Merge branch 'master' into fix-9478-partial-ec-with-sibling-dat

This commit is contained in:
Chris Lu 2026-05-13 00:19:36 -07:00 committed by GitHub
commit d105b58417
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
7 changed files with 190 additions and 33 deletions

View File

@ -86,8 +86,12 @@ func backdateVersionedMtime(t *testing.T, fc filer_pb.SeaweedFilerClient, bucket
func runLifecycleShard(t *testing.T) string {
t.Helper()
// -refresh 1s drives an inter-pass loop inside the 10s runtime so
// rules that need two passes to settle (e.g.
// TestLifecycleExpiredDeleteMarkerCleanup: pass 1 removes v1, pass 2
// removes the now-orphaned marker) get the second pass.
out := runShellCommand(t, fmt.Sprintf(
"s3.lifecycle.run-shard -shards 0-15 -s3 %s -events 0 -dispatch 200ms -checkpoint 5s -runtime 10s",
"s3.lifecycle.run-shard -shards 0-15 -s3 %s -events 0 -refresh 1s -runtime 10s",
envOr("S3_GRPC_ENDPOINT", defaultS3GrpcEndpoint),
))
require.NotContains(t, out, "FATAL", "shell output:\n%s", out)

View File

@ -104,6 +104,7 @@ func walkBucketTree(ctx context.Context, client filer_pb.SeaweedFilerClient, dir
ModTime: time.Unix(e.Attributes.Mtime, int64(e.Attributes.MtimeNs)),
Size: int64(e.Attributes.FileSize),
IsLatest: true, // Non-versioned default.
Tags: extractTags(e.Extended),
}
return cb(entry)
})
@ -183,7 +184,7 @@ func expandVersionsDir(ctx context.Context, client filer_pb.SeaweedFilerClient,
// Resolve latest:
// 1. Pointer names a real id -> that wins.
// 2. Pointer absent (or stale: set but no sibling carries it)
// + items[0] is an EXPLICIT null -> null is latest.
// + any EXPLICIT null bare exists -> null is latest.
// 3. Otherwise -> newest sibling (latestPos = 0 by default).
//
// A stale pointer falls through to the no-pointer fallback rather
@ -203,8 +204,13 @@ func expandVersionsDir(ctx context.Context, client filer_pb.SeaweedFilerClient,
}
}
}
if !pointerResolved && len(items) > 0 && items[0].versionID == "null" && items[0].isExplicitNull {
latestPos = 0
if !pointerResolved {
for i, it := range items {
if it.versionID == "null" && it.isExplicitNull {
latestPos = i
break
}
}
}
if start != "" && logical <= start {
@ -228,6 +234,7 @@ func expandVersionsDir(ctx context.Context, client filer_pb.SeaweedFilerClient,
IsDeleteMarker: string(it.entry.Extended[s3_constants.ExtDeleteMarkerKey]) == "true",
NumVersions: len(items),
SuccessorModTime: successor,
Tags: extractTags(it.entry.Extended),
}
if !isLatest {
rank := i
@ -318,3 +325,21 @@ func isMPUInitDir(key string, entry *filer_pb.Entry) bool {
v, ok := entry.Extended[s3_constants.ExtMultipartObjectKey]
return ok && len(v) > 0
}
func extractTags(ext map[string][]byte) map[string]string {
if len(ext) == 0 {
return nil
}
prefix := s3_constants.AmzObjectTagging + "-"
var tags map[string]string
for k, v := range ext {
if !strings.HasPrefix(k, prefix) {
continue
}
if tags == nil {
tags = make(map[string]string)
}
tags[k[len(prefix):]] = string(v)
}
return tags
}

View File

@ -126,6 +126,25 @@ func TestFilerListFunc_EmitsFlatFiles(t *testing.T) {
assert.Equal(t, []string{"a.txt", "b.txt"}, got)
}
func TestFilerListFunc_PropagatesTagsOnFlatFiles(t *testing.T) {
mtime := time.Now().Add(-7 * 24 * time.Hour)
tagged := fileWithExt("tagged.txt", mtime, 10, map[string][]byte{
s3_constants.AmzObjectTagging + "-env": []byte("temp"),
s3_constants.AmzObjectTagging + "-tier": []byte("cold"),
})
client := &fakeFiler{tree: map[string][]*filer_pb.Entry{
"/buckets/bkt": {tagged},
}}
listFn := FilerListFunc(client, "/buckets")
var got *bootstrap.Entry
require.NoError(t, listFn(context.Background(), "bkt", "", func(e *bootstrap.Entry) error {
got = e
return nil
}))
require.NotNil(t, got)
assert.Equal(t, map[string]string{"env": "temp", "tier": "cold"}, got.Tags)
}
func TestFilerListFunc_RecursesIntoSubdirs(t *testing.T) {
mtime := time.Now()
client := &fakeFiler{tree: map[string][]*filer_pb.Entry{
@ -360,6 +379,40 @@ func TestFilerListFunc_VersionedExpansionExplicitNullIsLatestWhenPointerMissing(
assert.Equal(t, 2, count)
}
func TestFilerListFunc_VersionedExpansionExplicitNullBeatsNewerSiblingWhenPointerMissing(t *testing.T) {
// Suspended-versioning current-ness is carried by the explicit bare
// null, not by mtime ordering. Backdating the null to make it due
// must not demote it behind a newer noncurrent sibling when the
// .versions pointer is absent.
tNull := time.Date(2026, 1, 1, 0, 0, 0, 0, time.UTC) // older
tV1 := time.Date(2026, 3, 1, 0, 0, 0, 0, time.UTC) // newer
bareNull := fileWithExt("foo", tNull, 1, map[string][]byte{s3_constants.ExtVersionIdKey: []byte("null")})
client := &fakeFiler{tree: map[string][]*filer_pb.Entry{
"/buckets/bkt": {
bareNull,
versionsDir("foo"+s3_constants.VersionsFolder, ""),
},
"/buckets/bkt/foo" + s3_constants.VersionsFolder: {
fileWithExt("v1", tV1, 1, map[string][]byte{s3_constants.ExtVersionIdKey: []byte("v1")}),
},
}}
listFn := FilerListFunc(client, "/buckets")
var got []*bootstrap.Entry
require.NoError(t, listFn(context.Background(), "bkt", "", func(e *bootstrap.Entry) error {
got = append(got, e)
return nil
}))
require.Len(t, got, 2)
byID := map[string]*bootstrap.Entry{}
for _, e := range got {
byID[e.VersionID] = e
}
require.NotNil(t, byID["null"])
require.NotNil(t, byID["v1"])
assert.True(t, byID["null"].IsLatest, "explicit null must stay latest even when older than noncurrent siblings")
assert.False(t, byID["v1"].IsLatest)
}
func TestFilerListFunc_VersionsDirWithoutMarkersRecursesAsRegular(t *testing.T) {
// A `.versions`-named folder whose children have no
// ExtVersionIdKey is a coincidence (user folder). Recurse into

View File

@ -204,6 +204,20 @@ func runShard(ctx context.Context, cfg Config, snap *engine.Snapshot, runNow tim
promoted := engine.PromotedHash(snap, retentionWindow)
if rsh == [32]byte{} {
// No replay-eligible rules. Walker-only rules
// (ExpirationDate / ExpiredDeleteMarker / NewerNoncurrent)
// or scan_only-promoted rules might still need a walk; run
// the steady-state walker before persisting the empty
// cursor and returning. Without this, a bucket whose only
// rule is walker-bound would never have it dispatched —
// the bug TestLifecycleExpirationDateInThePast caught.
if cfg.Walker != nil {
if _, walkView := snap.RulesForShard(shardID, retentionWindow); walkView != nil && len(walkView.AllActions()) > 0 {
if werr := cfg.Walker(ctx, walkView, shardID); werr != nil {
return fmt.Errorf("shard=%d: steady walk (empty replay): %w", shardID, werr)
}
}
}
return cfg.Persister.Save(ctx, shardID, Cursor{
TsNs: 0,
RuleSetHash: rsh,
@ -211,22 +225,36 @@ func runShard(ctx context.Context, cfg Config, snap *engine.Snapshot, runNow tim
})
}
// Recovery: rule-content edit (RuleSetHash mismatch) or partition
// flip (PromotedHash mismatch). Walk the rewritten rule set so
// already-due objects fire before the cursor rewinds; then rewind
// and let the sliding meta-log replay catch up steady state.
if found && (persisted.RuleSetHash != rsh || persisted.PromotedHash != promoted) {
// Recovery / cold-start walker:
// - found && hashes mismatch: rule edit or partition flip — walk
// the rewritten rule set so already-due objects fire before the
// cursor rewinds, then rewind for meta-log replay.
// - !found: first run for this shard. Pre-existing objects PUT
// before the rule was added live OUTSIDE the meta-log scan
// window (TsNs > runNow - maxTTL) and would never replay; the
// walker has to discover them. The streaming worker did this
// via BucketBootstrapper; daily-replay needs the same.
mustWalkRecovery := found && (persisted.RuleSetHash != rsh || persisted.PromotedHash != promoted)
mustWalkColdStart := !found
if mustWalkRecovery || mustWalkColdStart {
if cfg.Walker != nil {
if werr := cfg.Walker(ctx, engine.RecoveryView(snap), shardID); werr != nil {
return fmt.Errorf("shard=%d: recovery walk: %w", shardID, werr)
}
}
next := Cursor{
TsNs: runNow.Add(-maxTTL).UnixNano(),
RuleSetHash: rsh,
PromotedHash: promoted,
if mustWalkRecovery {
// Rule changed: rewind cursor so the sliding replay re-scans
// the new max-TTL window and the persisted hashes match the
// new rule set.
next := Cursor{
TsNs: runNow.Add(-maxTTL).UnixNano(),
RuleSetHash: rsh,
PromotedHash: promoted,
}
return cfg.Persister.Save(ctx, shardID, next)
}
return cfg.Persister.Save(ctx, shardID, next)
// Cold start: keep TsNs=0 so the drain below floors to
// runNow - maxTTL and the cursor is saved fresh after the run.
}
// Steady-state walker for walker-bound and scan_only-promoted rules.
@ -244,20 +272,39 @@ func runShard(ctx context.Context, cfg Config, snap *engine.Snapshot, runNow tim
}
// Cold start: scan from now-maxTTL so already-due objects within
// meta-log retention still expire.
// meta-log retention still expire. In steady state honor the
// cursor as-is: the drain freezes the cursor at the last pre-skip
// event so pending matches with DueTime == TsNs+maxTTL stay in
// scope across passes. Bumping forward to runNow-maxTTL would
// orphan exactly those events (the test_lifecyclev2_expiration
// regression: cursor saved at the no-match event right before
// the not-yet-due expire3 PUT, then floor at runNow=PUT+maxTTL
// equals PUT — bumping past the expire3 event itself).
startTsNs := persisted.TsNs
floor := runNow.Add(-maxTTL).UnixNano()
if startTsNs < floor {
startTsNs = floor
if !found {
startTsNs = runNow.Add(-maxTTL).UnixNano()
}
lastOK, _, drainErr := drainShardEvents(ctx, cfg, runNow, shardID, snap, startTsNs)
// Cursor save uses a fresh ctx because the steady-state drain exits
// via passCtx cancellation (the only signal the filer subscription
// gets when no new events arrive). Saving with the canceled passCtx
// would silently drop the cursor and the next pass would re-replay
// from the same floor — defeating advancement entirely.
saveCtx, saveCancel := context.WithTimeout(context.Background(), 5*time.Second)
defer saveCancel()
if drainErr != nil {
_ = cfg.Persister.Save(ctx, shardID, Cursor{TsNs: lastOK, RuleSetHash: rsh, PromotedHash: promoted})
_ = cfg.Persister.Save(saveCtx, shardID, Cursor{TsNs: lastOK, RuleSetHash: rsh, PromotedHash: promoted})
// passCtx timeout is the expected end-of-pass for an idle
// subscription; not a real error. Other drain errors still
// propagate.
if errors.Is(drainErr, context.DeadlineExceeded) || errors.Is(drainErr, context.Canceled) {
return nil
}
return fmt.Errorf("shard=%d: drain: %w", shardID, drainErr)
}
return cfg.Persister.Save(ctx, shardID, Cursor{
return cfg.Persister.Save(saveCtx, shardID, Cursor{
TsNs: lastOK,
RuleSetHash: rsh,
PromotedHash: promoted,

View File

@ -35,16 +35,18 @@ func (d *WalkerDispatcher) Delete(ctx context.Context, action *engine.CompiledAc
}
objectPath := entry.Path
if entry.IsMPUInit {
// Rule-prefix matching used DestKey; the server takes the
// canonical object path for the LifecycleDelete RPC, which
// is also DestKey. The walker hits the .uploads/<id>
// directory itself only when ActionKind=ABORT_MPU, and the
// server resolves the upload from (bucket, object_path) +
// the init record's metadata.
// Rule-prefix matching uses DestKey (the user's intended
// object key); dispatch uses entry.Path (.uploads/<id>),
// which is what the server's ABORT_MPU handler expects in
// req.ObjectPath — it strips the .uploads/ prefix to get
// the upload id and reads the init record from that
// directory. DestKey is the dispatch ANTI-pattern here: it
// looks like a regular object path, the server's check for
// the .uploads/ prefix fails, and the dispatch comes back
// as BLOCKED FATAL_EVENT_ERROR.
if entry.DestKey == "" {
return fmt.Errorf("walker dispatch: MPU init entry with empty DestKey: %s", entry.Path)
}
objectPath = entry.DestKey
}
rh := action.Key.RuleHash
req := &s3_lifecycle_pb.LifecycleDeleteRequest{

View File

@ -72,9 +72,11 @@ func TestWalkerDispatcher_VersionedPassesVersionID(t *testing.T) {
assert.Equal(t, "v-abc", c.lastReq.VersionId)
}
func TestWalkerDispatcher_MPUInitUsesDestKey(t *testing.T) {
// Rule-prefix matching used DestKey; the RPC ObjectPath must match
// so the server resolves the upload from the user's intended key.
func TestWalkerDispatcher_MPUInitUsesUploadsPath(t *testing.T) {
// Rule-prefix matching uses DestKey to decide IF this MPU init
// matches; dispatch uses Path (.uploads/<id>) because the server's
// ABORT_MPU handler strips the .uploads/ prefix to get the upload
// id. Sending DestKey here would BLOCK with FATAL_EVENT_ERROR.
c := &walkerStubClient{outcome: s3_lifecycle_pb.LifecycleDeleteOutcome_DONE}
d := &WalkerDispatcher{Client: c}
a := sampleAction(t, s3lifecycle.ActionKindAbortMPU)
@ -84,7 +86,7 @@ func TestWalkerDispatcher_MPUInitUsesDestKey(t *testing.T) {
IsMPUInit: true,
})
require.NoError(t, err)
assert.Equal(t, "user/path/object", c.lastReq.ObjectPath)
assert.Equal(t, ".uploads/abc123", c.lastReq.ObjectPath)
}
func TestWalkerDispatcher_MPUInitEmptyDestKeyErrors(t *testing.T) {

View File

@ -122,9 +122,24 @@ func (c *commandS3LifecycleRunShard) Do(args []string, env *CommandEnv, writer i
fmt.Fprintf(writer, "running shards %s (event budget=%d, runtime=%s, refresh=%s)…\n",
formatShardLabel(shards), *eventBudget, *runtime, *cadence)
announcedLoad := false
runPass := func() error {
// When -refresh is set we MUST cap each pass; otherwise
// drainShardEvents blocks until the outer runtime ctx
// expires and the loop's ticker never fires. With cadence=0
// (one-shot) the pass uses the full ctx and returns when
// the runtime cap hits.
passCtx := ctx
if *cadence > 0 {
var passCancel context.CancelFunc
// Pass budget = cadence + grace. Grace gives the
// drain time to actually process events that arrived
// during the cadence window.
passCtx, passCancel = context.WithTimeout(ctx, *cadence+5*time.Second)
defer passCancel()
}
eng := engine.New()
inputs, parseErrors, err := scheduler.LoadCompileInputs(ctx, filerClient, bucketsPath)
inputs, parseErrors, err := scheduler.LoadCompileInputs(passCtx, filerClient, bucketsPath)
if err != nil {
return fmt.Errorf("load lifecycle configs: %w", err)
}
@ -140,6 +155,10 @@ func (c *commandS3LifecycleRunShard) Do(args []string, env *CommandEnv, writer i
if len(inputs) == 0 {
return nil
}
if !announcedLoad {
fmt.Fprintf(writer, "loaded lifecycle for %d bucket(s)\n", len(inputs))
announcedLoad = true
}
buckets := make([]string, 0, len(inputs))
for _, in := range inputs {
if in.Bucket != "" {
@ -149,7 +168,7 @@ func (c *commandS3LifecycleRunShard) Do(args []string, env *CommandEnv, writer i
walker := dailyrun.WalkerFunc(func(walkCtx context.Context, view *engine.Snapshot, shardID int) error {
return dailyrun.WalkBuckets(walkCtx, view, shardID, buckets, listFn, walkerDispatch)
})
return dailyrun.Run(ctx, dailyrun.Config{
return dailyrun.Run(passCtx, dailyrun.Config{
Shards: shards,
BucketsPath: bucketsPath,
Engine: eng,
@ -157,6 +176,11 @@ func (c *commandS3LifecycleRunShard) Do(args []string, env *CommandEnv, writer i
Client: client,
Persister: &dailyrun.FilerCursorPersister{Store: dispatcher.NewFilerStoreClient(filerClient)},
Lister: dispatcher.NewFilerSiblingLister(filerClient, bucketsPath),
// The shell command is used for bounded one-shot sweeps in
// integration tests and CI. Fan out across the selected shards
// so recovery walks do not serialize 16 shard scans into a 10s
// timeout budget.
Workers: len(shards),
Walker: walker,
EventBudget: *eventBudget,
ClientName: fmt.Sprintf("shell-lifecycle-%s", formatShardLabel(shards)),