format strings (Nick Craig-Wood)+.RE+.IP \[bu] 2+Chunker+.RS 2+.IP \[bu] 2+Fix panic listing a directory after an interrupted upload with+meta_format none (Nick Craig-Wood)+.RE+.IP \[bu] 2+Compress+.RS 2+.IP \[bu] 2+Fix crash on ranged reads when gzip metadata is corrupted (Acts1631)+.RE+.IP \[bu] 2+Drime+.RS 2+.IP \[bu] 2+Fix server-side copy over an existing file leaving a \[dq]name (1)\[dq]+copy (Nick Craig-Wood)+.IP \[bu] 2+Fix deleted files and directories still being listed with hard_delete+(Nick Craig-Wood)+.RE+.IP \[bu] 2+Drive+.RS 2+.IP \[bu] 2+Document the Branding step needed to publish your own client_id (Sanjay+Kanth A)+.RE+.IP \[bu] 2+Dropbox+.RS 2+.IP \[bu] 2+Fix shared folder mount for roots nested more than one level deep+(phatlc)+.IP \[bu] 2+Match shared-folder and received-file names case-insensitively (phatlc)+.RE+.IP \[bu] 2+Filescom+.RS 2+.IP \[bu] 2+Fix corrupted uploads after a retried upload error (Nick Craig-Wood)+.IP \[bu] 2+Document that names with / or \ in can\[aq]t be stored (Nick Craig-Wood)+.RE+.IP \[bu] 2+HTTP+.RS 2+.IP \[bu] 2+Fix crash when the server fails to start (Nick Craig-Wood)+.RE+.IP \[bu] 2+Iclouddrive+.RS 2+.IP \[bu] 2+Fix potential crash looking up items (Nick Craig-Wood)+.RE+.IP \[bu] 2+Imagekit+.RS 2+.IP \[bu] 2+Fix potential crash in rmdir and purge (Nick Craig-Wood)+.RE+.IP \[bu] 2+Internxt+.RS 2+.IP \[bu] 2+Fix lookups of files starting with a dot and dropped uploads (jzunigax2)+.IP \[bu] 2+Fix sync with \f[V]--backup-dir\f[R] or \f[V]--suffix\f[R] deleting the+backed up file (Nick Craig-Wood)+.IP \[bu] 2+Fix server-side moves failing with \[dq]Not Found\[dq] or \[dq]already+exists\[dq] (Nick Craig-Wood)+.IP \[bu] 2+Fix \[dq]directory not empty\[dq] and stale directories after moves and+deletes (Nick Craig-Wood)+.IP \[bu] 2+Fix server-side directory move failing after a gateway timeout (Nick+Craig-Wood)+.RE+.IP \[bu] 2+Onedrive+.RS 2+.IP \[bu] 2+Update docs for versions, links and time precision on personal accounts+(Nick Craig-Wood)+.IP \[bu] 2+Document that shared with me shortcuts may fail to list on personal+(Nick Craig-Wood)+.IP \[bu] 2+Update \f[V]--onedrive-hard-delete\f[R] docs to mention personal account+support (foecmke)+.RE+.IP \[bu] 2+Oracleobjectstorage+.RS 2+.IP \[bu] 2+Fix SSE-C server-side copies (tomaszni)+.IP \[bu] 2+Upload empty streams without multipart (tomaszni)+.RE+.IP \[bu] 2+Pixeldrain+.RS 2+.IP \[bu] 2+Fix corrupted uploads after a retried upload error (Nick Craig-Wood)+.RE+.IP \[bu] 2+Premiumizeme+.RS 2+.IP \[bu] 2+Fix uploading files with \[dq];\[dq] in their names (Nick Craig-Wood)+.RE+.IP \[bu] 2+S3+.RS 2+.IP \[bu] 2+Disable signing Accept-Encoding for Ceph and Linode (Aditya)+.IP \[bu] 2+Fix version-at listings with URL encoded keys (jxj)+.IP \[bu] 2+Fix storage class missing from \f[V]--s3-versions\f[R] listings (Dirk+Petersen)+.RE+.IP \[bu] 2+Seafile+.RS 2+.IP \[bu] 2+Fix corrupted uploads after a retried upload error (Nick Craig-Wood)+.RE+.IP \[bu] 2+SFTP+.RS 2+.IP \[bu] 2+Fix failed uploads leaving disk space in use on the server (Nick+Craig-Wood)+.RE+.IP \[bu] 2+Shade+.RS 2+.IP \[bu] 2+Fix potential crash in directory move (Nick Craig-Wood)+.RE+.IP \[bu] 2+Smb+.RS 2+.IP \[bu] 2+Save the user name in the config even if it matches the current user+(phatlc)+.RE+.IP \[bu] 2+WebDAV+.RS 2+.IP \[bu] 2+Fix duplicated listing entries after retried PROPFIND (ZRHann)+.IP \[bu] 2+Fix \[dq]XML syntax error\[dq] on paths which don\[aq]t exist with+ownCloud 10.16 (Nick Craig-Wood)+.RE+.SS v1.75.1 - 2026-09-04+.PP+See commits (https://github.com/rclone/rclone/compare/v1.75.0...v1.75.1)+.IP \[bu] 2+Security+.RS 2+.IP \[bu] 2+archive+.RS 2+.IP \[bu] 2+Fix zip slip path traversal in untrusted zip files GHSA-66hp-wgxq-6f5q+CVE-2026-88014 (Nick Craig-Wood)+.IP \[bu] 2+Hide any archive entry which escapes the directory being listed+GHSA-66hp-wgxq-6f5q (Nick Craig-Wood)+.IP \[bu] 2+Reject unsafe entry names when mounting squashfs images+GHSA-66hp-wgxq-6f5q (Nick Craig-Wood)+.IP \[bu] 2+Fix zip subdirectory root matching sibling directories+GHSA-66hp-wgxq-6f5q (Nick Craig-Wood)+.IP \[bu] 2+Fix zip entry named \[dq].\[dq] hiding every other file+GHSA-66hp-wgxq-6f5q (Nick Craig-Wood)+.IP \[bu] 2+Fix \[dq]directory not found\[dq] for archive paths containing+\[dq]./\[dq] or \[dq]//\[dq] GHSA-66hp-wgxq-6f5q (Nick Craig-Wood)+.RE+.IP \[bu] 2+build+.RS 2+.IP \[bu] 2+Fix multiple CVEs by upgrading to go1.26.6 (Nick Craig-Wood)+.RS 2+.IP \[bu] 2+CVE-2026-56860: net/url: quadratic complexity in resolvePath+.IP \[bu] 2+CVE-2026-56858: html/template: JavaScript regexp context tracking+.IP \[bu] 2+CVE-2026-56862: crypto/tls: limit handshake messages accepted+post-handshake+.IP \[bu] 2+CVE-2026-56853: net/http: apply ReadHeaderTimeout to unencrypted HTTP/2+check+.IP \[bu] 2+CVE-2026-56859: encoding/xml: recursion depth guard during decode+.IP \[bu] 2+CVE-2026-33818: encoding/asn1: enforce maximum recursion depth+.IP \[bu] 2+CVE-2026-46600: net: panic parsing an invalid SVCB or HTTPS RR in+dnsmessage+.IP \[bu] 2+CVE-2026-39821: net/http: reject ASCII-only Punycode-encoded labels in+idna+.RE+.IP \[bu] 2+Update golang.org/x/crypto to v0.56.0 to fix multiple CVEs (Nick+Craig-Wood)+.RS 2+.IP \[bu] 2+CVE-2026-56854: ssh: source-address critical option not enforced for+non-public-key auth callbacks+.IP \[bu] 2+CVE-2026-78662: ssh: a malicious peer could flood an undecided+channel\[aq]s incoming requests, deadlocking the connection+.IP \[bu] 2+CVE-2026-56855: ssh: a malicious peer could send crafted messages on an+established channel, deadlocking the connection+.RE+.IP \[bu] 2+Update golang.org/x/image to v0.45.0 to fix CVE-2026-46603 (Nick+Craig-Wood)+.RS 2+.IP \[bu] 2+CVE-2026-46603: excessive memory allocation during VP8L decoding+.RE+.RE+.IP \[bu] 2+fs: Confine directory listing entries that escape the root+GHSA-3vxh-3pcx-9m8q CVE-2026-93986 GHSA-38xv-hf3p-h7mq CVE-2026-88046+(Nick Craig-Wood)+.IP \[bu] 2+fshttp: Don\[aq]t send \f[V]--header\f[R] values to other hosts on+redirect GHSA-486v-q2wf-fp2r CVE-2026-88013 (Nick Craig-Wood)+.IP \[bu] 2+http: Don\[aq]t leak configured headers to other hosts or over plaintext+on redirect GHSA-486v-q2wf-fp2r CVE-2026-88013 (Nick Craig-Wood)+.IP \[bu] 2+lib/rest: Check HTTPS downgrades against the original request on+redirect GHSA-486v-q2wf-fp2r CVE-2026-88013 (Nick Craig-Wood)+.IP \[bu] 2+local+.RS 2+.IP \[bu] 2+Fix dir metadata escaping the root through a planted symlink+GHSA-f8g7-2xjc-7mfh CVE-2026-88016 (Nick Craig-Wood)+.IP \[bu] 2+Fix btime escaping the root via a planted symlink GHSA-f8g7-2xjc-7mfh+CVE-2026-88016 (Nick Craig-Wood)+.IP \[bu] 2+Fix panic on Range request past the end of a symlink GHSA-p6m2-r3w9-mpxw+CVE-2026-88015 (Nick Craig-Wood)+.RE+.IP \[bu] 2+serve docker+.RS 2+.IP \[bu] 2+Reject volume names that escape the base directory GHSA-p6vx-hf7p-98j6+CVE-2026-93987 (Nick Craig-Wood)+.IP \[bu] 2+Reject volume names resolving to the base directory itself+GHSA-p6vx-hf7p-98j6 CVE-2026-93987 (Nick Craig-Wood)+.IP \[bu] 2+Re-derive volume mountpoint from name when restoring state+GHSA-p6vx-hf7p-98j6 CVE-2026-93987 (Nick Craig-Wood)+.RE+.IP \[bu] 2+serve ftp: Fix auth-proxy sessions sharing credentials by username+GHSA-c476-6w5q-jw77 CVE-2026-88017 (Nick Craig-Wood)+.IP \[bu] 2+serve s3+.RS 2+.IP \[bu] 2+Fix memory exhaustion from client-declared multipart part size+GHSA-2p48-j3qc-rx9f CVE-2026-88045 (Nick Craig-Wood)+.IP \[bu] 2+Reject bogus multipart part sizes in the reorder buffer+GHSA-2p48-j3qc-rx9f (Nick Craig-Wood)+.IP \[bu] 2+Fix auth proxy accepting any request signed with an empty secret+GHSA-xwwr-4h3p-r22c CVE-2026-88018 (Nick Craig-Wood)+.RS 2+.IP \[bu] 2+\f[B]NB\f[R] the auth proxy protocol for \f[V]serve s3\f[R] has changed+- the proxy program is now given the access key ID as \f[V]user\f[R] and+must return the secret as \f[V]_secret_access_key\f[R]+.RE+.IP \[bu] 2+Fix each server accepting the \f[V]--auth-key\f[R] credentials of all+the others (Nick Craig-Wood)+.IP \[bu] 2+Fix misleading anonymous access log when using an auth proxy via rc+GHSA-p569-5gjg-9cmj CVE-2026-88044 (Nick Craig-Wood)+.RE+.IP \[bu] 2+serve sftp: Fix auth proxy configured via rc being silently ignored+GHSA-p569-5gjg-9cmj CVE-2026-88044 (Nick Craig-Wood)+.RE+.IP \[bu] 2+Bug Fixes+.RS 2+.IP \[bu] 2+accounting+.RS 2+.IP \[bu] 2+Fix memory leak on long-running rcd (nielash)+.IP \[bu] 2+Fix memory leak from stats groups on long-running rcd (nielash)+.IP \[bu] 2+Fix bwlimit burst overflow (Rayan Salhab)+.RE+.IP \[bu] 2+bisync+.RS 2+.IP \[bu] 2+Fix memory leak when running via the rc (nielash)+.IP \[bu] 2+Fix failed transfers of empty files being recorded as synced (Nick+Craig-Wood)+.RE+.IP \[bu] 2+build: Make go1.26 the minimum required version as needed by+golang.org/x/crypto v0.56.0 (Nick Craig-Wood)+.IP \[bu] 2+config: Redact env var config values in logs (Pastalikek65)+.IP \[bu] 2+doc fixes (Anton Karpov, CAOShurong, Dean Chen, Nick Craig-Wood,+Recoordinate, Rodrigo Rodrigues, Shantanav Mukherjee, shaurya)+.IP \[bu] 2+lib/batcher: Prevent commits racing shutdown (Loi Nguyen)+.IP \[bu] 2+lib/transform: Fix panic in \f[V]truncate_keep_extension\f[R] (VXNCXNX)+.IP \[bu] 2+multipart: Fix chunked uploads storing truncated objects when the source+ends early (Nick Craig-Wood)+.IP \[bu] 2+operations: Fix silent truncation of streaming uploads whose source ends+early (Nick Craig-Wood)+.IP \[bu] 2+serve+.RS 2+.IP \[bu] 2+Fix VFS instance leaks on server startup failures and shutdown (Hakan+İSMAİL)+.IP \[bu] 2+Pass the client IP address to the auth proxy (am-at-enrollvb)+.RE+.IP \[bu] 2+serve http: Prevent scrolling to the top on page reload (Sune Mølgaard)+.IP \[bu] 2+serve nfs: Fix EIO when creating symlinks with \f[V]--vfs-links\f[R]+(SillyZir)+.IP \[bu] 2+serve s3+.RS 2+.IP \[bu] 2+Fix failed uploads deleting or corrupting the object at the key (Nick+Craig-Wood)+.IP \[bu] 2+Fix crash when a multipart upload is aborted while a part is uploading+(Nick Craig-Wood)+.IP \[bu] 2+Fix modtime not being set when only mtime metadata is supplied on PUT+(Nick Craig-Wood)+.IP \[bu] 2+Upload all multipart uploads via the VFS so they obey+\f[V]--bwlimit\f[R] and show in stats (Nick Craig-Wood)+.IP \[bu] 2+Reserve the \f[V].rclone_temp_\f[R] prefix for temporary objects (Nick+Craig-Wood)+.IP \[bu] 2+Clean up abandoned multipart uploads after \f[V]--multipart-expiry\f[R]+(Nick Craig-Wood)+.RE+.IP \[bu] 2+vfscache+.RS 2+.IP \[bu] 2+Fix reader deadlock when the item size drops below the read offset+(Dave)+.IP \[bu] 2+Fix log message growing without bound on repeated write errors (Vijay+Misal)+.RE+.IP \[bu] 2+walk: Stop directory traversal when the context is cancelled (Rahman+Yilmaz)+.RE.IP \[bu] 2-Env Var: RCLONE_LOCAL_NO_PREALLOCATE+VFS+.RS 2.IP \[bu] 2-Type: bool+Synchronize poll updates with shutdown (Loi Nguyen).IP \[bu] 2-Default: false-.SS --local-no-sparse-.PP-Disable sparse files for multi-thread downloads.-.PP-On Windows platforms rclone will make sparse files when doing-multi-thread downloads.-This avoids long pauses on large files where the OS zeros the file.-However sparse files may be undesirable as they cause disk fragmentation-and can be slow to work with.-.PP-Properties:+Make poll shutdown lifecycle deterministic (Loi Nguyen)+.RE.IP \[bu] 2-Config: no_sparse+Crypt+.RS 2.IP \[bu] 2-Env Var: RCLONE_LOCAL_NO_SPARSE+Fix hash mismatches with \f[V]no_data_encryption\f[R] on backends which+check upload hashes (Nick Craig-Wood).IP \[bu] 2-Type: bool+Fix directory names which look like versioned file names (TowyTowy).IP \[bu] 2-Default: false-.SS --local-no-set-modtime-.PP-Disable setting modtime.-.PP-Normally rclone updates modification time of files after they are done-uploading.-This can cause permissions issues on Linux platforms when the user-rclone is running as does not own the file uploaded, such as when-copying to a CIFS mount owned by another user.-If this option is enabled, rclone will no longer update the modtime-after copying a file.-.PP-Properties:+Warn about directories with legacy version-like encrypted names (Nick+Craig-Wood)+.RE.IP \[bu] 2-Config: no_set_modtime+Azure Blob+.RS 2.IP \[bu] 2-Env Var: RCLONE_LOCAL_NO_SET_MODTIME+Fix Entra ID server-side copy source authentication (Edward Klesel).IP \[bu] 2-Type: bool+Fix spurious vfs cache corruption errors during chunked reads (Nick+Craig-Wood)+.RE.IP \[bu] 2-Default: false-.SS --local-metadata-restore-special-bits-.PP-Restore the setuid, setgid and sticky bits from metadata.-.PP-When restoring metadata with --metadata rclone applies the-\[dq]mode\[dq] from the source.-By default rclone applies only the permission bits and strips the-setuid, setgid and sticky bits.-.PP-The \[dq]mode\[dq] comes from the source remote which may not be-trusted.-Restoring a setuid or setgid bit onto freshly written, source-controlled-content can plant a setuid binary, which is dangerous in particular when-restoring from an untrusted source while running as root.-For this reason these bits are not restored by default.-.PP-If you trust the source and want the setuid, setgid and sticky bits-restored - for example when restoring a system backup made by rclone --set this flag.-.PP-Properties:+Azurefiles+.RS 2.IP \[bu] 2-Config: metadata_restore_special_bits+Fix zero padded files being created when the source ends early (Nick+Craig-Wood)+.RE.IP \[bu] 2-Env Var: RCLONE_LOCAL_METADATA_RESTORE_SPECIAL_BITS+Box+.RS 2.IP \[bu] 2-Type: bool+Fix truncated files being uploaded successfully when the source ends+early (Rohit Behera)+.RE.IP \[bu] 2-Default: false-.SS --local-fatal-if-no-space-.PP-Make out-of-space errors fatal during transfers.-.PP-When enabled, an ENOSPC error during a write returns a fatal error so-that rclone aborts rather than retrying the operation.-Useful for backup scripts that should halt loudly on a full disk rather-than spin retrying.-.PP-Properties:+Compress+.RS 2.IP \[bu] 2-Config: fatal_if_no_space+Fix corrupted objects being created when the source ends early (Nick+Craig-Wood)+.RE.IP \[bu] 2-Env Var: RCLONE_LOCAL_FATAL_IF_NO_SPACE+Drive+.RS 2.IP \[bu] 2-Type: bool+Don\[aq]t list trashed files when removing a directory into the trash+(alliasgher)+.RE.IP \[bu] 2-Default: false-.SS --local-time-type-.PP-Set what kind of time is returned.-.PP-Normally rclone does all operations on the mtime or Modification time.-.PP-If you set this flag then rclone will return the Modified time as-whatever you set here.-So if you use \[dq]rclone lsl --local-time-type ctime\[dq] then you will-see ctimes in the listing.-.PP-If the OS doesn\[aq]t support returning the time_type specified then-rclone will silently replace it with the modification time which all-OSes support.+Dropbox+.RS 2.IP \[bu] 2-mtime is supported by all OSes+Preserve Paper export paths on lookup (Loi Nguyen).IP \[bu] 2-atime is supported on all OSes except: plan9, js+Fix context cancellation (e.g.+\f[V]--max-duration\f[R] limit) not stopping in-flight requests+(debaditya).IP \[bu] 2-btime is only supported on: Windows, macOS, freebsd, netbsd+Fix chunked uploads of truncated files never finishing (Nick Craig-Wood).IP \[bu] 2-ctime is supported on all Oses except: Windows, plan9, js-.PP-Note that setting the time will still set the modified time so this is-only useful for reading.-.PP-Properties:+Don\[aq]t retry chunked upload requests when the upload has been+cancelled (Nick Craig-Wood).IP \[bu] 2-Config: time_type+Decode received shared-file names (Sanjay Kanth A).IP \[bu] 2-Env Var: RCLONE_LOCAL_TIME_TYPE+Fix ChangeNotify when the root\[aq]s case differs from Dropbox\[aq]s+(Loi Nguyen)+.RE.IP \[bu] 2-Type: mtime|atime|btime|ctime+Filelu+.RS 2.IP \[bu] 2-Default: mtime+Fix truncated files being uploaded successfully when the source ends+early (Nick Craig-Wood).IP \[bu] 2-Examples:+Fix duplicate root path during multipart folder creation (kingston125)+.RE+.IP \[bu] 2+Huaweidrive.RS 2.IP \[bu] 2-\[dq]mtime\[dq]+Fix truncated files being uploaded successfully when the source ends+early (Rohit Behera)+.RE+.IP \[bu] 2+Iclouddrive.RS 2.IP \[bu] 2-The last modification time.+Fix uploads into an app container failing with 412 (Christian De Santis).RE.IP \[bu] 2-\[dq]atime\[dq]+Internetarchive.RS 2.IP \[bu] 2-The last access time.+Fix corrupted files being created when the source ends early (Nick+Craig-Wood).RE.IP \[bu] 2-\[dq]btime\[dq]+Internxt.RS 2.IP \[bu] 2-The creation time.+Persist rotated token returned by the user info call (0rangeSeaW0lf).RE.IP \[bu] 2-\[dq]ctime\[dq]+Onedrive.RS 2.IP \[bu] 2-The last status change time.+Fix 403 Forbidden for configuration personal onedrive (machsix)+.IP \[bu] 2+Fall back to manual drive ID entry when drive listing fails (SillyZir)+.IP \[bu] 2+Don\[aq]t retry multipart upload chunk on 404 (upload session not found)+(water).RE+.IP \[bu] 2+Overview+.RS 2+.IP \[bu] 2+Fix \[dq]internal error: no overview data found\[dq] on 32 bit+architectures (Nick Craig-Wood).RE-.SS --local-hashes-.PP-Comma separated list of supported checksum types.-.PP-Properties:.IP \[bu] 2-Config: hashes+Pikpak+.RS 2.IP \[bu] 2-Env Var: RCLONE_LOCAL_HASHES+Fix truncated files being created when the source ends early (Nick+Craig-Wood).IP \[bu] 2-Type: CommaSepList+Fix truncated single part uploads reported as ok when source ends early+(Nick Craig-Wood)+.RE.IP \[bu] 2-Default:-.SS --local-encoding-.PP-The encoding for the backend.-.PP-See the encoding section in the-overview (https://rclone.org/overview/#encoding) for more info.-.PP-Properties:+Protondrive+.RS 2.IP \[bu] 2-Config: encoding+Fix files uploaded with v1.75.0 not being readable in the Proton apps+(Nick Craig-Wood).IP \[bu] 2-Env Var: RCLONE_LOCAL_ENCODING+Fix corrupted uploads after a retried upload error (Nick Craig-Wood)+.RE.IP \[bu] 2-Type: Encoding+Quatrix+.RS 2.IP \[bu] 2-Default: Slash,Dot-.SS --local-description-.PP-Description of the remote.-.PP-Properties:+Fix chunk upload retries and fix memory leak (Nick Craig-Wood)+.RE.IP \[bu] 2-Config: description+S3+.RS 2.IP \[bu] 2-Env Var: RCLONE_LOCAL_DESCRIPTION+Update Mega endpoints (Nick Craig-Wood).IP \[bu] 2-Type: string+Treat UploadPart success without ETag as retryable error (CAOShurong).IP \[bu] 2-Required: false-.SS Metadata-.PP-Depending on which OS is in use the local backend may return only some-of the system metadata.-Setting system metadata is supported on all OSes but setting user-metadata is only supported on linux, freebsd, netbsd, macOS and Solaris.-It is \f[B]not\f[R] supported on Windows yet (see-pkg/attrs#47 (https://github.com/pkg/xattr/issues/47)).-.PP-User metadata is stored as extended attributes (which may not be-supported by all file systems) under the \[dq]user.*\[dq] prefix.-.PP-Metadata is supported on files and directories.-.PP-When restoring metadata with \f[V]--metadata\f[R] rclone applies the-\[dq]mode\[dq], \[dq]uid\[dq] and \[dq]gid\[dq] from the source.-These come from the source remote which may not be trusted, so restoring-metadata as root from an untrusted source can change file ownership and-is not recommended.-The setuid, setgid and sticky bits are not restored by default - see the-\f[V]--local-metadata-restore-special-bits\f[R] flag.-.PP-Here are the possible system metadata items for the local backend.-.PP-.TS-tab(@);-lw(11.1n) lw(11.1n) lw(11.1n) lw(16.6n) lw(20.3n).-T{-Name-T}@T{-Help-T}@T{-Type-T}@T{-Example-T}@T{-Read Only-T}-_-T{-atime-T}@T{-Time of last access-T}@T{-RFC 3339-T}@T{-2006-01-02T15:04:05.999999999Z07:00-T}@T{-N-T}-T{-btime-T}@T{-Time of file birth (creation)-T}@T{-RFC 3339-T}@T{-2006-01-02T15:04:05.999999999Z07:00-T}@T{-N-T}-T{-gid-T}@T{-Group ID of owner-T}@T{-decimal number-T}@T{-500-T}@T{-N-T}-T{-mode-T}@T{-File type and mode-T}@T{-octal, unix style-T}@T{-0100664-T}@T{-N-T}-T{-mtime-T}@T{-Time of last modification-T}@T{-RFC 3339-T}@T{-2006-01-02T15:04:05.999999999Z07:00-T}@T{-N-T}-T{-rdev-T}@T{-Device ID (if special file)-T}@T{-hexadecimal-T}@T{-1abc-T}@T{-N-T}-T{-uid-T}@T{-User ID of owner-T}@T{-decimal number-T}@T{-500-T}@T{-N-T}-.TE-.PP-See the metadata (https://rclone.org/docs/#metadata) docs for more info.-.SS Backend commands-.PP-Here are the commands specific to the local backend.-.PP-Run them with:-.IP-.nf-\f[C]-rclone backend COMMAND remote:-\f[R]-.fi-.PP-The help below will explain what arguments each command takes.-.PP-See the backend (https://rclone.org/commands/rclone_backend/) command-for more info on how to pass options and arguments.-.PP-These can be run on a running backend using the rc command-backend/command (https://rclone.org/rc/#backend-command).-.SS noop-.PP-A null operation for testing backend commands.-.IP-.nf-\f[C]-rclone backend noop remote: [options] [<arguments>+]-\f[R]-.fi-.PP-This is a test command which has some options you can try to change the-output.-.PP-Options:+Fix server side copy failing with \f[V]--s3-no-head-object\f[R] (Anatoly+Tarnavsky)+.RE.IP \[bu] 2-\[dq]echo\[dq]: Echo the input arguments.+Sia+.RS 2.IP \[bu] 2-\[dq]error\[dq]: Return an error based on option value.-.SH Changelog+Fix corrupted files being created when the source ends early (Nick+Craig-Wood)+.RE+.IP \[bu] 2+Smb+.RS 2+.IP \[bu] 2+Reuse the upload connection for SetModTime (alliasgher)+.RE+.IP \[bu] 2+WebDAV+.RS 2+.IP \[bu] 2+Fix SetModTime failing and hashes missing on Nextcloud (Nick Craig-Wood)+.RE+.IP \[bu] 2+Yandex+.RS 2+.IP \[bu] 2+Fix truncated files being uploaded successfully when the source ends+early (Rohit Behera)+.RE.SS v1.75.0 - 2026-07-31.PPSee commits (https://github.com/rclone/rclone/compare/v1.74.0...v1.75.0)@@ -91570,66 +97914,67 @@ Security.RS 2.IP \[bu] 2archive: Don\[aq]t crash on malformed squashfs images-GHSA-6jcg-q3wp-x2f4 CVE-PENDING (Nick Craig-Wood)+GHSA-6jcg-q3wp-x2f4 CVE-2026-79775 (Nick Craig-Wood).IP \[bu] 2ftp: Fix ftp command injection when encoding doesn\[aq]t include CRLF-GHSA-8c48-q9wj-3w37 CVE-PENDING (Nick Craig-Wood)+GHSA-8c48-q9wj-3w37 CVE-2026-71311 (Nick Craig-Wood).IP \[bu] 2lib/http: Use TLS on all \f[V]--addr\f[R] listeners when\f[V]--cert\f[R] and \f[V]--key\f[R] are set GHSA-mfvx-7rcj-9m5g (NickCraig-Wood).IP \[bu] 2lib/proxy: Fix unbounded HTTP CONNECT headers causing OOM-GHSA-xhf4-832v-7xcr CVE-PENDING (Nick Craig-Wood)+GHSA-xhf4-832v-7xcr CVE-2026-71310 (Nick Craig-Wood).IP \[bu] 2local: Stop source file names escaping the destination directory-GHSA-7p4m-qxvv-g567 CVE-PENDING (Nick Craig-Wood)+GHSA-7p4m-qxvv-g567 CVE-2026-71313 (Nick Craig-Wood).IP \[bu] 2rc.RS 2.IP \[bu] 2Don\[aq]t expose pprof debug handlers on an unauthenticated server-GHSA-mfvx-7rcj-9m5g CVE-PENDING (Nick Craig-Wood)+GHSA-mfvx-7rcj-9m5g CVE-2026-79776 (Nick Craig-Wood).IP \[bu] 2Require authentication to list the remotes with \f[V]--rc-serve\f[R]GHSA-mfvx-7rcj-9m5g (Nick Craig-Wood).IP \[bu] 2-Fix leaking stack traces on panics GHSA-gwfq-86j8-7qhv (Nick Craig-Wood)+Fix leaking stack traces on panics GHSA-gwfq-86j8-7qhv CVE-2026-79777+(Nick Craig-Wood).RE.IP \[bu] 2s3.RS 2.IP \[bu] 2Fix redirect credential leaks, reject HTTPS->HTTP and strip secrets-GHSA-8mxv-9xhp-86h4 (Nick Craig-Wood)+GHSA-8mxv-9xhp-86h4 CVE-2026-79780 (Nick Craig-Wood).IP \[bu] 2Strip S3 Express session token on cross-host redirects-GHSA-8mxv-9xhp-86h4 (Nick Craig-Wood)+GHSA-8mxv-9xhp-86h4 CVE-2026-79780 (Nick Craig-Wood).RE.IP \[bu] 2serve ftp: Use constant time comparison for password checkGHSA-mfvx-7rcj-9m5g (Nick Craig-Wood).IP \[bu] 2serve restic: Fix path traversal above the served directory-GHSA-45pq-889g-fcgh CVE-PENDING (Nick Craig-Wood)+GHSA-45pq-889g-fcgh CVE-2026-71309 (Nick Craig-Wood).IP \[bu] 2serve sftp: Don\[aq]t crash the whole server on a bad request-GHSA-6jcg-q3wp-x2f4 (Nick Craig-Wood)+GHSA-6jcg-q3wp-x2f4 CVE-2026-79775 (Nick Craig-Wood).IP \[bu] 2sftp: Fix command injection via crafted filenames on PowerShell remotes-GHSA-2m8m-jhrm-w6j2 CVE-PENDING (Nick Craig-Wood)+GHSA-2m8m-jhrm-w6j2 CVE-2026-71312 (Nick Craig-Wood).IP \[bu] 2vfs: Don\[aq]t crash the process if a backend panics on a background-goroutine GHSA-6jcg-q3wp-x2f4 (Nick Craig-Wood)+goroutine GHSA-6jcg-q3wp-x2f4 CVE-2026-79775 (Nick Craig-Wood).IP \[bu] 2webdav.RS 2.IP \[bu] 2Fix HTTPS to HTTP redirects leaking credentials GHSA-h4mf-4v27-hggj-(Nick Craig-Wood)+CVE-2026-79779 (Nick Craig-Wood).IP \[bu] 2-Tus: fix potential nil pointer crash GHSA-3x6r-wxxg-53vv (Nick-Craig-Wood)+Tus: fix potential nil pointer crash GHSA-3x6r-wxxg-53vv CVE-2026-79778+(Nick Craig-Wood).RE.IP \[bu] 2Update \f[V]google.golang.org/grpc\f[R] to fix multiple security@@ -92304,7 +98649,7 @@ serve s3Fix spurious 404 on HEAD/GET during VFS writeback (max).IP \[bu] 2Fix path traversal letting clients see files in the root-GHSA-8v25-v8p6-qf7v (Nick Craig-Wood)+GHSA-8v25-v8p6-qf7v CVE-2026-79781 (Nick Craig-Wood).RE.IP \[bu] 2serve webdav: Fix MOVE overwrite failing without Overwrite header@@ -92328,7 +98673,7 @@ Stop \f[V]--links\f[R] symlinks escaping the destination directoryCVE-2026-54572 (Nick Craig-Wood).IP \[bu] 2Don\[aq]t restore setuid/setgid/sticky bits from metadata by default-GHSA-945v-v9p3-v5xw (Nick Craig-Wood)+GHSA-945v-v9p3-v5xw CVE-2026-79783 (Nick Craig-Wood).RE.IP \[bu] 2Drive@@ -109997,7 +116342,7 @@ the config file.The FTP and SFTP backends have their own \f[V]http_proxy\f[R] settingsto support an HTTP CONNECT proxy (--ftp-http-proxy (https://rclone.org/ftp/#ftp-http-proxy) and---sftp-http-proxy (https://rclone.org/ftp/#sftp-http-proxy) )+--sftp-http-proxy (https://rclone.org/sftp/#sftp-http-proxy) ).SS Rclone gives x509 SSL root certificates error.PP\f[V]x509: failed to load system roots and no roots provided\f[R] means@@ -112635,6 +118980,277 @@ Socialpranker <[email protected]>phatlc <[email protected]>.IP \[bu] 2Punya Jain <[email protected]>+.IP \[bu] 2+acoeur <[email protected]>+.IP \[bu] 2+SillyZir <[email protected]>+.IP \[bu] 2+am-at-enrollvb <[email protected]>+.IP \[bu] 2+Loi Nguyen <[email protected]>+.IP \[bu] 2+debaditya <[email protected]>+.IP \[bu] 2+alliasgher <[email protected]>+.IP \[bu] 2+Anton Karpov <[email protected]>+.IP \[bu] 2+Edward Klesel <[email protected]>+.IP \[bu] 2+Christian De Santis+.IP \[bu] 2+Rodrigo Rodrigues <[email protected]>+.IP \[bu] 2+Recoordinate <[email protected]>+.IP \[bu] 2+Jiacheng Xu <[email protected]>+.IP \[bu] 2+Teddy Tennant <[email protected]>+.IP \[bu] 2+Dean Chen <[email protected]>+.IP \[bu] 2+Pastalikek65 <[email protected]>+.IP \[bu] 2+Morax <[email protected]>+.IP \[bu] 2+Shurong Cao <[email protected]>+.IP \[bu] 2+Dave <[email protected]>+.IP \[bu] 2+max <[email protected]>+.IP \[bu] 2+Hakan İSMAİL <[email protected]>+.IP \[bu] 2+Shantanav Mukherjee <[email protected]>+.IP \[bu] 2+VXNCXNX <[email protected]>+.IP \[bu] 2+Rohit Behera <[email protected]>+.IP \[bu] 2+Rahman Yilmaz <[email protected]>+.IP \[bu] 2+Sanjay Kanth A <[email protected]>+.IP \[bu] 2+machsix <[email protected]>+.IP \[bu] 2+CAOShurong <[email protected]>+.IP \[bu] 2+shaurya <[email protected]>+.IP \[bu] 2+water <[email protected]>+.IP \[bu] 2+Rayan Salhab <[email protected]>+.IP \[bu] 2+Vijay Misal <[email protected]>+.IP \[bu] 2+Sune Mølgaard <[email protected]>+.IP \[bu] 2+Anatoly Tarnavsky <[email protected]>+.IP \[bu] 2+cyphercodes <[email protected]>+.IP \[bu] 2+shaurya <[email protected]>+.IP \[bu] 2+no-hup <[email protected]>+.IP \[bu] 2+Dhevenddra <[email protected]>+.IP \[bu] 2+Splainte <[email protected]>+.IP \[bu] 2+PSR94 <[email protected]>+.IP \[bu] 2+Shane McCarron <[email protected]>+.IP \[bu] 2+Dhevenddra K G <[email protected]>+.IP \[bu] 2+Vaibhav Mashal <[email protected]>+.IP \[bu] 2+Murat Topcu <[email protected]>+.IP \[bu] 2+tomaszni <[email protected]>+.IP \[bu] 2+eliotee <[email protected]>+.IP \[bu] 2+Aditya <[email protected]>+.IP \[bu] 2+Nicholas Velten <[email protected]>+.IP \[bu] 2+youdie006 <[email protected]>+.IP \[bu] 2+subomi <[email protected]>+.IP \[bu] 2+Can Arslan <[email protected]>+.IP \[bu] 2+Can Arslan <[email protected]>+.IP \[bu] 2+ZRHann <[email protected]>+.IP \[bu] 2+enkvadrat <[email protected]>+.IP \[bu] 2+Vladimir Babin <[email protected]>+.IP \[bu] 2+Eugene <[email protected]>+.IP \[bu] 2+Kunpeng Xie <[email protected]>+.IP \[bu] 2+foecmke <[email protected]>+.IP \[bu] 2+Vasek Sraier <[email protected]>+.IP \[bu] 2+Kalin Stoyanov <[email protected]>+.IP \[bu] 2+NytePlus <[email protected]>+.IP \[bu] 2+Jeremy Schoemaker <[email protected]>+.IP \[bu] 2+solunolab <[email protected]>+.IP \[bu] 2+Mattias Michaux <[email protected]>+.IP \[bu] 2+jxj <[email protected]>+.IP \[bu] 2+Harsh Raj Singhania+.IP \[bu] 2+Roland <[email protected]>+.IP \[bu] 2+interested.tortoise <[email protected]>+.IP \[bu] 2+hsdfat <[email protected]>+.IP \[bu] 2+Dirk Petersen <[email protected]>+.IP \[bu] 2+TastyHeadphones <[email protected]>+.IP \[bu] 2+Neil Cawse <[email protected]>+.IP \[bu] 2+Lev Devaev <[email protected]>+.IP \[bu] 2+GhostCoder6969 <[email protected]>+.IP \[bu] 2+bounty-agent <[email protected]>+.IP \[bu] 2+mygrexit <[email protected]>+.IP \[bu] 2+Kaya <[email protected]>+.IP \[bu] 2+somaz <[email protected]>+.IP \[bu] 2+xiaozou-wine <[email protected]>+.IP \[bu] 2+Roman Butor <[email protected]>+.IP \[bu] 2+kaii9 <[email protected]>+.IP \[bu] 2+Stochastic <[email protected]>+.IP \[bu] 2+James Truitt <[email protected]>+.IP \[bu] 2+firatkaya <[email protected]>+.IP \[bu] 2+kevin9327 <[email protected]>+.IP \[bu] 2+drakeo338 <[email protected]>+.IP \[bu] 2+Alessio Margelli <[email protected]>+.IP \[bu] 2+Ludovic Bondon <[email protected]>+.IP \[bu] 2+fmterrors <[email protected]>+.IP \[bu] 2+GingerAdonis <[email protected]>+.IP \[bu] 2+Satvik Mishra <[email protected]>+.IP \[bu] 2+wjc <[email protected]>+.IP \[bu] 2+Maksym Tymoshyk <[email protected]>+.IP \[bu] 2+Aniket <[email protected]>+.IP \[bu] 2+Oshino Penguin <[email protected]>+.IP \[bu] 2+Yousef Issa <[email protected]>+.IP \[bu] 2+ghwany <[email protected]>+.IP \[bu] 2+DEVN <[email protected]>+.IP \[bu] 2+47th <[email protected]>+.IP \[bu] 2+Sash <[email protected]>+.IP \[bu] 2+Shreyansh Jain <[email protected]>+.IP \[bu] 2+Biagio Peccerillo <[email protected]>+.IP \[bu] 2+gomitrah <[email protected]>+.IP \[bu] 2+Jim Counter <[email protected]>+.IP \[bu] 2+Dirk Einecke <[email protected]>+.IP \[bu] 2+ASRodrigo1 <[email protected]>+.IP \[bu] 2+chaogeshuodiannao <[email protected]>+.IP \[bu] 2+David Legrand <[email protected]>+.IP \[bu] 2+Azhar <[email protected]>+.IP \[bu] 2+Daniel Bae <[email protected]>+.IP \[bu] 2+frodleTian <[email protected]>+.IP \[bu] 2+Cestercian <[email protected]>+.IP \[bu] 2+Ayush Sharma <[email protected]>+.IP \[bu] 2+Dirk Petersen <[email protected]>+.IP \[bu] 2+Codex <[email protected]>+.IP \[bu] 2+bunnysayzz <[email protected]>+.IP \[bu] 2+hahahahahayesyeseys <[email protected]>+.IP \[bu] 2+qodelicious <[email protected]>+.IP \[bu] 2+fixfon <[email protected]>+.IP \[bu] 2+berce <[email protected]>+.IP \[bu] 2+Fred <[email protected]>+.IP \[bu] 2+Florian Josef Altendorfer <[email protected]>+.IP \[bu] 2+ashwinibailangadi <[email protected]>+.IP \[bu] 2+ilya lavrenov <[email protected]>+.IP \[bu] 2+Soujas K <[email protected]>+.IP \[bu] 2+Robert Kubiś <[email protected]>+.IP \[bu] 2+b84330808 <[email protected]>+.IP \[bu] 2+Guillaume MALLERET <[email protected]>+.IP \[bu] 2+Louis <[email protected]>+.IP \[bu] 2+唐僧洗发用飘柔 <[email protected]>+.IP \[bu] 2+Seth Westphal <[email protected]>+.IP \[bu] 2+ZainnQureshii <[email protected]>+.IP \[bu] 2+Rishi Raj Jain <[email protected]>+.IP \[bu] 2+Enbiya Göral <[email protected]>+.IP \[bu] 2+Florian Josef Altendorfer.SH Contact the rclone project.SS Forum.PP