s3api: preserve remote object ETags across cache fills (#11704)

* s3api: preserve remote object ETags across cache fills * s3api: align remote copy validators and reject stale ETags * s3api: keep remote ETag through metadata touches, drop it after resizes The remote-ETag check required Mtime to match RemoteMtime, so a utimens on a cold remote object returned an empty ETag until a later cache fill restored the remote value. It also trusted LastLocalSyncTsNs alone, so a setattr resize that leaves chunks and the stamp untouched kept serving the old remote ETag for different local bytes. Require FileSize to match RemoteSize for both the synced and remote-only paths and stop comparing modification times: content changes already surface through cleared sync stamps, chunks, or inline content, while a metadata-only touch leaves the remote bytes authoritative. Regression tests cover touched cold and cached entries and resized cached entries. --------- Co-authored-by: Chris Lu <chrislusf@users.noreply.github.com>
2026-10-12 01:07:35 +02:00 · 2026-10-10 23:58:22 +08:00
1 parent 17ba8b3c7c
commit 547befc070
@@ -3,6 +3,7 @@ package s3api
 import (
 	"net/http"
 	"net/http/httptest"
+	"strings"
 	"testing"
 	"time"

@@ -58,3 +59,182 @@ func TestValidateConditionalHeadersForReadsMissingObject(t *testing.T) {
 		})
 	}
 }
+
+func TestRemoteObjectETagAcrossCacheFill(t *testing.T) {
+	s3a := &S3ApiServer{}
+	entry := &filer_pb.Entry{
+		Name:        "object",
+		Attributes:  &filer_pb.FuseAttributes{FileSize: 2048, Mtime: 100},
+		RemoteEntry: &filer_pb.RemoteEntry{RemoteSize: 2048, RemoteMtime: 100, RemoteETag: "remote-multipart-2"},
+	}
+	head := httptest.NewRecorder()
+	s3a.setResponseHeaders(head, httptest.NewRequest(http.MethodHead, "/bucket/object", nil), entry, 2048)
+	headETag := head.Header().Get("ETag")
+	if headETag != `"remote-multipart-2"` {
+		t.Errorf("cold ETag = %q", headETag)
+	}
+	entry.Chunks = []*filer_pb.FileChunk{
+		{Offset: 0, Size: 1024, ETag: "1B2M2Y8AsgTpgAmY7PhCfg=="},
+		{Offset: 1024, Size: 1024, ETag: "1B2M2Y8AsgTpgAmY7PhCfg=="},
+	}
+	entry.RemoteEntry.LastLocalSyncTsNs = 200
+	if got := s3a.getObjectETag(entry); got != headETag {
+		t.Errorf("cached ETag = %q, cold ETag = %q", got, headETag)
+	}
+	for _, method := range []string{http.MethodGet, http.MethodHead} {
+		r := httptest.NewRequest(method, "/bucket/object", nil)
+		r.Header.Set(s3_constants.IfMatch, headETag)
+		r.Header.Set("Range", "bytes=1024-2047")
+		headers, code := parseConditionalHeaders(r)
+		if code != s3err.ErrNone {
+			t.Fatal(code)
+		}
+		result := s3a.validateConditionalHeadersForReads(r, headers, entry, "bucket", "object")
+		if result.ErrorCode != s3err.ErrNone {
+			t.Errorf("%s after cache fill: %v", method, result.ErrorCode)
+		}
+	}
+	entry.RemoteEntry.RemoteETag = `"remote-multipart-2"`
+	if got := s3a.getObjectETag(entry); got != headETag {
+		t.Errorf("quoted remote ETag = %q", got)
+	}
+	entry.Extended = map[string][]byte{s3_constants.ExtETagKey: []byte("local-etag")}
+	if got := s3a.getObjectETag(entry); got != `"local-etag"` {
+		t.Errorf("explicit ETag = %q", got)
+	}
+	entry.Extended = nil
+	entry.RemoteEntry.LastLocalSyncTsNs = 0
+	if got, want := s3a.getObjectETag(entry), s3a.calculateETagFromChunks(entry.Chunks); got != want {
+		t.Errorf("unsynced local ETag = %q, want %q", got, want)
+	}
+}
+
+// A metadata-only update such as utimens changes Mtime without touching the
+// remote bytes, so the object must keep its remote ETag; otherwise a later
+// cache fill restores it and breaks clients holding the interim value.
+func TestRemoteObjectETagMetadataOnlyTouch(t *testing.T) {
+	s3a := &S3ApiServer{}
+	for _, cached := range []bool{false, true} {
+		entry := &filer_pb.Entry{
+			Name:       "object",
+			Attributes: &filer_pb.FuseAttributes{FileSize: 2048, Mtime: 100},
+			RemoteEntry: &filer_pb.RemoteEntry{
+				RemoteSize: 2048, RemoteMtime: 100, RemoteETag: "remote-multipart-2",
+			},
+		}
+		if cached {
+			entry.Chunks = []*filer_pb.FileChunk{{Size: 2048, ETag: "1B2M2Y8AsgTpgAmY7PhCfg=="}}
+			entry.RemoteEntry.LastLocalSyncTsNs = 200
+		}
+		entry.Attributes.Mtime = 101
+		entry.Attributes.MtimeNs = 500
+		if got := s3a.getObjectETag(entry); got != `"remote-multipart-2"` {
+			t.Errorf("cached=%v touched ETag = %q", cached, got)
+		}
+	}
+}
+
+// Extending or shrinking a cached remote file through setattr changes FileSize
+// while leaving its chunks and sync stamp alone; the remote ETag no longer
+// describes the local bytes and conditional reads must not answer 304 for it.
+func TestRemoteObjectETagLocalResize(t *testing.T) {
+	s3a := &S3ApiServer{}
+	for _, size := range []uint64{0, 1024, 4096} {
+		entry := &filer_pb.Entry{
+			Name:       "object",
+			Attributes: &filer_pb.FuseAttributes{FileSize: 2048, Mtime: 100},
+			Chunks:     []*filer_pb.FileChunk{{Size: 2048, ETag: "1B2M2Y8AsgTpgAmY7PhCfg=="}},
+			RemoteEntry: &filer_pb.RemoteEntry{
+				RemoteSize: 2048, RemoteMtime: 100, RemoteETag: "remote-multipart-2",
+				LastLocalSyncTsNs: 200,
+			},
+		}
+		entry.Attributes.FileSize = size
+		if got := s3a.getObjectETag(entry); got == `"remote-multipart-2"` {
+			t.Errorf("resized to %d retained remote ETag", size)
+		}
+		r := httptest.NewRequest(http.MethodHead, "/bucket/object", nil)
+		r.Header.Set(s3_constants.IfNoneMatch, `"remote-multipart-2"`)
+		headers, code := parseConditionalHeaders(r)
+		if code != s3err.ErrNone {
+			t.Fatal(code)
+		}
+		if got := s3a.validateConditionalHeadersForReads(r, headers, entry, "bucket", "object"); got.ErrorCode != s3err.ErrNone {
+			t.Errorf("resized to %d read: %v", size, got.ErrorCode)
+		}
+	}
+}
+
+func TestRemoteObjectETagCopyConditions(t *testing.T) {
+	s3a := &S3ApiServer{}
+	entry := &filer_pb.Entry{
+		Attributes:  &filer_pb.FuseAttributes{FileSize: 2048, Mtime: 100},
+		RemoteEntry: &filer_pb.RemoteEntry{RemoteSize: 2048, RemoteMtime: 100, RemoteETag: "remote-multipart-2"},
+	}
+	for _, cached := range []bool{false, true} {
+		if cached {
+			entry.Chunks = []*filer_pb.FileChunk{{Size: 2048, ETag: "1B2M2Y8AsgTpgAmY7PhCfg=="}}
+			entry.RemoteEntry.LastLocalSyncTsNs = 200
+		}
+		for _, condition := range []string{s3_constants.AmzCopySourceIfMatch, s3_constants.AmzCopySourceIfNoneMatch} {
+			r := httptest.NewRequest(http.MethodPut, "/bucket/copy", nil)
+			r.Header.Set(condition, s3a.getObjectETag(entry))
+			want := s3err.ErrNone
+			if condition == s3_constants.AmzCopySourceIfNoneMatch {
+				want = s3err.ErrPreconditionFailed
+			}
+			if got := s3a.validateConditionalCopyHeaders(r, entry); got != want {
+				t.Errorf("cached=%v %s: got %v, want %v", cached, condition, got, want)
+			}
+		}
+	}
+}
+
+func TestRemoteObjectETagLocalChanges(t *testing.T) {
+	s3a := &S3ApiServer{}
+	for _, content := range [][]byte{nil, []byte("new local content")} {
+		entry := &filer_pb.Entry{
+			Attributes:  &filer_pb.FuseAttributes{FileSize: uint64(len(content)), Mtime: 101},
+			Content:     content,
+			RemoteEntry: &filer_pb.RemoteEntry{RemoteSize: 2048, RemoteMtime: 100, RemoteETag: "old-origin"},
+		}
+		if got, want := copyEntryETag(entry), strings.Trim(s3a.getObjectETag(entry), `"`); got != want {
+			t.Errorf("local copy ETag = %q, read ETag = %q", got, want)
+		}
+		if got := s3a.getObjectETag(entry); got == `"old-origin"` {
+			t.Errorf("local content %q retained remote ETag", content)
+		}
+		r := httptest.NewRequest(http.MethodHead, "/bucket/object", nil)
+		r.Header.Set(s3_constants.IfNoneMatch, `"old-origin"`)
+		headers, code := parseConditionalHeaders(r)
+		if code != s3err.ErrNone {
+			t.Fatal(code)
+		}
+		if got := s3a.validateConditionalHeadersForReads(r, headers, entry, "bucket", "object"); got.ErrorCode != s3err.ErrNone {
+			t.Errorf("local content %q read: %v", content, got.ErrorCode)
+		}
+		r.Header.Set(s3_constants.AmzCopySourceIfMatch, `"old-origin"`)
+		if got := s3a.validateConditionalCopyHeaders(r, entry); got != s3err.ErrPreconditionFailed {
+			t.Errorf("local content %q copy: %v", content, got)
+		}
+	}
+}
+
+func TestRemoteObjectETagFallback(t *testing.T) {
+	s3a := &S3ApiServer{}
+	entry := &filer_pb.Entry{
+		Attributes:  &filer_pb.FuseAttributes{Md5: []byte{0xab, 0xcd}},
+		RemoteEntry: &filer_pb.RemoteEntry{},
+	}
+	if got := s3a.getObjectETag(entry); got != `"abcd"` {
+		t.Errorf("missing remote ETag = %q", got)
+	}
+	entry.RemoteEntry.RemoteETag = "empty-origin"
+	if got := s3a.getObjectETag(entry); got != `"empty-origin"` {
+		t.Errorf("empty remote object ETag = %q", got)
+	}
+	entry.RemoteEntry = nil
+	if got := s3a.getObjectETag(entry); got != `"abcd"` {
+		t.Errorf("local object ETag = %q", got)
+	}
+}