Skip to content

SeaweedFS is a distributed storage system for object storage (S3), file systems, Iceberg table buckets, and Lance vector buckets, designed to handle billions of files with O(1) disk access and effortless horizontal scaling.

fix(s3): keep server-side copy data in the bucket collection (#9607)

Chris Lu83b7ea5e7bca
diff --git a/weed/s3api/s3api_object_handlers_copy.go b/weed/s3api/s3api_object_handlers_copy.go
index 383ef21314..59338ac150 100644
--- a/weed/s3api/s3api_object_handlers_copy.go
+++ b/weed/s3api/s3api_object_handlers_copy.go
@@ -627,6 +627,17 @@ type CopyPartResult struct {
ETag string `xml:"ETag"`
}
+// copyPartLocation returns the destination directory and filename for a
+// server-side copy part under the destination bucket's .uploads folder. Copy
+// parts carry a fixed "copy" suffix (rather than the random suffix
+// genPartUploadPath mints for client uploads) so re-copying a part replaces
+// its predecessor in place.
+func (s3a *S3ApiServer) copyPartLocation(dstBucket, uploadID string, partID int) (uploadDir, partName string) {
+ uploadDir = s3a.genUploadsFolder(dstBucket) + "/" + uploadID
+ partName = fmt.Sprintf("%04d_%s.part", partID, "copy")
+ return
+}
+
func (s3a *S3ApiServer) CopyObjectPartHandler(w http.ResponseWriter, r *http.Request) {
t := time.Now().UTC().Truncate(time.Millisecond)
// https://docs.aws.amazon.com/AmazonS3/latest/dev/CopyingObjctsUsingRESTMPUapi.html
@@ -845,6 +856,15 @@ func (s3a *S3ApiServer) CopyObjectPartHandler(w http.ResponseWriter, r *http.Req
Extended: make(map[string][]byte),
}
+ // The copied part lives under the destination bucket's .uploads folder.
+ // Assign destination volumes against that real filer path so they land in
+ // the destination bucket's collection. r.URL.Path is the S3 request URI
+ // (e.g. /bucket/key), not a filer path, so passing it would skip the
+ // filer's bucket-to-collection mapping and route the copied bytes to the
+ // default collection.
+ uploadDir, partName := s3a.copyPartLocation(dstBucket, uploadID, partID)
+ dstPartPath := uploadDir + "/" + partName
+
// Handle zero-size files or empty ranges
if entry.Attributes.FileSize == 0 || endOffset < startOffset {
// For zero-size files or invalid ranges, create an empty part with size 0
@@ -852,7 +872,7 @@ func (s3a *S3ApiServer) CopyObjectPartHandler(w http.ResponseWriter, r *http.Req
dstEntry.Chunks = nil
} else {
// Copy chunks that overlap with the range
- dstChunks, err := s3a.copyChunksForRange(entry, startOffset, endOffset, r.URL.Path)
+ dstChunks, err := s3a.copyChunksForRange(entry, startOffset, endOffset, dstPartPath)
if err != nil {
glog.Errorf("CopyObjectPartHandler copy chunks error: %v", err)
s3err.WriteErrorResponse(w, r, s3err.ErrInternalError)
@@ -862,9 +882,6 @@ func (s3a *S3ApiServer) CopyObjectPartHandler(w http.ResponseWriter, r *http.Req
}
// Save the part entry to the multipart uploads folder
- uploadDir := s3a.genUploadsFolder(dstBucket) + "/" + uploadID
- partName := fmt.Sprintf("%04d_%s.part", partID, "copy")
-
// Check if part exists and remove it first (allow re-copying same part)
if exists, _ := s3a.exists(uploadDir, partName, false); exists {
if err := s3a.rm(uploadDir, partName, false, false); err != nil {
@@ -2276,7 +2293,10 @@ func (s3a *S3ApiServer) copyCrossEncryptionChunk(chunk *filer_pb.FileChunk, sour
// copyChunksWithSSEC handles SSE-C aware copying with smart fast/slow path selection
// Returns chunks and destination metadata that should be applied to the destination entry
-func (s3a *S3ApiServer) copyChunksWithSSEC(entry *filer_pb.Entry, r *http.Request) ([]*filer_pb.FileChunk, map[string][]byte, error) {
+// dstPath is the destination object's filer path, so volume assignment targets
+// the destination bucket's collection. The caller must not pass r.URL.Path (the
+// S3 request URI), which would route copied chunks to the default collection.
+func (s3a *S3ApiServer) copyChunksWithSSEC(entry *filer_pb.Entry, r *http.Request, dstPath string) ([]*filer_pb.FileChunk, map[string][]byte, error) {
// Parse SSE-C headers
copySourceKey, err := ParseSSECCopySourceHeaders(r)
@@ -2304,7 +2324,7 @@ func (s3a *S3ApiServer) copyChunksWithSSEC(entry *filer_pb.Entry, r *http.Reques
if isMultipartSSEC {
glog.V(2).Infof("Detected multipart SSE-C object with %d encrypted chunks for copy", sseCChunks)
- return s3a.copyMultipartSSECChunks(entry, copySourceKey, destKey, r.URL.Path)
+ return s3a.copyMultipartSSECChunks(entry, copySourceKey, destKey, dstPath)
}
// Single-part SSE-C object: use original logic
@@ -2320,13 +2340,13 @@ func (s3a *S3ApiServer) copyChunksWithSSEC(entry *filer_pb.Entry, r *http.Reques
case SSECCopyStrategyDirect:
// FAST PATH: Direct chunk copy
glog.V(2).Infof("Using fast path: direct chunk copy for %s", r.URL.Path)
- chunks, err := s3a.copyChunks(entry, r.URL.Path)
+ chunks, err := s3a.copyChunks(entry, dstPath)
return chunks, nil, err
case SSECCopyStrategyDecryptEncrypt:
// SLOW PATH: Decrypt and re-encrypt
glog.V(2).Infof("Using slow path: decrypt/re-encrypt for %s", r.URL.Path)
- chunks, destIV, err := s3a.copyChunksWithReencryption(entry, copySourceKey, destKey, r.URL.Path)
+ chunks, destIV, err := s3a.copyChunksWithReencryption(entry, copySourceKey, destKey, dstPath)
if err != nil {
return nil, nil, err
}
diff --git a/weed/s3api/s3api_object_handlers_copy_collection_test.go b/weed/s3api/s3api_object_handlers_copy_collection_test.go
new file mode 100644
index 0000000000..79d0bcac69
--- /dev/null
+++ b/weed/s3api/s3api_object_handlers_copy_collection_test.go
@@ -0,0 +1,45 @@
+package s3api
+
+import (
+ "fmt"
+ "testing"
+
+ "github.com/seaweedfs/seaweedfs/weed/filer"
+ "github.com/seaweedfs/seaweedfs/weed/util"
+)
+
+// TestCopyDestinationPathResolvesBucketCollection guards against server-side
+// copies routing data into the default collection.
+//
+// The filer maps a write to a bucket's collection only when the AssignVolume
+// Path sits under its buckets folder (/buckets/<bucket>/...). UploadPartCopy
+// and SSE-C CopyObject used to assign destination volumes against r.URL.Path,
+// the S3 request URI (e.g. /bucket/key), which never has that prefix; the
+// copied bytes therefore landed in the default collection instead of the
+// destination bucket's. The handlers must assign against the real filer path
+// of the destination.
+func TestCopyDestinationPathResolvesBucketCollection(t *testing.T) {
+ const bucket = "docker-registry"
+ s3a := &S3ApiServer{option: &S3ApiServerOption{BucketsPath: "/buckets"}}
+ f := &filer.Filer{DirBucketsPath: "/buckets"}
+
+ // The S3 request URI is not a filer path: the filer cannot derive the
+ // destination bucket from it. This is the shape that caused the leak.
+ if got := f.DetectBucket(util.FullPath("/" + bucket + "/blobs/data")); got != "" {
+ t.Fatalf("S3 request URI unexpectedly mapped to collection %q; the test no longer reproduces the bug", got)
+ }
+
+ // UploadPartCopy assigns against the destination part path under .uploads.
+ uploadDir, partName := s3a.copyPartLocation(bucket, "uploadid", 1)
+ partPath := uploadDir + "/" + partName
+ if got := f.DetectBucket(util.FullPath(partPath)); got != bucket {
+ t.Fatalf("UploadPartCopy dst path %q resolved to collection %q, want %q", partPath, got, bucket)
+ }
+
+ // CopyObject (including the SSE-C paths) assigns against the destination
+ // object path.
+ objPath := fmt.Sprintf("%s/%s", s3a.bucketDir(bucket), "blobs/data")
+ if got := f.DetectBucket(util.FullPath(objPath)); got != bucket {
+ t.Fatalf("CopyObject dst path %q resolved to collection %q, want %q", objPath, got, bucket)
+ }
+}
diff --git a/weed/s3api/s3api_object_handlers_copy_part_sse.go b/weed/s3api/s3api_object_handlers_copy_part_sse.go
index 36519446e7..097dcc26c1 100644
--- a/weed/s3api/s3api_object_handlers_copy_part_sse.go
+++ b/weed/s3api/s3api_object_handlers_copy_part_sse.go
@@ -407,8 +407,7 @@ func (s3a *S3ApiServer) copyObjectPartViaReencryption(
// writeEmptyCopyPart writes a 0-byte part entry for an empty UploadPartCopy
// range, mirroring the legacy fast path's handling of endOffset < startOffset.
func (s3a *S3ApiServer) writeEmptyCopyPart(dstBucket, uploadID string, partID int) (string, s3err.ErrorCode) {
- uploadDir := s3a.genUploadsFolder(dstBucket) + "/" + uploadID
- partName := fmt.Sprintf("%04d_%s.part", partID, "copy")
+ uploadDir, partName := s3a.copyPartLocation(dstBucket, uploadID, partID)
if exists, _ := s3a.exists(uploadDir, partName, false); exists {
if err := s3a.rm(uploadDir, partName, false, false); err != nil {
return "", s3err.ErrInternalError
diff --git a/weed/s3api/s3api_object_handlers_copy_unified.go b/weed/s3api/s3api_object_handlers_copy_unified.go
index 223e42ac99..7146f5b5bf 100644
--- a/weed/s3api/s3api_object_handlers_copy_unified.go
+++ b/weed/s3api/s3api_object_handlers_copy_unified.go
@@ -45,7 +45,7 @@ func (s3a *S3ApiServer) executeUnifiedCopyStrategy(entry *filer_pb.Entry, r *htt
return chunks, nil, err
case CopyStrategyKeyRotation:
- return s3a.executeKeyRotation(entry, r, state)
+ return s3a.executeKeyRotation(entry, r, state, dstBucket, dstPath)
case CopyStrategyEncrypt:
return s3a.executeEncryptCopy(entry, r, state, dstBucket, dstPath)
@@ -90,13 +90,13 @@ func (s3a *S3ApiServer) mapCopyErrorToS3Error(err error) s3err.ErrorCode {
}
// executeKeyRotation handles key rotation for same-object copies
-func (s3a *S3ApiServer) executeKeyRotation(entry *filer_pb.Entry, r *http.Request, state *EncryptionState) ([]*filer_pb.FileChunk, map[string][]byte, error) {
+func (s3a *S3ApiServer) executeKeyRotation(entry *filer_pb.Entry, r *http.Request, state *EncryptionState, dstBucket, dstPath string) ([]*filer_pb.FileChunk, map[string][]byte, error) {
// For key rotation, we only need to update metadata, not re-copy chunks
// This is a significant optimization for same-object key changes
if state.SrcSSEC && state.DstSSEC {
// SSE-C key rotation - need to handle new key/IV, use reencrypt logic
- return s3a.executeReencryptCopy(entry, r, state, "", "")
+ return s3a.executeReencryptCopy(entry, r, state, dstBucket, dstPath)
}
if state.SrcSSEKMS && state.DstSSEKMS {
@@ -105,14 +105,14 @@ func (s3a *S3ApiServer) executeKeyRotation(entry *filer_pb.Entry, r *http.Reques
}
// Fallback to reencrypt if we can't do metadata-only rotation
- return s3a.executeReencryptCopy(entry, r, state, "", "")
+ return s3a.executeReencryptCopy(entry, r, state, dstBucket, dstPath)
}
// executeEncryptCopy handles plain → encrypted copies
func (s3a *S3ApiServer) executeEncryptCopy(entry *filer_pb.Entry, r *http.Request, state *EncryptionState, dstBucket, dstPath string) ([]*filer_pb.FileChunk, map[string][]byte, error) {
if state.DstSSEC {
// Use existing SSE-C copy logic
- return s3a.copyChunksWithSSEC(entry, r)
+ return s3a.copyChunksWithSSEC(entry, r, dstPath)
}
if state.DstSSEKMS {
@@ -145,7 +145,7 @@ func (s3a *S3ApiServer) executeDecryptCopy(entry *filer_pb.Entry, r *http.Reques
func (s3a *S3ApiServer) executeReencryptCopy(entry *filer_pb.Entry, r *http.Request, state *EncryptionState, dstBucket, dstPath string) ([]*filer_pb.FileChunk, map[string][]byte, error) {
// Use chunk-by-chunk approach for all cross-encryption scenarios (consistent behavior)
if state.SrcSSEC && state.DstSSEC {
- return s3a.copyChunksWithSSEC(entry, r)
+ return s3a.copyChunksWithSSEC(entry, r, dstPath)
}
if state.SrcSSEKMS && state.DstSSEKMS {