etcdutl: Fix snapshot restore memory alloc issue

When running the snapshot command, allow receiving an initial memory map
allocation for the database, avoiding future memory allocation issues.

Co-authored-by: Benjamin Wang <benjamin.wang@broadcom.com>
Signed-off-by: Fatih USTA <fatihusta86@gmail.com>
Signed-off-by: Ivan Valdes <ivan@vald.es>

etcdutl/etcdutl/migrate_command.go

@@ -92,7 +92,7 @@ func (o *migrateOptions) Config() (*migrateConfig, error) {
 	}
 
 	dbPath := datadir.ToBackendFileName(o.dataDir)
-	c.be = backend.NewDefaultBackend(GetLogger(), dbPath)
+	c.be = backend.NewDefaultBackend(backend.BackendConfig{Logger: GetLogger(), Path: dbPath})
 
 	walPath := datadir.ToWalDir(o.dataDir)
 	w, err := wal.OpenForRead(c.lg, walPath, walpb.Snapshot{})

etcdutl/etcdutl/snapshot_command.go

@@ -38,6 +38,7 @@ var (
 	restorePeerURLs     string
 	restoreName         string
 	skipHashCheck       bool
+	initialMmapSize     = uint64(10 * 1024 * 1024 * 1024)
 	markCompacted       bool
 	revisionBump        uint64
 )
@@ -77,6 +78,7 @@ func NewSnapshotRestoreCommand() *cobra.Command {
 	cmd.Flags().StringVar(&restorePeerURLs, "initial-advertise-peer-urls", defaultInitialAdvertisePeerURLs, "List of this member's peer URLs to advertise to the rest of the cluster")
 	cmd.Flags().StringVar(&restoreName, "name", defaultName, "Human-readable name for this member")
 	cmd.Flags().BoolVar(&skipHashCheck, "skip-hash-check", false, "Ignore snapshot integrity hash value (required if copied from data directory)")
+	cmd.Flags().Uint64Var(&initialMmapSize, "initial-memory-map-size", initialMmapSize, "Initial memory map allocation size(bytes) for the DB. It uses the default value if not defined or defined to 0")
 	cmd.Flags().Uint64Var(&revisionBump, "bump-revision", 0, "How much to increase the latest revision after restore")
 	cmd.Flags().BoolVar(&markCompacted, "mark-compacted", false, "Mark the latest revision after restore as the point of scheduled compaction (required if --bump-revision > 0, disallowed otherwise)")
 
@@ -104,7 +106,7 @@ func SnapshotStatusCommandFunc(cmd *cobra.Command, args []string) {
 
 func snapshotRestoreCommandFunc(_ *cobra.Command, args []string) {
 	SnapshotRestoreCommandFunc(restoreCluster, restoreClusterToken, restoreDataDir, restoreWalDir,
-		restorePeerURLs, restoreName, skipHashCheck, revisionBump, markCompacted, args)
+		restorePeerURLs, restoreName, skipHashCheck, initialMmapSize, revisionBump, markCompacted, args)
 }
 
 func SnapshotRestoreCommandFunc(restoreCluster string,
@@ -114,6 +116,7 @@ func SnapshotRestoreCommandFunc(restoreCluster string,
 	restorePeerURLs string,
 	restoreName string,
 	skipHashCheck bool,
+	initialMmapSize uint64,
 	revisionBump uint64,
 	markCompacted bool,
 	args []string) {
@@ -149,6 +152,7 @@ func SnapshotRestoreCommandFunc(restoreCluster string,
 		InitialCluster:      restoreCluster,
 		InitialClusterToken: restoreClusterToken,
 		SkipHashCheck:       skipHashCheck,
+		InitialMmapSize:     initialMmapSize,
 		RevisionBump:        revisionBump,
 		MarkCompacted:       markCompacted,
 	}); err != nil {

etcdutl/snapshot/v3_snapshot.go

@@ -83,7 +83,8 @@ type v3Manager struct {
 	snapDir   string
 	cl        *membership.RaftCluster
 
-	skipHashCheck bool
+	skipHashCheck   bool
+	initialMmapSize uint64
 }
 
 // hasChecksum returns "true" if the file size "n"
@@ -204,6 +205,9 @@ type RestoreConfig struct {
 	// (required if copied from data directory).
 	SkipHashCheck bool
 
+	// InitialMmapSize is the database initial memory map size.
+	InitialMmapSize uint64
+
 	// RevisionBump is the amount to increase the latest revision after restore,
 	// to allow administrators to trick clients into thinking that revision never decreased.
 	// If 0, revision bumping is skipped.
@@ -233,6 +237,7 @@ func (s *v3Manager) Restore(cfg RestoreConfig) error {
 		PeerURLs:            pURLs,
 		InitialPeerURLsMap:  ics,
 		InitialClusterToken: cfg.InitialClusterToken,
+		InitialMmapSize:     cfg.InitialMmapSize,
 	}
 	if err = srv.VerifyBootstrap(); err != nil {
 		return err
@@ -263,13 +268,15 @@ func (s *v3Manager) Restore(cfg RestoreConfig) error {
 	s.walDir = walDir
 	s.snapDir = filepath.Join(dataDir, "member", "snap")
 	s.skipHashCheck = cfg.SkipHashCheck
+	s.initialMmapSize = cfg.InitialMmapSize
 
 	s.lg.Info(
 		"restoring snapshot",
 		zap.String("path", s.srcDbPath),
 		zap.String("wal-dir", s.walDir),
 		zap.String("data-dir", dataDir),
 		zap.String("snap-dir", s.snapDir),
+		zap.Uint64("initial-memory-map-size", s.initialMmapSize),
 	)
 
 	if err = s.saveDB(); err != nil {
@@ -297,6 +304,7 @@ func (s *v3Manager) Restore(cfg RestoreConfig) error {
 		zap.String("wal-dir", s.walDir),
 		zap.String("data-dir", dataDir),
 		zap.String("snap-dir", s.snapDir),
+		zap.Uint64("initial-memory-map-size", s.initialMmapSize),
 	)
 
 	return verify.VerifyIfEnabled(verify.Config{
@@ -317,7 +325,7 @@ func (s *v3Manager) saveDB() error {
 		return err
 	}
 
-	be := backend.NewDefaultBackend(s.lg, s.outDbPath())
+	be := backend.NewDefaultBackend(backend.BackendConfig{Logger: s.lg, Path: s.outDbPath(), MmapSize: s.initialMmapSize})
 	defer be.Close()
 
 	err = schema.NewMembershipBackend(s.lg, be).TrimMembershipFromBackend()
@@ -331,7 +339,7 @@ func (s *v3Manager) saveDB() error {
 // modifyLatestRevision can increase the latest revision by the given amount and sets the scheduled compaction
 // to that revision so that the server will consider this revision compacted.
 func (s *v3Manager) modifyLatestRevision(bumpAmount uint64) error {
-	be := backend.NewDefaultBackend(s.lg, s.outDbPath())
+	be := backend.NewDefaultBackend(backend.BackendConfig{Logger: s.lg, Path: s.outDbPath()})
 	defer func() {
 		be.ForceCommit()
 		be.Close()
@@ -471,8 +479,8 @@ func (s *v3Manager) saveWALAndSnap() (*raftpb.HardState, error) {
 		return nil, err
 	}
 
-	// add members again to persist them to the backend we create.
-	be := backend.NewDefaultBackend(s.lg, s.outDbPath())
+	// add members again to persist them to the store we create.
+	be := backend.NewDefaultBackend(backend.BackendConfig{Logger: s.lg, Path: s.outDbPath(), MmapSize: s.initialMmapSize})
 	defer be.Close()
 	s.cl.SetBackend(schema.NewMembershipBackend(s.lg, be))
 	for _, m := range s.cl.Members() {
@@ -551,7 +559,7 @@ func (s *v3Manager) saveWALAndSnap() (*raftpb.HardState, error) {
 }
 
 func (s *v3Manager) updateCIndex(commit uint64, term uint64) error {
-	be := backend.NewDefaultBackend(s.lg, s.outDbPath())
+	be := backend.NewDefaultBackend(backend.BackendConfig{Logger: s.lg, Path: s.outDbPath(), MmapSize: s.initialMmapSize})
 	defer be.Close()
 
 	cindex.UpdateConsistentIndexForce(be.BatchTx(), commit, term)

server/config/config.go

@@ -119,6 +119,7 @@ type ServerConfig struct {
 	CompactionBatchLimit    int
 	CompactionSleepInterval time.Duration
 	QuotaBackendBytes       int64
+	InitialMmapSize         uint64
 	MaxTxnOps               uint
 
 	// MaxRequestBytes is the maximum request size to send over raft.

server/etcdserver/cindex/cindex_test.go

@@ -56,7 +56,7 @@ func TestConsistentIndex(t *testing.T) {
 	be.ForceCommit()
 	be.Close()
 
-	b := backend.NewDefaultBackend(zaptest.NewLogger(t), tmpPath)
+	b := backend.NewDefaultBackend(backend.BackendConfig{Logger: zaptest.NewLogger(t), Path: tmpPath})
 	defer b.Close()
 	ci.SetBackend(b)
 	index = ci.ConsistentIndex()
@@ -108,7 +108,7 @@ func TestConsistentIndexDecrease(t *testing.T) {
 			be.ForceCommit()
 			be.Close()
 
-			be = backend.NewDefaultBackend(zaptest.NewLogger(t), tmpPath)
+			be = backend.NewDefaultBackend(backend.BackendConfig{Logger: zaptest.NewLogger(t), Path: tmpPath})
 			defer be.Close()
 			ci := NewConsistentIndex(be)
 			ci.SetConsistentIndex(tc.index, tc.term)

server/storage/backend/backend.go

@@ -164,9 +164,17 @@ func New(bcfg BackendConfig) Backend {
 	return newBackend(bcfg)
 }
 
-func NewDefaultBackend(lg *zap.Logger, path string) Backend {
-	bcfg := DefaultBackendConfig(lg)
-	bcfg.Path = path
+func NewDefaultBackend(bc BackendConfig) Backend {
+
+	bcfg := DefaultBackendConfig(bc.Logger)
+	bcfg.Path = bc.Path
+
+	if bc.MmapSize > 0 {
+		bcfg.MmapSize = bc.MmapSize
+	} else {
+		bcfg.MmapSize = initialMmapSize
+	}
+
 	return newBackend(bcfg)
 }
 

server/storage/mvcc/store_test.go

@@ -61,7 +61,7 @@ func TestScheduledCompact(t *testing.T) {
 			be.ForceCommit()
 			be.Close()
 
-			b := backend.NewDefaultBackend(lg, tmpPath)
+			b := backend.NewDefaultBackend(backend.BackendConfig{Logger: lg, Path: tmpPath})
 			defer b.Close()
 			v, found := UnsafeReadScheduledCompact(b.BatchTx())
 			assert.Equal(t, true, found)
@@ -103,7 +103,7 @@ func TestFinishedCompact(t *testing.T) {
 			be.ForceCommit()
 			be.Close()
 
-			b := backend.NewDefaultBackend(lg, tmpPath)
+			b := backend.NewDefaultBackend(backend.BackendConfig{Logger: lg, Path: tmpPath})
 			defer b.Close()
 			v, found := UnsafeReadFinishedCompact(b.BatchTx())
 			assert.Equal(t, true, found)

server/storage/schema/auth_roles_test.go

@@ -122,7 +122,7 @@ func TestGetAllRoles(t *testing.T) {
 			abe.ForceCommit()
 			be.Close()
 
-			be2 := backend.NewDefaultBackend(lg, tmpPath)
+			be2 := backend.NewDefaultBackend(backend.BackendConfig{Logger: lg, Path: tmpPath})
 			defer be2.Close()
 			abe2 := NewAuthBackend(lg, be2)
 			users := abe2.GetAllRoles()
@@ -219,7 +219,7 @@ func TestGetRole(t *testing.T) {
 			abe.ForceCommit()
 			be.Close()
 
-			be2 := backend.NewDefaultBackend(lg, tmpPath)
+			be2 := backend.NewDefaultBackend(backend.BackendConfig{Logger: lg, Path: tmpPath})
 			defer be2.Close()
 			abe2 := NewAuthBackend(lg, be2)
 			users := abe2.GetRole("role1")

server/storage/schema/auth_test.go

@@ -66,7 +66,7 @@ func TestAuthEnabled(t *testing.T) {
 			abe.ForceCommit()
 			be.Close()
 
-			be2 := backend.NewDefaultBackend(lg, tmpPath)
+			be2 := backend.NewDefaultBackend(backend.BackendConfig{Logger: lg, Path: tmpPath})
 			defer be2.Close()
 			abe2 := NewAuthBackend(lg, be2)
 			tx = abe2.BatchTx()
@@ -117,7 +117,7 @@ func TestAuthRevision(t *testing.T) {
 			abe.ForceCommit()
 			be.Close()
 
-			be2 := backend.NewDefaultBackend(lg, tmpPath)
+			be2 := backend.NewDefaultBackend(backend.BackendConfig{Logger: lg, Path: tmpPath})
 			defer be2.Close()
 			abe2 := NewAuthBackend(lg, be2)
 			tx := abe2.BatchTx()

server/storage/schema/auth_users_test.go

@@ -110,7 +110,7 @@ func TestGetAllUsers(t *testing.T) {
 			abe.ForceCommit()
 			be.Close()
 
-			be2 := backend.NewDefaultBackend(lg, tmpPath)
+			be2 := backend.NewDefaultBackend(backend.BackendConfig{Logger: lg, Path: tmpPath})
 			defer be2.Close()
 			abe2 := NewAuthBackend(lg, be2)
 			users := abe2.ReadTx().UnsafeGetAllUsers()
@@ -195,7 +195,7 @@ func TestGetUser(t *testing.T) {
 			abe.ForceCommit()
 			be.Close()
 
-			be2 := backend.NewDefaultBackend(lg, tmpPath)
+			be2 := backend.NewDefaultBackend(backend.BackendConfig{Logger: lg, Path: tmpPath})
 			defer be2.Close()
 			abe2 := NewAuthBackend(lg, be2)
 			users := abe2.GetUser("alice")

server/storage/schema/lease_test.go

@@ -99,7 +99,7 @@ func TestLeaseBackend(t *testing.T) {
 			be.ForceCommit()
 			be.Close()
 
-			be2 := backend.NewDefaultBackend(lg, tmpPath)
+			be2 := backend.NewDefaultBackend(backend.BackendConfig{Logger: lg, Path: tmpPath})
 			defer be2.Close()
 			leases := MustUnsafeGetAllLeases(be2.ReadTx())
 

server/storage/schema/schema_test.go

@@ -83,7 +83,7 @@ func TestValidate(t *testing.T) {
 			lg := zap.NewNop()
 			dataPath := setupBackendData(t, tc.version, tc.overrideKeys)
 
-			b := backend.NewDefaultBackend(lg, dataPath)
+			b := backend.NewDefaultBackend(backend.BackendConfig{Logger: lg, Path: dataPath})
 			defer b.Close()
 			err := Validate(lg, b.ReadTx())
 			if (err != nil) != tc.expectError {
@@ -211,7 +211,7 @@ func TestMigrate(t *testing.T) {
 				t.Fatal(err)
 			}
 
-			b := backend.NewDefaultBackend(lg, dataPath)
+			b := backend.NewDefaultBackend(backend.BackendConfig{Logger: lg, Path: dataPath})
 			defer b.Close()
 
 			err = Migrate(lg, b.BatchTx(), walVersion, tc.targetVersion)
@@ -255,7 +255,7 @@ func TestMigrateIsReversible(t *testing.T) {
 			lg := zap.NewNop()
 			dataPath := setupBackendData(t, tc.initialVersion, nil)
 
-			be := backend.NewDefaultBackend(lg, dataPath)
+			be := backend.NewDefaultBackend(backend.BackendConfig{Logger: lg, Path: dataPath})
 			defer be.Close()
 			tx := be.BatchTx()
 			tx.Lock()

server/storage/schema/version_test.go

@@ -69,7 +69,7 @@ func TestVersion(t *testing.T) {
 			be.ForceCommit()
 			be.Close()
 
-			b := backend.NewDefaultBackend(lg, tmpPath)
+			b := backend.NewDefaultBackend(backend.BackendConfig{Logger: lg, Path: tmpPath})
 			defer b.Close()
 			v := UnsafeReadStorageVersion(b.BatchTx())
 

server/verify/verify.go

@@ -74,8 +74,7 @@ func Verify(cfg Config) error {
 			lg.Info("verification of persisted state successful", zap.String("data-dir", cfg.DataDir))
 		}
 	}()
-
-	be := backend.NewDefaultBackend(lg, datadir.ToBackendFileName(cfg.DataDir))
+	be := backend.NewDefaultBackend(backend.BackendConfig{Logger: lg, Path: datadir.ToBackendFileName(cfg.DataDir)})
 	defer be.Close()
 
 	snapshot, hardstate, err := validateWal(cfg)

tests/e2e/utl_migrate_test.go

@@ -166,7 +166,7 @@ func TestEtctlutlMigrate(t *testing.T) {
 			}
 
 			t.Log("etcdutl migrate...")
-			be := backend.NewDefaultBackend(lg, filepath.Join(memberDataDir, "member/snap/db"))
+			be := backend.NewDefaultBackend(backend.BackendConfig{Logger: lg, Path: filepath.Join(memberDataDir, "member/snap/db")})
 			defer be.Close()
 
 			ver := schema.ReadStorageVersion(be.ReadTx())

tests/integration/clientv3/maintenance_test.go

@@ -277,7 +277,7 @@ func testMaintenanceSnapshotErrorInflight(t *testing.T, snapshot func(context.Co
 	// take about 1-second to read snapshot
 	clus.Members[0].Stop(t)
 	dpath := filepath.Join(clus.Members[0].DataDir, "member", "snap", "db")
-	b := backend.NewDefaultBackend(lg, dpath)
+	b := backend.NewDefaultBackend(backend.BackendConfig{Logger: lg, Path: dpath})
 	s := mvcc.NewStore(lg, b, &lease.FakeLessor{}, mvcc.StoreConfig{CompactionBatchLimit: math.MaxInt32})
 	rev := 100000
 	for i := 2; i <= rev; i++ {

tests/integration/v3_alarm_test.go

@@ -194,7 +194,7 @@ func TestV3CorruptAlarm(t *testing.T) {
 	// Corrupt member 0 by modifying backend offline.
 	clus.Members[0].Stop(t)
 	fp := filepath.Join(clus.Members[0].DataDir, "member", "snap", "db")
-	be := backend.NewDefaultBackend(lg, fp)
+	be := backend.NewDefaultBackend(backend.BackendConfig{Logger: lg, Path: fp})
 	s := mvcc.NewStore(lg, be, nil, mvcc.StoreConfig{})
 	// NOTE: cluster_proxy mode with namespacing won't set 'k', but namespace/'k'.
 	s.Put([]byte("abc"), []byte("def"), 0)

tools/etcd-dump-db/backend.go

@@ -179,7 +179,7 @@ func iterateBucket(dbPath, bucket string, limit uint64, decode bool) (err error)
 }
 
 func getHash(dbPath string) (hash uint32, err error) {
-	b := backend.NewDefaultBackend(zap.NewNop(), dbPath)
+	b := backend.NewDefaultBackend(backend.BackendConfig{Logger: zap.NewNop(), Path: dbPath})
 	return b.Hash(schema.DefaultIgnores)
 }