Real-time monitoring dashboard for Proxmox VE, PBS, Docker, Kubernetes, TrueNAS and vSphere. Self-hosted, with alerts for failed backups, full disks, unhealthy containers and offline nodes.
Go aialertsdashboarddockerdocker-monitoringgohomelabinfrastructure-monitoringkubernetesmonitoringmspproxmoxproxmox-backup-serverproxmox-mail-gatewayproxmox-veself-hostedtruenastypescriptvspherewebhooks
Prepare v6.6.0-beta.2
pulse-triage[bot]dae9cb86bdec
diff --git a/VERSION b/VERSIONindex 6b546cf64d..056498ecbd 100644--- a/VERSION+++ b/VERSION@@ -1 +1 @@-6.6.0-beta.1+6.6.0-beta.2diff --git a/deploy/helm/pulse/Chart.yaml b/deploy/helm/pulse/Chart.yamlindex 84946d166f..4c43ab0287 100644--- a/deploy/helm/pulse/Chart.yaml+++ b/deploy/helm/pulse/Chart.yaml@@ -2,9 +2,9 @@ apiVersion: v2name: pulsedescription: Helm chart for deploying the Pulse hub and optional Docker, Kubernetes, or OpenShift monitoring agent.type: application-version: 6.6.0-beta.1-appVersion: "6.6.0-beta.1"-icon: https://raw.githubusercontent.com/rcourtman/Pulse/v6.6.0-beta.1/docs/images/pulse-logo.svg+version: 6.6.0-beta.2+appVersion: "6.6.0-beta.2"+icon: https://raw.githubusercontent.com/rcourtman/Pulse/v6.6.0-beta.2/docs/images/pulse-logo.svgkeywords:- monitoring- proxmox@@ -32,7 +32,7 @@ annotations:description: Smoke tests with kind cluster deploymentartifacthub.io/links: |- name: Documentation- url: https://github.com/rcourtman/Pulse/blob/v6.6.0-beta.1/docs/KUBERNETES.md+ url: https://github.com/rcourtman/Pulse/blob/v6.6.0-beta.2/docs/KUBERNETES.md- name: Supporturl: https://github.com/rcourtman/Pulse/discussionsartifacthub.io/maintainers: |diff --git a/deploy/helm/pulse/README.md b/deploy/helm/pulse/README.mdindex 090ce2156e..3b6db1b77b 100644--- a/deploy/helm/pulse/README.md+++ b/deploy/helm/pulse/README.md@@ -1,6 +1,6 @@# pulse-  +  Helm chart for deploying the Pulse hub and optional Docker, Kubernetes, or OpenShift monitoring agent.diff --git a/docker-compose.yml b/docker-compose.ymlindex 46d87c7386..0bb8a8322a 100644--- a/docker-compose.yml+++ b/docker-compose.yml@@ -2,7 +2,7 @@ version: '3.8'services:pulse:- image: ${PULSE_IMAGE:-rcourtman/pulse:6.6.0-beta.1}+ image: ${PULSE_IMAGE:-rcourtman/pulse:6.6.0-beta.2}container_name: pulserestart: unless-stoppedlogging:diff --git a/docs/RELEASE_NOTES.md b/docs/RELEASE_NOTES.mdindex 93f3125bdb..b2cebed2a9 100644--- a/docs/RELEASE_NOTES.md+++ b/docs/RELEASE_NOTES.md@@ -7,8 +7,8 @@ For historical v4 notes that previously lived in this repo, see:`docs/releases/RELEASE_NOTES_v4.md`For the current v6 beta packet, see:-- `docs/releases/RELEASE_NOTES_v6.6.0-beta.1.md`-- `docs/releases/V6_CHANGELOG_v6.6.0-beta.1.md`+- `docs/releases/RELEASE_NOTES_v6.6.0-beta.2.md`+- `docs/releases/V6_CHANGELOG_v6.6.0-beta.2.md`For the current stable v6 packet, see:- `docs/releases/RELEASE_NOTES_v6.5.0.md`diff --git a/docs/UPGRADE_v6.md b/docs/UPGRADE_v6.mdindex cbbccbcc56..3a244366ab 100644--- a/docs/UPGRADE_v6.md+++ b/docs/UPGRADE_v6.md@@ -4,8 +4,8 @@ This guide covers practical upgrade steps for existing Pulse installs moving toFor the current v6 beta packet, see:-- `docs/releases/RELEASE_NOTES_v6.6.0-beta.1.md`-- `docs/releases/V6_CHANGELOG_v6.6.0-beta.1.md`+- `docs/releases/RELEASE_NOTES_v6.6.0-beta.2.md`+- `docs/releases/V6_CHANGELOG_v6.6.0-beta.2.md`For the current stable v6 packet, see:diff --git a/docs/releases/RELEASE_NOTES_v6.6.0-beta.2.md b/docs/releases/RELEASE_NOTES_v6.6.0-beta.2.mdnew file mode 100644index 0000000000..3fdd997ae2--- /dev/null+++ b/docs/releases/RELEASE_NOTES_v6.6.0-beta.2.md@@ -0,0 +1,15 @@+# Pulse v6.6.0-beta.2 Release Notes++This beta follows `v6.6.0-beta.1` on the way to stable `v6.6.0`. Stable `v6.5.0` remains the rollback target.++This beta carries 322 changes merged since stable `v6.5.0`.++## What's improved++- **Fixes and improvements since v6.5.0.** The changelog for this beta lists every change it carries.++## Before you upgrade++- Windows Unified Agent binaries are not Authenticode-signed while SignPath remains unavailable and may show an Unknown Publisher warning. Verify downloads with the published checksums and detached signatures.+- Pulse Mobile remains compatible. This candidate does not require a companion mobile release.+- The rollback target is stable `v6.5.0`. On systemd and Proxmox LXC installs, use this command only when `/bin/update` was installed by the Pulse server installer: `sudo /bin/update --version v6.5.0` to return to the previous stable release. For Docker Compose, pin `rcourtman/pulse:6.5.0` and recreate the container.diff --git a/docs/releases/RELEASE_NOTES_v6.6.0-beta.2.md.visuals.json b/docs/releases/RELEASE_NOTES_v6.6.0-beta.2.md.visuals.jsonnew file mode 100644index 0000000000..dc2038ee88--- /dev/null+++ b/docs/releases/RELEASE_NOTES_v6.6.0-beta.2.md.visuals.json@@ -0,0 +1,5 @@+{+ "schema_version": 1,+ "captures": [],+ "decision": "Zero captures: the notes describe the changes in text."+}diff --git a/docs/releases/V6_CHANGELOG_v6.6.0-beta.2.md b/docs/releases/V6_CHANGELOG_v6.6.0-beta.2.mdnew file mode 100644index 0000000000..1f87567b75--- /dev/null+++ b/docs/releases/V6_CHANGELOG_v6.6.0-beta.2.md@@ -0,0 +1,339 @@+# Pulse v6.6.0-beta.2++This changelog describes the changes since `v6.6.0-beta.1`. Stable `v6.5.0` remains the rollback target.++## Changes since v6.5.0++- Teach the pre-release metadata test about a second beta (#2869)+- Explain backup freshness without changing alert policy (#2868)+- Retry registry reads when verifying release container images (#2866)+- Retire Patrol's Safe auto-fix and Autopilot modes (#2863)+- Retire the obsolete governed worktree workflow (#2865)+- Share demo generation and stabilize workload parent groups (#2861)+- Retry the release listing so a GitHub 504 cannot fail a release (#2862)+- Keep guest readings and recovery guidance trustworthy (#2825)+- Release v6.6.0-beta.1 (#2860)+- Treat no matching unit files as an empty service inventory (#2859)+- Remove AI Discovery (#2848)+- Let backup health counts filter Coverage to their workloads (#2857)+- Keep phone alert rows to one action row (#2856)+- Read staged governance inputs from the Git index (#2852)+- Release v6.6.0-beta.1 (#2853)+- Advance the Go toolchain to 1.26.9 and x/net to v0.60.0 (#2849)+- Keep the phone backup indicator from taking over guest rows (#2850)+- Wait for reporting hook loads instead of counting ticks (#2847)+- Release v6.6.0-beta.1 (#2846)+- Release v6.6.0-beta.1 (#2845)+- Let the release metadata tests accept beta versions (#2844)+- Update source-map-js in the Cloud portal lockfile to 1.2.2 (#2843)+- Show diagnostics callers only the API tokens they may manage (#2841)+- docs: describe the token authority rules that now ship (#2842)+- Require ai:execute for AI narration of reports by token (#2836)+- Keep token management inside the caller's organizations (#2837)+- Make alert threshold Off stick for Docker, PMG and backups (#2839)+- Require instance-wide authority to rotate credentials by token (#2838)+- Require an admin for non-token agent uninstall (#2833)+- docs: state the scopes for token regeneration and resource links (#2834)+- Admit attention lifecycle POSTs on the mutation scope (#2832)+- docs: stop labelling free Patrol routes as Pro (#2831)+- Stop token callers widening or bypassing their scopes (#2830)+- docs: correct API reference auth labels and add missing routes (#2829)+- docs: correct customer docs that drifted from the source (#2826)+- Remove the Assistant mutation handlers nothing calls (#2824)+- Recover Windows guest readings and keep monitoring evidence trustworthy (#2817)+- Give Alert History messages room to wrap instead of truncating (#2823)+- Say when a Proxmox guest's memory reading may include cache (#2820)+- Make the first-run Infrastructure screen readable and truthful (#2819)+- Keep end-to-end sign-in and agent removal checks reliable (#2816)+- Record why a disk's operator state does not follow its ID (#2815)+- Keep tenant cleanup safe and preserve unattended-update completion evidence (#2814)+- Keep notification recovery feedback in its view and release discarded history storage (#2813)+- Bound alert-history caching and keep updater tests and agent help accurate (#2812)+- Keep updates recoverable and alert settings and History trustworthy (#2795)+- Rebuild the mock view when only a link's creation time changes (#2811)+- Stop pin succession erasing an operator split (#2810)+- Keep a remediation lock readable on an ID a succession retired (#2809)+- Drop the dead inputs from useWorkloadsState (#2807)+- Port the inline selection scroll E2E spec to the platform pages (#2803)+- Stop the Windows polling contract depending on host speed (#2804)+- Build the mock linked view at the rebuild's link boundary (#2799)+- Require page-owned state on WorkloadsSurface (#2802)+- Drop the unread postures response meta from the frontend (#2801)+- Stop mock-mode callers that read identity rebuilding the estate view (#2798)+- Keep a remediation lock when a reporter gains an identity key (#2800)+- Hide section tabs and lift table preview caps in kiosk (#2797)+- Carry the PBS-reported node name into the connections ledger (#2794)+- Fold chained manual links the same way in any link order (#2793)+- Keep an operator-split host pair as two rows in the browser (#2792)+- Carry operator decisions across a rebuild that cannot read them (#2791)+- Keep guest disk totals, email settings and recovery guidance trustworthy (#2790)+- Keep a split PVE disk off the agent's SMART row (#2789)+- Judge adaptive-polling freshness by the scheduler's cadence (#2788)+- Drop the unread points response meta from the recovery decode (#2787)+- Stop monitoring linking a node to an agent the operator split off (#2786)+- Keep a remediation lock across links and ID successions (#2783)+- Open the Patrol mode selector when a link lands on its anchor (#2784)+- Admit human action dispatch under the policy coordinator (#2781)+- Stop Storage group headers pinning the group they collapse (#2782)+- Undo only the links that hold a reported source in report-merge (#2780)+- Preserve notification edits and truthful monitoring recovery (#2768)+- Stop mock target lookups rebuilding the estate view per tick (#2778)+- Remove the inert Assistant Protected guests setting (#2776)+- Hide every filter toolbar in kiosk mode (#2779)+- Judge connection freshness by the monitor's live poll cadence (#2774)+- Stop TempDir-backed alert managers in two tests (#2775)+- Trim useRecoveryPoints to what the TrueNAS Protection tab reads (#2773)+- Show drawer identity values whole instead of cutting them (#2772)+- Correct Patrol mode selector and dialog contract prose (#2771)+- Drop audit rules whose replacement modules were deleted (#2770)+- Keep guest memory and issue reports truthful (#2767)+- Keep guest memory readings and issue versions accurate, with complete parallel CI smoke checks (#2766)+- Make Assistant planning controls and guest-memory reporting guidance accurate (#2765)+- Keep guest memory evidence honest across AI queries and observers (#2764)+- Keep Patrol guest memory warnings grounded and reject native image drift earlier (#2763)+- Keep alert timing and workload attribution truthful and notification redirects private (#2761)+- Preserve certificate trust and keep guest readings and AI controls honest (#2760)+- Keep guest readings, monitoring and update recovery trustworthy (#2576)+- Fix Patrol investigation of host filesystem alerts (#2754)+- Clarify unavailable Docker container updates (#2753)+- Drop audit rules and allowlist entries for deleted recovery modules (#2752)+- Remove recovery exports and fallbacks the Recovery page left dead (#2750)+- Remove WorkloadsSurface's unreachable built-in filter toolbar (#2749)+- Let an operator split override a Proxmox node's agent link (#2741)+- Let a clean full Patrol clear the stale coverage caveat (#2697)+- Point dev:verify at the Proxmox backups layout spec (#2747)+- Honour operator splits in the linked physical disk join (#2748)+- Judge PBS and PMG freshness by the live polling interval (#2746)+- Apply operator links in the mock-mode unified view (#2732)+- Judge resources API freshness by the monitor's thresholds (#2745)+- Remove the summary chart routes no client calls (#2720)+- Start fresh installs with factory Docker alert thresholds (#2683)+- Point Patrol trigger contracts at the Patrol settings page (#2691)+- Resolve saved hosts an operator link joins to the link's primary (#2744)+- Resolve Patrol finding references before reading operator state (#2743)+- Resolve references to resources a manual link folded (#2721)+- Record the browser pass for main's merged Patrol state hook (#2742)+- Remove test-only Patrol autonomy save helpers (#2740)+- Stop Workloads row focus staging a stale shell scroll restore (#2736)+- Split operator-linked resources when a merge is reported (#2739)+- Remove the unreachable Workloads chart visibility toggle (#2734)+- Stop Patrol fetching the circuit-breaker status nobody reads (#2704)+- Drop Patrol state members no surface reads (#2695)+- Colour pulse-relaxed guests by the thresholds their alerts use (#2735)+- Stop Storage scrolling back to open pools and reopening groups (#2733)+- Bind SAML logins to the browser that started them (#2729)+- Wrap long identity chips inside phone drawers (#2731)+- Stop sending the ignored autonomous_mode chat field (#2730)+- Name sensor setup notice nodes the way the nodes table does (#2727)+- Drop the dead recovery serializer and its stale contract prose (#2728)+- Stop hidden tooltips and breakpoints listening to resize (#2724)+- Correct stale platform-page embed contract prose (#2726)+- Split a manually linked pair when the API unlinks it (#2725)+- Remove the unreachable Workloads group drawer and group pin (#2723)+- Retire the summary hover E2E spec pinned to removed surfaces (#2722)+- Pin each manual-link side from its own identity (#2719)+- Correct stale Storage embed and History range contract prose (#2718)+- Let only POST narrate single-resource reports with Pulse Assistant (#2716)+- Judge source merges by the ingest's stale thresholds (#2717)+- Drop the unused Patrol readiness-failure object behind the toggle toast (#2687)+- Stop tenant incident stores racing on shared ID counters (#2711)+- Remove unused CorrelationsResponse type (#2710)+- Alert on the Unraid temperature an agent disk shows (#2701)+- Drop the summary focus bridge members nothing reads (#2709)+- Hide pprof routes on demo-mode instances (#2708)+- Judge a linked agent's guest disks by that agent everywhere (#2712)+- Drop contract prose for the deleted problem-resource surfaces (#2706)+- Keep alert settings a config save leaves out (#2689)+- Judge pulse-relaxed guests by one threshold on poll and save (#2705)+- Raise Proxmox disk temperature alerts under the per-type policy (#2680)+- Count down to future Patrol times instead of reading just now (#2702)+- Fill the table's visible width with phone drawers (#2703)+- Require POST for Assistant session mutation routes (#2670)+- Judge filesystem alerts by their own threshold on config save (#2700)+- Remove unread Patrol summaryStats and its audit rule (#2690)+- Stop Patrol fetching the correlation list nobody reads (#2677)+- Read controller-member disk history at its written key (#2694)+- Remove dead Patrol assessment and save-failure handoffs (#2682)+- Keep Off alert defaults off after saving (#2684)+- Remove exports orphaned by the summary component deletions (#2671)+- Keep a low alert default's clear below its trigger (#2678)+- Remove test-only Patrol context summary helpers (#2674)+- Read the shared wall clock on Alerts overview cards (#2673)+- Keep platform metric alerts through Proxmox node cleanup (#2672)+- Read a silent in-guest agent's guest drawer readings as last known (#2669)+- Stop websocket upgrade headers exempting demo-mode writes (#2666)+- Refuse a seeded disk slot mapping when the disk was replaced (#2667)+- Show the outdated sensor setup notice on every Proxmox tab (#2668)+- Switch every running tenant monitor when mock mode flips (#2665)+- Rank a confirmed availability outage first in the summary check (#2664)+- Stop a failing availability check keeping a node online (#2663)+- Show retained disk temperatures as last known in drawers and tables (#2659)+- Drop the summary chart caches nothing reads (#2662)+- Keep same-serial disks on different machines apart (#2660)+- Keep watchdog, delivery log and Patrol ages moving (#2598)+- Lead open History rows with the live reading of held alerts (#2638)+- Stop a stale agent SMART row lending a replaced disk its identity (#2658)+- Judge disk heat under the reporting host's Disk Temp override (#2654)+- Keep fixture alerts from outliving a mock-mode switch (#2656)+- Show a silent agent's retained RAID state as last known (#2652)+- Stop reporting unavailable storage as disabled to AI tools (#2657)+- Remove status and disk helpers left unused by the selector removal (#2651)+- Stop QEMU default disk serials merging nested Proxmox disks (#2653)+- Remove contract prose and badge helpers for deleted surfaces (#2655)+- Remove the unreachable resource policy summary card (#2650)+- Remove exports orphaned by the unified resource table removal (#2649)+- Stop a replaced PVE disk inheriting the old disk's evidence (#2648)+- Show a remote unacknowledge in a tab that acknowledged locally (#2647)+- Own PVE disk alert history rows by recorded hardware (#2646)+- Pair a merged disk's temperature with its rows' own states (#2645)+- Show a silent agent's retained disk usage as last known (#2644)+- Keep a source's own offline verdict through the stale pass (#2643)+- Move static inline styles to classes the production CSP allows (#2642)+- Name disk wearout as life remaining and log disk alerts once (#2640)+- Stop reports and resource history calling a moved alert recovered (#2607)+- Lead alert cards and Patrol with the live reading of held alerts (#2637)+- Keep PVE disk alert references out of resource history (#2627)+- Replace stored alerts with each payload instead of merging (#2631)+- Give Docker containers and services without IDs their own alert refs (#2629)+- Keep the cleared history hint box so hovered guests stay put (#2624)+- Judge disk heat by the alert disk temperature thresholds (#2636)+- Keep retained disk temperatures out of AI, Patrol and reports (#2625)+- Show a host whose agent stopped reporting as offline (#2634)+- Raise disk health alerts for mock physical disks (#2633)+- Poll only the history each metric table renders (#2632)+- Call a disk that is only hot Running Hot, not Replace Now (#2623)+- Remove the unreachable host metric cells from grouped workload rows (#2622)+- Show real CPU low and record in the Proxmox node drawer (#2594)+- Keep a merged disk's temperature under its own row's state (#2618)+- Record Docker host, service and sub-resource alert history (#2620)+- Keep node temperatures off same-named nodes at other sites (#2617)+- Say a reviewed Patrol issue stays reviewed when suppression ends (#2615)+- Chart Net and Disk I/O in the row history lens (#2614)+- Show the live reading of threshold alerts held below trigger (#2567)+- Remove the never-mounted unified resource table (#2612)+- Settle slow polled reads and keep replication links honest (#2610)+- Judge disk risk only on a temperature collected now (#2611)+- Show retained disk temperatures as last known, not live (#2603)+- Stop serving retained disk temperatures as current readings (#2609)+- Record Proxmox and agent alert history on the canonical resource (#2608)+- Key previous Proxmox nodes by source ID so poll carry-overs match (#2604)+- Read a linked agent's own heartbeat in host snapshots (#2601)+- Stop presenting a silent agent's disk readings as collected (#2593)+- Record the browser pass for the combined Machines table (#2597)+- Show a moved alert as moved in Patrol, history and assistant (#2591)+- Tint Kubernetes node rows for their drawer's alerts (#2596)+- Clear agent disk temperature alerts when the disk leaves (#2595)+- Clear agent disk temperature alerts when their threshold is off (#2589)+- Carry a typed reason when an alert closes without recovering (#2580)+- Show each identifier once in the drawer's Identity card (#2590)+- Keep the availability result readable and give names room (#2588)+- Hand a node only the usage metrics its agent evaluates (#2581)+- Evaluate host agents in mock mode like live agents (#2585)+- Keep open row drawers spanning the table when columns change (#2571)+- Stop presenting a silent agent's node temperature as live (#2582)+- Give agent disk, array and sensor alerts their own resource type (#2583)+- Say what an agent update brings and what the CPU count is (#2584)+- Keep row, drawer and replication ages moving on the shared clock (#2578)+- Tint Docker host and Machines rows for their drawer's alerts (#2579)+- Dedup node alerts by agent link and keep node thresholds on save (#2577)+- Revalidate restored backup alerts before startup notifications (#2572)+- Stamp alert platforms so cards stop guessing the policy owner (#2574)+- Show open alerts in the Machines and Docker host drawers (#2575)+- Read platform alert Started as an age that keeps moving (#2573)+- Open an issue when the weekly signing preflight fails (#2570)+- Keep node temperature alerts when a host agent runs on the node (#2569)+- Run the release signing preflight every Monday (#2568)+- Show the vCenter column only when several are connected (#2566)+- Describe releases as the fixed release train everywhere (#2565)+- Link alert cards to their platform from incident metadata (#2564)+- Read vSphere activity as vCenter's own sentences (#2563)+- Collect LXC filesystems without pmxcfs or LXC socket IPC (#2562)+- Keep parent node identity out of Proxmox child aliases (#2561)+- Name vSphere, not Proxmox, as the owner of vSphere VM records (#2560)+- Say why vSphere datastores and networks need attention (#2559)+- Close the harness command client before Windows removes its state dir (#2558)+- Keep Docker update alerts open when the local digest is unknown (#2557)+- Stop the Restarts and Lifecycle headers clipping at tablet widths (#2555)+- Resolve masked webhook values per key and keep responses masked (#2556)+- Size Alert History columns to content so messages read (#2537)+- Bump golang in /.devcontainer (#2554)+- Register agent command channels under the identity Pulse resolves (#2550)+- Use the shared link and button primitives in two remaining spots (#2549)+- Paint the backgrounds the dead bg-base class dropped (#2548)+- Let agents without a command channel reduce their authority (#2547)+- Move three forms onto shared controls and draw the select arrow (#2546)+- Send built-in webhook template headers on real alerts (#2545)+- Show a Docker container's health check in its drawer (#2534)+- Keep paired PR benchmarks inside their time budget (#2544)+- Explain why the Container update risk switch is locked (#2541)+- Release v6.5.0 (#2536)+- Keep deployed agents' command channel across restarts (#2542)+- Keep agent-reported disks out of cluster member state (#2543)+- Keep Actions target names whole on narrow cards (#2539)+- Keep a disabled toggle's track visible and true to its state (#2535)+- Keep VM Discovery safe during backups and document precise Patrol rule removal (#2533)+- Give organization deletion time for a busy tenant loop to stop (#2529)+- Stop mock metrics seeding when its monitor is cancelled (#2532)+- Flag an outdated agent once on the Infrastructure page (#2531)+- Give Patrol autopilot tests their own session store (#2530)+- Reduce temporary memory churn in complete monitoring updates (#2527)+- Show Docker container health and uptime in the containers table (#2526)+- Keep the Proxmox guest Backup badge whole at laptop widths (#2525)+- Keep a text-selection drag from toggling table rows (#2524)+- Exchange the bootstrap token before any agent module starts (#2520)+- Keep Docker image names readable by their repository and tag (#2521)+- Copy Kubernetes names without a break between head and tail (#2522)+- Install cluster member agents in one step from a covered sibling (#2508)+- Preserve guest identity and History consistency with accessible backup evidence (#2518)+- Keep Docker update and Kubernetes node status badges whole (#2517)+- Keep the Docker image update badge readable at tablet widths (#2516)+- Bulk-edit Docker container links and hide edits from read-only tokens (#2515)+- Truncate platform summary cells inside their own column (#2514)+- Require empty subsystem default proof lists the guard never reads (#2512)+- Stop Kubernetes Config and Deployments tables clipping names (#2509)+- Stop Kubernetes names clipping at tablet widths (#2506)+- Retire the same-subsystem test wildcard the guards never honoured (#2505)+- Stop the SSE connect-snapshot test racing a queued duplicate (#2503)+- Read the guard's registry and contracts from the evaluated commit (#2504)+- Judge each commit's contract edit against that commit, not HEAD (#2502)+- Keep guest readings truthful and diagnostic help safe (#2486)+- Say why TrueNAS storage is unhealthy, drop the healthy pills (#2497)+- Point Audit Webhooks visitors to alert notifications (#2500)+- Call plain HTTP critical only when Pulse is publicly reachable (#2499)+- List every reason a resource needs attention in its drawer (#2498)+- Mark the alerting Mail Gateway number, keep rows single-line (#2496)+- Say which TrueNAS services and VMs should be running (#2495)+- Make availability checks agree across Settings and Machines (#2493)+- Show Mail Gateway alerts on the gateway they belong to (#2490)+- Explain backup protection in plain words (#2492)+- Reload the Patrol briefing E2E page by navigation on WebKit (#2491)+- Lead the Kubernetes overview with the nodes that are not Ready (#2489)+- Say when a Docker host goes silent, and stop repeating host columns (#2482)+- Make Patrol and Actions views linkable (#2488)+- Lead Notifications with where alerts go (#2487)+- Render light-mode banner and callout text at full strength (#2485)+- Lead the vSphere overview with vCenter's health signals (#2481)+- Lead alert thresholds with each group's default limits (#2484)+- Say what backup health counts mean and drop per-source age columns (#2483)+- Keep deferred guest readings, alerts and History truthful (#2472)+- Count alerting guests under Attention on the Proxmox overview (#2480)+- Add a bulk web-link editor to the workloads table (#2477)+- Turn the older-agent notice into one quiet line (#2479)+- Make the Alerts overview answer what is broken first (#2476)+- Lead the storage table with what is filling up, and say why (#2478)+- Stop repeating Kubernetes workload names in the demo (#2475)+- Open a guest's saved web interface from its drawer header (#2474)+- Label vSphere host groups with their cluster, not a VM (#2473)+- Keep guest monitoring truthful and preserve TrueNAS CORE inventory (#2471)+- Release v6.5.0-rc.1 (#2470)++## Release Metadata++- Version: `v6.6.0-beta.2`+- Previous candidate: `v6.6.0-beta.1`+- Previous stable: `v6.5.0`+- Rollback target: `v6.5.0`+- Rollback command: `sudo /bin/update --version v6.5.0`+- Promotion path: exact-SHA single-build release candidate from `main`+- Windows signing decision: prereleases publish checksum- and detached-signature-verified Windows agents without Authenticode while SignPath remains unavailable. Windows Unified Agent binaries may display an Unknown Publisher warning+- Mobile decision: `no-mobile-impact`. No governed mobile route, payload, Relay, pairing, approval, push or onboarding contract changed from `v6.5.0`, so no companion mobile build or store rollout is requireddiff --git a/frontend-modern/public/docs/UPGRADE_v6.md b/frontend-modern/public/docs/UPGRADE_v6.mdindex cbbccbcc56..3a244366ab 100644--- a/frontend-modern/public/docs/UPGRADE_v6.md+++ b/frontend-modern/public/docs/UPGRADE_v6.md@@ -4,8 +4,8 @@ This guide covers practical upgrade steps for existing Pulse installs moving toFor the current v6 beta packet, see:-- `docs/releases/RELEASE_NOTES_v6.6.0-beta.1.md`-- `docs/releases/V6_CHANGELOG_v6.6.0-beta.1.md`+- `docs/releases/RELEASE_NOTES_v6.6.0-beta.2.md`+- `docs/releases/V6_CHANGELOG_v6.6.0-beta.2.md`For the current stable v6 packet, see:diff --git a/scripts/install-docker.sh b/scripts/install-docker.shindex 01f5d1aab8..60ba0180cd 100755--- a/scripts/install-docker.sh+++ b/scripts/install-docker.sh@@ -6,7 +6,7 @@ set -euo pipefailSCRIPT_DIR="$(CDPATH= cd -- "$(dirname -- "$0")" && pwd)"DOCKER_IMAGE_REPO="${DOCKER_IMAGE_REPO:-rcourtman/pulse}"-CANONICAL_DEFAULT_PULSE_VERSION="6.6.0-beta.1"+CANONICAL_DEFAULT_PULSE_VERSION="6.6.0-beta.2"resolve_default_pulse_version() {if [ -n "${PULSE_IMAGE_VERSION:-}" ]; then