fix: order bucket deletion and metadata publication safely

Signed-off-by: Feng Ruohang <rh@vonng.com>

cmd/bucket-metadata-publication_test.go

@@ -0,0 +1,201 @@
+// Copyright 2026 PGSTY contributors.
+// SPDX-License-Identifier: AGPL-3.0-or-later
+
+package cmd
+
+import (
+	"bytes"
+	"context"
+	"encoding/json"
+	"errors"
+	"fmt"
+	"io"
+	"net/http"
+	"reflect"
+	"sync"
+	"testing"
+	"time"
+
+	"github.com/minio/madmin-go/v3"
+	"github.com/minio/minio/internal/auth"
+	"github.com/minio/minio/internal/grid"
+)
+
+func TestDeleteBucketMetadataLockCancellation(t *testing.T) {
+	defer DetectTestLeak(t)()
+	ExecObjectLayerAPITest(ExecObjectLayerAPITestArgs{t: t, objAPITest: testDeleteBucketMetadataLockCancellation})
+}
+
+func testDeleteBucketMetadataLockCancellation(obj ObjectLayer, instanceType, bucket string, _ http.Handler, _ auth.Credentials, t *testing.T) {
+	_, unlock, err := lockBucketMetadata(t.Context(), obj, bucket)
+	if err != nil {
+		t.Fatal(err)
+	}
+	release := sync.OnceFunc(unlock)
+	defer release()
+	ctx, cancel := context.WithTimeout(t.Context(), 250*time.Millisecond)
+	defer cancel()
+	done := make(chan error, 1)
+	go func() { done <- obj.DeleteBucket(ctx, bucket, DeleteBucketOptions{Force: true, NoLock: true}) }()
+	<-ctx.Done()
+	// A metadata-lock failure must happen before the destructive operation.
+	if _, err := obj.GetBucketInfo(t.Context(), bucket, BucketOptions{}); err != nil {
+		t.Errorf("%s: bucket disappeared while metadata.lock was unavailable: %v", instanceType, err)
+	}
+	release()
+	if err := <-done; err == nil {
+		t.Errorf("%s: canceled deletion succeeded", instanceType)
+	}
+	if _, err := readBucketMetadata(t.Context(), obj, bucket); err != nil {
+		t.Errorf("%s: canceled deletion removed metadata: %v", instanceType, err)
+	}
+}
+
+func TestQueuedMetadataUpdateAfterDelete(t *testing.T) {
+	defer DetectTestLeak(t)()
+	for _, expiry := range []bool{false, true} {
+		t.Run(fmt.Sprintf("expiry=%v", expiry), func(t *testing.T) {
+			ExecObjectLayerAPITest(ExecObjectLayerAPITestArgs{t: t, objAPITest: func(obj ObjectLayer, instanceType, bucket string, _ http.Handler, _ auth.Credentials, t *testing.T) {
+				previous := newObjectLayerFn()
+				barrier := &lcMergeBarrier{ObjectLayer: obj, bucket: bucket, mAtLock: make(chan struct{}), mProceed: make(chan struct{})}
+				setObjectLayer(barrier)
+				defer setObjectLayer(previous)
+				release := sync.OnceFunc(func() { close(barrier.mProceed) })
+				defer release()
+				ctx, cancel := context.WithTimeout(t.Context(), 10*time.Second)
+				defer cancel()
+				ctx = context.WithValue(ctx, lcMergeWriterKey{}, "M")
+				done := make(chan error, 1)
+				go func() {
+					if expiry {
+						done <- globalBucketMetadataSys.UpdateExpiryLCConfig(ctx, bucket, nil, UTCNow())
+						return
+					}
+					_, err := globalBucketMetadataSys.Update(ctx, bucket, bucketTaggingConfig, []byte(`<Tagging><TagSet/></Tagging>`))
+					done <- err
+				}()
+				select {
+				case <-barrier.mAtLock:
+				case <-ctx.Done():
+					t.Fatal("writer did not reach metadata.lock")
+				}
+				if err := obj.DeleteBucket(t.Context(), bucket, DeleteBucketOptions{Force: true}); err != nil {
+					t.Fatal(err)
+				}
+				release()
+				if err := <-done; !isErrBucketNotFound(err) {
+					t.Errorf("%s: queued update should reject a deleted bucket, got %v", instanceType, err)
+				}
+				if _, err := readBucketMetadata(t.Context(), obj, bucket); !errors.Is(err, errConfigNotFound) && !isErrBucketNotFound(err) {
+					t.Errorf("%s: queued update recreated metadata: %v", instanceType, err)
+				}
+			}})
+		})
+	}
+}
+
+// Pause the first actual peer-handler read, without replacing the handler or
+// requiring it to accept a test-only context.
+type peerMetadataReadBarrier struct {
+	ObjectLayer
+	bucket           string
+	once             sync.Once
+	reading, release chan struct{}
+}
+
+func (o *peerMetadataReadBarrier) GetObjectNInfo(ctx context.Context, bucket, object string, rs *HTTPRangeSpec, h http.Header, opts ObjectOptions) (*GetObjectReader, error) {
+	first := false
+	if bucket == minioMetaBucket && object == pathJoin(bucketMetaPrefix, o.bucket, bucketMetadataFile) {
+		o.once.Do(func() { first = true })
+	}
+	gr, err := o.ObjectLayer.GetObjectNInfo(ctx, bucket, object, rs, h, opts)
+	if err != nil || !first {
+		return gr, err
+	}
+	data, err := io.ReadAll(gr)
+	oi := gr.ObjInfo
+	gr.Close()
+	if err != nil {
+		return nil, err
+	}
+	close(o.reading)
+	select {
+	case <-o.release:
+	case <-ctx.Done():
+		return nil, ctx.Err()
+	}
+	return NewGetObjectReaderFromReader(bytes.NewReader(data), oi, opts)
+}
+
+func TestPeerMetadataReloadPreservesCurrentTargets(t *testing.T) {
+	defer DetectTestLeak(t)()
+	ExecObjectLayerAPITest(ExecObjectLayerAPITestArgs{t: t, objAPITest: testPeerMetadataReloadPreservesCurrentTargets})
+}
+
+func testPeerMetadataReloadPreservesCurrentTargets(obj ObjectLayer, instanceType, bucket string, _ http.Handler, _ auth.Credentials, t *testing.T) {
+	seed := func(revision string) {
+		t.Helper()
+		notification := []byte(`<NotificationConfiguration><QueueConfiguration><Id>` + revision + `</Id><Queue>arn:minio:sqs::` + revision + `:webhook</Queue><Event>s3:ObjectCreated:*</Event></QueueConfiguration></NotificationConfiguration>`)
+		targets, err := json.Marshal(madmin.BucketTargets{Targets: []madmin.BucketTarget{{
+			SourceBucket: bucket, TargetBucket: bucket, Endpoint: "127.0.0.1:9000", Arn: revision,
+			Credentials: &madmin.Credentials{AccessKey: "fixture", SecretKey: "fixture-secret"},
+		}}})
+		if err != nil {
+			t.Fatal(err)
+		}
+		for _, config := range []struct {
+			name string
+			data []byte
+		}{{bucketNotificationConfig, notification}, {bucketTargetsFile, targets}} {
+			if _, err := globalBucketMetadataSys.Update(t.Context(), bucket, config.name, config.data); err != nil {
+				t.Fatal(err)
+			}
+		}
+	}
+	reload := func() error {
+		args := grid.MSS{peerRESTBucket: bucket}
+		_, err := (&peerRESTServer{}).LoadBucketMetadataHandler(&args)
+		if err != nil {
+			return err
+		}
+		return nil
+	}
+	seed("old")
+	previous := newObjectLayerFn()
+	barrier := &peerMetadataReadBarrier{ObjectLayer: obj, bucket: bucket, reading: make(chan struct{}), release: make(chan struct{})}
+	setObjectLayer(barrier)
+	defer setObjectLayer(previous)
+	release := sync.OnceFunc(func() { close(barrier.release) })
+	defer release()
+	done := make(chan error, 1)
+	go func() { done <- reload() }()
+	select {
+	case <-barrier.reading:
+	case <-time.After(10 * time.Second):
+		t.Fatal("peer handler did not read metadata")
+	}
+	seed("new")
+	if err := reload(); err != nil {
+		t.Fatal(err)
+	}
+	release()
+	if err := <-done; err != nil {
+		t.Fatal(err)
+	}
+	meta, err := globalBucketMetadataSys.Get(bucket)
+	if err != nil {
+		t.Fatal(err)
+	}
+	globalEventNotifier.RLock()
+	rules := globalEventNotifier.bucketRulesMap[bucket].Clone()
+	globalEventNotifier.RUnlock()
+	if !reflect.DeepEqual(rules, meta.notificationConfig.ToRulesMap()) {
+		t.Errorf("%s: stale peer reload replaced current notification rules", instanceType)
+	}
+	globalBucketTargetSys.RLock()
+	targets := append([]madmin.BucketTarget(nil), globalBucketTargetSys.targetsMap[bucket]...)
+	globalBucketTargetSys.RUnlock()
+	if len(targets) != 1 || targets[0].Arn != "new" {
+		t.Errorf("%s: stale peer reload replaced current replication targets: %+v", instanceType, targets)
+	}
+}

cmd/bucket-metadata-sys.go

@@ -125,13 +125,10 @@ func (sys *BucketMetadataSys) Set(bucket string, meta BucketMetadata) {
 	}
 }
 
-// setReloaded publishes a record freshly loaded from disk into the resident
-// cache without letting an older on-disk revision overwrite a newer resident
-// one. Overlapping peer reloads (LoadBucketMetadataHandler) and cache-miss
-// loads can finish out of order, so an unconditional Set can leave the resident
-// cache a revision behind until the next refresh (issue #105). The authoritative
-// read-modify-write save path uses Set directly and always wins because it
-// stamps a fresh updatedAt. Only a shallow copy is stored, exactly like Set.
+// setReloaded publishes the newest known revision and its derived registries
+// together. An old reload may finish after a local save or another reload;
+// applying its notification/replication targets would regress live behavior
+// even if the metadata cache itself rejected that old revision.
 func (sys *BucketMetadataSys) setReloaded(bucket string, meta BucketMetadata) {
 	if isMinioMetaBucketName(bucket) {
 		return
@@ -139,12 +136,21 @@ func (sys *BucketMetadataSys) setReloaded(bucket string, meta BucketMetadata) {
 	sys.Lock()
 	defer sys.Unlock()
 	if cur, ok := sys.metadataMap[bucket]; ok && !cur.lastUpdate().Before(meta.lastUpdate()) {
-		// A resident revision that is at least as new is already published; do
-		// not regress it to the older reload.
-		return
+		meta = cur
+	} else {
+		sys.metadataMap[bucket] = meta
 	}
-	sys.metadataMap[bucket] = meta
 	sys.clearLoadFailure(bucket)
+	// These registry updates only change local state; no peer/network I/O runs
+	// under the metadata mutex. Keep publication ordered against Set/Remove.
+	if globalEventNotifier != nil {
+		if meta.notificationConfig != nil {
+			globalEventNotifier.AddRulesMap(bucket, meta.notificationConfig.ToRulesMap())
+		} else {
+			globalEventNotifier.RemoveNotification(bucket)
+		}
+	}
+	globalBucketTargetSys.UpdateAllTargets(bucket, meta.bucketTargetConfig)
 }
 
 func (sys *BucketMetadataSys) updateAndParse(ctx context.Context, bucket string, configFile string, configData []byte, parse, lifecycleDelete bool) (updatedAt time.Time, err error) {
@@ -251,6 +257,11 @@ func (sys *BucketMetadataSys) save(ctx context.Context, meta BucketMetadata) err
 // saveMetadata persists and publishes metadata locally. Callers performing a
 // read-modify-write must hold metadata.lock and release it before peer fan-out.
 func (sys *BucketMetadataSys) saveMetadata(ctx context.Context, objAPI ObjectLayer, meta BucketMetadata) error {
+	// A writer may have queued for metadata.lock before DeleteBucket completed.
+	// Recheck the physical bucket under that lock, before recreating metadata.
+	if _, err := objAPI.GetBucketInfo(ctx, meta.Name, BucketOptions{NoMetadata: true}); err != nil {
+		return err
+	}
 	if err := meta.Save(ctx, objAPI); err != nil {
 		return err
 	}
@@ -782,8 +793,6 @@ func (sys *BucketMetadataSys) refreshBucketsMetadataLoop(ctx context.Context) {
 				wait := sleeper.Timer(ctx)
 
 				bucket := buckets[i].Name
-				updated := false
-
 				meta, err := loadBucketMetadata(ctx, sys.objAPI, bucket)
 				if err != nil {
 					internalLogIf(ctx, err, logger.WarningKind)
@@ -794,19 +803,7 @@ func (sys *BucketMetadataSys) refreshBucketsMetadataLoop(ctx context.Context) {
 					continue
 				}
 
-				sys.Lock()
-				// Update if the bucket metadata in the memory is older than on-disk one
-				if lu := sys.metadataMap[bucket].lastUpdate(); lu.Before(meta.lastUpdate()) {
-					updated = true
-					sys.metadataMap[bucket] = meta
-				}
-				sys.clearLoadFailure(bucket)
-				sys.Unlock()
-
-				if updated {
-					globalEventNotifier.set(bucket, meta)
-					globalBucketTargetSys.set(bucket, meta)
-				}
+				sys.setReloaded(bucket, meta)
 
 				wait() // wait to proceed to next entry.
 			}

cmd/bucket-metadata.go

@@ -303,6 +303,9 @@ func loadBucketMetadataParseUnderLock(ctx context.Context, objectAPI ObjectLayer
 		return newBucketMetadata(bucket), fmt.Errorf("%w: %v", errBucketMetadataMigrationLockUnavailable, err)
 	}
 	defer unlock()
+	if _, err := objectAPI.GetBucketInfo(ctx, bucket, BucketOptions{NoMetadata: true}); err != nil {
+		return newBucketMetadata(bucket), err
+	}
 	return loadBucketMetadataParse(ctx, objectAPI, bucket, parse)
 }
 

cmd/erasure-server-pool.go

@@ -2195,7 +2195,15 @@ func (z *erasureServerPools) DeleteBucket(ctx context.Context, bucket string, op
 		opts.Force = true
 	}
 
-	err := z.s3Peer.DeleteBucket(ctx, bucket, opts)
+	// Take the metadata writer lock before deleting anything. Failure or
+	// cancellation must leave both the bucket and its metadata intact.
+	ctx, unlock, err := lockBucketMetadata(ctx, z, bucket)
+	if err != nil {
+		return toObjectErr(err, bucket)
+	}
+	defer unlock()
+
+	err = z.s3Peer.DeleteBucket(ctx, bucket, opts)
 	if err == nil || isErrBucketNotFound(err) {
 		// If site replication is configured, hold on to deleted bucket state until sites sync
 		if opts.SRDeleteOp == MarkDelete {
@@ -2204,18 +2212,9 @@ func (z *erasureServerPools) DeleteBucket(ctx context.Context, bucket string, op
 	}
 
 	if err == nil {
-		// Purge the entire bucket metadata entirely. Hold metadata.lock across
-		// the purge so a bucket metadata writer that is mid-save cannot
-		// resurrect a ghost .metadata.bin after the prefix has been removed
-		// (issue #105). Lock order stays <bucket>.lck -> metadata.lock; a lock
-		// acquisition failure falls back to a best-effort unlocked purge so a
-		// delete is never blocked from completing.
-		if lctx, unlock, lerr := lockBucketMetadata(context.Background(), z, bucket); lerr == nil {
-			z.deleteAll(lctx, minioMetaBucket, pathJoin(bucketMetaPrefix, bucket))
-			unlock()
-		} else {
-			z.deleteAll(context.Background(), minioMetaBucket, pathJoin(bucketMetaPrefix, bucket))
-		}
+		// Finish cleanup after a committed delete even if the client disconnects.
+		// Both the bucket-name and metadata locks remain held until return.
+		z.deleteAll(context.WithoutCancel(ctx), minioMetaBucket, pathJoin(bucketMetaPrefix, bucket))
 	}
 
 	return toObjectErr(err, bucket)

cmd/peer-rest-server.go

@@ -544,18 +544,9 @@ func (s *peerRESTServer) LoadBucketMetadataHandler(mss *grid.MSS) (np grid.NoPay
 		return np, grid.NewRemoteErr(err)
 	}
 
-	// Publish monotonically: an overlapping reload that read an older revision
-	// must not overwrite a newer resident record (issue #105).
+	// Publish the metadata and derived registries from the same current revision.
 	globalBucketMetadataSys.setReloaded(bucketName, meta)
 
-	if meta.notificationConfig != nil {
-		globalEventNotifier.AddRulesMap(bucketName, meta.notificationConfig.ToRulesMap())
-	}
-
-	if meta.bucketTargetConfig != nil {
-		globalBucketTargetSys.UpdateAllTargets(bucketName, meta.bucketTargetConfig)
-	}
-
 	return np, nerr
 }