triedb/pathdb: introduce file-based state journal (#32060)

Introduce file-based state journal in path database, fixing
the Pebble restriction when the journal size exceeds 4GB.

---------

Signed-off-by: jsvisa <delweng@gmail.com>
Co-authored-by: Gary Rong <garyrong0905@gmail.com>

cmd/utils/flags.go

@@ -2198,6 +2198,12 @@ func MakeChain(ctx *cli.Context, stack *node.Node, readonly bool) (*core.BlockCh
 		StateHistory:   ctx.Uint64(StateHistoryFlag.Name),
 		// Disable transaction indexing/unindexing.
 		TxLookupLimit: -1,
+
+		// Enables file journaling for the trie database. The journal files will be stored
+		// within the data directory. The corresponding paths will be either:
+		// - DATADIR/triedb/merkle.journal
+		// - DATADIR/triedb/verkle.journal
+		TrieJournalDirectory: stack.ResolvePath("triedb"),
 	}
 	if options.ArchiveMode && !options.Preimages {
 		options.Preimages = true

core/blockchain.go

@@ -162,10 +162,11 @@ const (
 // BlockChainConfig contains the configuration of the BlockChain object.
 type BlockChainConfig struct {
 	// Trie database related options
-	TrieCleanLimit   int           // Memory allowance (MB) to use for caching trie nodes in memory
+	TrieCleanLimit       int           // Memory allowance (MB) to use for caching trie nodes in memory
-	TrieDirtyLimit   int           // Memory limit (MB) at which to start flushing dirty trie nodes to disk
+	TrieDirtyLimit       int           // Memory limit (MB) at which to start flushing dirty trie nodes to disk
-	TrieTimeLimit    time.Duration // Time limit after which to flush the current in-memory trie to disk
+	TrieTimeLimit        time.Duration // Time limit after which to flush the current in-memory trie to disk
-	TrieNoAsyncFlush bool          // Whether the asynchronous buffer flushing is disallowed
+	TrieNoAsyncFlush     bool          // Whether the asynchronous buffer flushing is disallowed
+	TrieJournalDirectory string        // Directory path to the journal used for persisting trie data across node restarts
 
 	Preimages    bool   // Whether to store preimage of trie key to the disk
 	StateHistory uint64 // Number of blocks from head whose state histories are reserved.
@@ -246,6 +247,7 @@ func (cfg *BlockChainConfig) triedbConfig(isVerkle bool) *triedb.Config {
 			EnableStateIndexing: cfg.ArchiveMode,
 			TrieCleanSize:       cfg.TrieCleanLimit * 1024 * 1024,
 			StateCleanSize:      cfg.SnapshotLimit * 1024 * 1024,
+			JournalDirectory:    cfg.TrieJournalDirectory,
 
 			// TODO(rjl493456442): The write buffer represents the memory limit used
 			// for flushing both trie data and state data to disk. The config name

core/rawdb/accessors_state.go

@@ -157,14 +157,6 @@ func WriteTrieJournal(db ethdb.KeyValueWriter, journal []byte) {
 	}
 }
 
-// DeleteTrieJournal deletes the serialized in-memory trie nodes of layers saved at
-// the last shutdown.
-func DeleteTrieJournal(db ethdb.KeyValueWriter) {
-	if err := db.Delete(trieJournalKey); err != nil {
-		log.Crit("Failed to remove tries journal", "err", err)
-	}
-}
-
 // ReadStateHistoryMeta retrieves the metadata corresponding to the specified
 // state history. Compute the position of state history in freezer by minus
 // one since the id of first state history starts from one(zero for initial

eth/backend.go

@@ -236,9 +236,13 @@ func New(stack *node.Node, config *ethconfig.Config) (*Ethereum, error) {
 			VmConfig: vm.Config{
 				EnablePreimageRecording: config.EnablePreimageRecording,
 			},
+			// Enables file journaling for the trie database. The journal files will be stored
+			// within the data directory. The corresponding paths will be either:
+			// - DATADIR/triedb/merkle.journal
+			// - DATADIR/triedb/verkle.journal
+			TrieJournalDirectory: stack.ResolvePath("triedb"),
 		}
 	)
-
 	if config.VMTrace != "" {
 		traceConfig := json.RawMessage("{}")
 		if config.VMTraceJsonConfig != "" {

triedb/pathdb/database.go

@@ -21,6 +21,7 @@ import (
 	"errors"
 	"fmt"
 	"io"
+	"path/filepath"
 	"sync"
 	"time"
 
@@ -120,6 +121,7 @@ type Config struct {
 	StateCleanSize      int    // Maximum memory allowance (in bytes) for caching clean state data
 	WriteBufferSize     int    // Maximum memory allowance (in bytes) for write buffer
 	ReadOnly            bool   // Flag whether the database is opened in read only mode
+	JournalDirectory    string // Absolute path of journal directory (null means the journal data is persisted in key-value store)
 
 	// Testing configurations
 	SnapshotNoBuild   bool // Flag Whether the state generation is allowed
@@ -156,6 +158,9 @@ func (c *Config) fields() []interface{} {
 	} else {
 		list = append(list, "history", fmt.Sprintf("last %d blocks", c.StateHistory))
 	}
+	if c.JournalDirectory != "" {
+		list = append(list, "journal-dir", c.JournalDirectory)
+	}
 	return list
 }
 
@@ -493,7 +498,6 @@ func (db *Database) Enable(root common.Hash) error {
 	// Drop the stale state journal in persistent database and
 	// reset the persistent state id back to zero.
 	batch := db.diskdb.NewBatch()
-	rawdb.DeleteTrieJournal(batch)
 	rawdb.DeleteSnapshotRoot(batch)
 	rawdb.WritePersistentStateID(batch, 0)
 	if err := batch.Write(); err != nil {
@@ -573,8 +577,6 @@ func (db *Database) Recover(root common.Hash) error {
 		// disk layer won't be accessible from outside.
 		db.tree.init(dl)
 	}
-	rawdb.DeleteTrieJournal(db.diskdb)
-
 	// Explicitly sync the key-value store to ensure all recent writes are
 	// flushed to disk. This step is crucial to prevent a scenario where
 	// recent key-value writes are lost due to an application panic, while
@@ -680,6 +682,20 @@ func (db *Database) modifyAllowed() error {
 	return nil
 }
 
+// journalPath returns the absolute path of journal for persisting state data.
+func (db *Database) journalPath() string {
+	if db.config.JournalDirectory == "" {
+		return ""
+	}
+	var fname string
+	if db.isVerkle {
+		fname = fmt.Sprintf("verkle.journal")
+	} else {
+		fname = fmt.Sprintf("merkle.journal")
+	}
+	return filepath.Join(db.config.JournalDirectory, fname)
+}
+
 // AccountHistory inspects the account history within the specified range.
 //
 // Start: State ID of the first history object for the query. 0 implies the first

triedb/pathdb/database_test.go

@@ -21,12 +21,16 @@ import (
 	"errors"
 	"fmt"
 	"math/rand"
+	"os"
+	"path/filepath"
+	"strconv"
 	"testing"
 
 	"github.com/ethereum/go-ethereum/common"
 	"github.com/ethereum/go-ethereum/core/rawdb"
 	"github.com/ethereum/go-ethereum/core/types"
 	"github.com/ethereum/go-ethereum/crypto"
+	"github.com/ethereum/go-ethereum/ethdb"
 	"github.com/ethereum/go-ethereum/internal/testrand"
 	"github.com/ethereum/go-ethereum/rlp"
 	"github.com/ethereum/go-ethereum/trie"
@@ -121,7 +125,7 @@ type tester struct {
 	snapStorages map[common.Hash]map[common.Hash]map[common.Hash][]byte // Keyed by the hash of account address and the hash of storage key
 }
 
-func newTester(t *testing.T, historyLimit uint64, isVerkle bool, layers int, enableIndex bool) *tester {
+func newTester(t *testing.T, historyLimit uint64, isVerkle bool, layers int, enableIndex bool, journalDir string) *tester {
 	var (
 		disk, _ = rawdb.Open(rawdb.NewMemoryDatabase(), rawdb.OpenOptions{Ancient: t.TempDir()})
 		db      = New(disk, &Config{
@@ -131,6 +135,7 @@ func newTester(t *testing.T, historyLimit uint64, isVerkle bool, layers int, ena
 			StateCleanSize:      256 * 1024,
 			WriteBufferSize:     256 * 1024,
 			NoAsyncFlush:        true,
+			JournalDirectory:    journalDir,
 		}, isVerkle)
 
 		obj = &tester{
@@ -466,7 +471,7 @@ func TestDatabaseRollback(t *testing.T) {
 	}()
 
 	// Verify state histories
-	tester := newTester(t, 0, false, 32, false)
+	tester := newTester(t, 0, false, 32, false, "")
 	defer tester.release()
 
 	if err := tester.verifyHistory(); err != nil {
@@ -500,7 +505,7 @@ func TestDatabaseRecoverable(t *testing.T) {
 	}()
 
 	var (
-		tester = newTester(t, 0, false, 12, false)
+		tester = newTester(t, 0, false, 12, false, "")
 		index  = tester.bottomIndex()
 	)
 	defer tester.release()
@@ -544,7 +549,7 @@ func TestDisable(t *testing.T) {
 		maxDiffLayers = 128
 	}()
 
-	tester := newTester(t, 0, false, 32, false)
+	tester := newTester(t, 0, false, 32, false, "")
 	defer tester.release()
 
 	stored := crypto.Keccak256Hash(rawdb.ReadAccountTrieNode(tester.db.diskdb, nil))
@@ -586,7 +591,7 @@ func TestCommit(t *testing.T) {
 		maxDiffLayers = 128
 	}()
 
-	tester := newTester(t, 0, false, 12, false)
+	tester := newTester(t, 0, false, 12, false, "")
 	defer tester.release()
 
 	if err := tester.db.Commit(tester.lastHash(), false); err != nil {
@@ -610,20 +615,25 @@ func TestCommit(t *testing.T) {
 }
 
 func TestJournal(t *testing.T) {
+	testJournal(t, "")
+	testJournal(t, filepath.Join(t.TempDir(), strconv.Itoa(rand.Intn(10000))))
+}
+
+func testJournal(t *testing.T, journalDir string) {
 	// Redefine the diff layer depth allowance for faster testing.
 	maxDiffLayers = 4
 	defer func() {
 		maxDiffLayers = 128
 	}()
 
-	tester := newTester(t, 0, false, 12, false)
+	tester := newTester(t, 0, false, 12, false, journalDir)
 	defer tester.release()
 
 	if err := tester.db.Journal(tester.lastHash()); err != nil {
 		t.Errorf("Failed to journal, err: %v", err)
 	}
 	tester.db.Close()
-	tester.db = New(tester.db.diskdb, nil, false)
+	tester.db = New(tester.db.diskdb, tester.db.config, false)
 
 	// Verify states including disk layer and all diff on top.
 	for i := 0; i < len(tester.roots); i++ {
@@ -640,13 +650,30 @@ func TestJournal(t *testing.T) {
 }
 
 func TestCorruptedJournal(t *testing.T) {
+	testCorruptedJournal(t, "", func(db ethdb.Database) {
+		// Mutate the journal in disk, it should be regarded as invalid
+		blob := rawdb.ReadTrieJournal(db)
+		blob[0] = 0xa
+		rawdb.WriteTrieJournal(db, blob)
+	})
+
+	directory := filepath.Join(t.TempDir(), strconv.Itoa(rand.Intn(10000)))
+	testCorruptedJournal(t, directory, func(_ ethdb.Database) {
+		f, _ := os.OpenFile(filepath.Join(directory, "merkle.journal"), os.O_WRONLY, 0644)
+		f.WriteAt([]byte{0xa}, 0)
+		f.Sync()
+		f.Close()
+	})
+}
+
+func testCorruptedJournal(t *testing.T, journalDir string, modifyFn func(database ethdb.Database)) {
 	// Redefine the diff layer depth allowance for faster testing.
 	maxDiffLayers = 4
 	defer func() {
 		maxDiffLayers = 128
 	}()
 
-	tester := newTester(t, 0, false, 12, false)
+	tester := newTester(t, 0, false, 12, false, journalDir)
 	defer tester.release()
 
 	if err := tester.db.Journal(tester.lastHash()); err != nil {
@@ -655,13 +682,10 @@ func TestCorruptedJournal(t *testing.T) {
 	tester.db.Close()
 	root := crypto.Keccak256Hash(rawdb.ReadAccountTrieNode(tester.db.diskdb, nil))
 
-	// Mutate the journal in disk, it should be regarded as invalid
+	modifyFn(tester.db.diskdb)
-	blob := rawdb.ReadTrieJournal(tester.db.diskdb)
-	blob[0] = 0xa
-	rawdb.WriteTrieJournal(tester.db.diskdb, blob)
 
 	// Verify states, all not-yet-written states should be discarded
-	tester.db = New(tester.db.diskdb, nil, false)
+	tester.db = New(tester.db.diskdb, tester.db.config, false)
 	for i := 0; i < len(tester.roots); i++ {
 		if tester.roots[i] == root {
 			if err := tester.verifyState(root); err != nil {
@@ -694,7 +718,7 @@ func TestTailTruncateHistory(t *testing.T) {
 		maxDiffLayers = 128
 	}()
 
-	tester := newTester(t, 10, false, 12, false)
+	tester := newTester(t, 10, false, 12, false, "")
 	defer tester.release()
 
 	tester.db.Close()

triedb/pathdb/fileutils_unix.go

@@ -0,0 +1,57 @@
+// Copyright 2025 The go-ethereum Authors
+// This file is part of the go-ethereum library.
+//
+// The go-ethereum library is free software: you can redistribute it and/or modify
+// it under the terms of the GNU Lesser General Public License as published by
+// the Free Software Foundation, either version 3 of the License, or
+// (at your option) any later version.
+//
+// The go-ethereum library is distributed in the hope that it will be useful,
+// but WITHOUT ANY WARRANTY; without even the implied warranty of
+// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+// GNU Lesser General Public License for more details.
+//
+// You should have received a copy of the GNU Lesser General Public License
+// along with the go-ethereum library. If not, see <http://www.gnu.org/licenses/>.
+
+//go:build !windows
+// +build !windows
+
+package pathdb
+
+import (
+	"errors"
+	"os"
+	"syscall"
+)
+
+func isErrInvalid(err error) bool {
+	if errors.Is(err, os.ErrInvalid) {
+		return true
+	}
+	// Go >= 1.8 returns *os.PathError instead
+	if patherr, ok := err.(*os.PathError); ok && patherr.Err == syscall.EINVAL {
+		return true
+	}
+	return false
+}
+
+func syncDir(name string) error {
+	// As per fsync manpage, Linux seems to expect fsync on directory, however
+	// some system don't support this, so we will ignore syscall.EINVAL.
+	//
+	// From fsync(2):
+	//   Calling fsync() does not necessarily ensure that the entry in the
+	//   directory containing the file has also reached disk. For that an
+	//   explicit fsync() on a file descriptor for the directory is also needed.
+	f, err := os.Open(name)
+	if err != nil {
+		return err
+	}
+	defer f.Close()
+
+	if err := f.Sync(); err != nil && !isErrInvalid(err) {
+		return err
+	}
+	return nil
+}

triedb/pathdb/fileutils_windows.go

@@ -0,0 +1,25 @@
+// Copyright 2025 The go-ethereum Authors
+// This file is part of the go-ethereum library.
+//
+// The go-ethereum library is free software: you can redistribute it and/or modify
+// it under the terms of the GNU Lesser General Public License as published by
+// the Free Software Foundation, either version 3 of the License, or
+// (at your option) any later version.
+//
+// The go-ethereum library is distributed in the hope that it will be useful,
+// but WITHOUT ANY WARRANTY; without even the implied warranty of
+// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+// GNU Lesser General Public License for more details.
+//
+// You should have received a copy of the GNU Lesser General Public License
+// along with the go-ethereum library. If not, see <http://www.gnu.org/licenses/>.
+
+//go:build windows
+// +build windows
+
+package pathdb
+
+func syncDir(name string) error {
+	// On Windows, fsync on directories is not supported
+	return nil
+}

triedb/pathdb/history_reader_test.go

@@ -126,7 +126,7 @@ func testHistoryReader(t *testing.T, historyLimit uint64) {
 	}()
 	//log.SetDefault(log.NewLogger(log.NewTerminalHandlerWithLevel(os.Stderr, log.LevelDebug, true)))
 
-	env := newTester(t, historyLimit, false, 64, true)
+	env := newTester(t, historyLimit, false, 64, true, "")
 	defer env.release()
 	waitIndexing(env.db)
 

triedb/pathdb/journal.go

@@ -21,6 +21,7 @@ import (
 	"errors"
 	"fmt"
 	"io"
+	"os"
 	"time"
 
 	"github.com/ethereum/go-ethereum/common"
@@ -31,6 +32,8 @@ import (
 	"github.com/ethereum/go-ethereum/rlp"
 )
 
+const tempJournalSuffix = ".tmp"
+
 var (
 	errMissJournal       = errors.New("journal not found")
 	errMissVersion       = errors.New("version not found")
@@ -51,11 +54,25 @@ const journalVersion uint64 = 3
 
 // loadJournal tries to parse the layer journal from the disk.
 func (db *Database) loadJournal(diskRoot common.Hash) (layer, error) {
-	journal := rawdb.ReadTrieJournal(db.diskdb)
+	var reader io.Reader
-	if len(journal) == 0 {
+	if path := db.journalPath(); path != "" && common.FileExist(path) {
-		return nil, errMissJournal
+		// If a journal file is specified, read it from there
+		log.Info("Load database journal from file", "path", path)
+		f, err := os.OpenFile(path, os.O_RDONLY, 0644)
+		if err != nil {
+			return nil, fmt.Errorf("failed to read journal file %s: %w", path, err)
+		}
+		defer f.Close()
+		reader = f
+	} else {
+		log.Info("Load database journal from disk")
+		journal := rawdb.ReadTrieJournal(db.diskdb)
+		if len(journal) == 0 {
+			return nil, errMissJournal
+		}
+		reader = bytes.NewReader(journal)
 	}
-	r := rlp.NewStream(bytes.NewReader(journal), 0)
+	r := rlp.NewStream(reader, 0)
 
 	// Firstly, resolve the first element as the journal version
 	version, err := r.Uint64()
@@ -297,9 +314,9 @@ func (db *Database) Journal(root common.Hash) error {
 	}
 	disk := db.tree.bottom()
 	if l, ok := l.(*diffLayer); ok {
-		log.Info("Persisting dirty state to disk", "head", l.block, "root", root, "layers", l.id-disk.id+disk.buffer.layers)
+		log.Info("Persisting dirty state", "head", l.block, "root", root, "layers", l.id-disk.id+disk.buffer.layers)
 	} else { // disk layer only on noop runs (likely) or deep reorgs (unlikely)
-		log.Info("Persisting dirty state to disk", "root", root, "layers", disk.buffer.layers)
+		log.Info("Persisting dirty state", "root", root, "layers", disk.buffer.layers)
 	}
 	// Block until the background flushing is finished and terminate
 	// the potential active state generator.
@@ -316,8 +333,37 @@ func (db *Database) Journal(root common.Hash) error {
 	if db.readOnly {
 		return errDatabaseReadOnly
 	}
+
+	// Store the journal into the database and return
+	var (
+		file        *os.File
+		journal     io.Writer
+		journalPath = db.journalPath()
+	)
+	if journalPath != "" {
+		// Write into a temp file first
+		err := os.MkdirAll(db.config.JournalDirectory, 0755)
+		if err != nil {
+			return err
+		}
+		tmp := journalPath + tempJournalSuffix
+		file, err = os.OpenFile(tmp, os.O_WRONLY|os.O_CREATE|os.O_TRUNC, 0644)
+		if err != nil {
+			return fmt.Errorf("failed to open journal file %s: %w", tmp, err)
+		}
+		defer func() {
+			if file != nil {
+				file.Close()
+				os.Remove(tmp) // Clean up temp file if we didn't successfully rename it
+				log.Warn("Removed leftover temporary journal file", "path", tmp)
+			}
+		}()
+		journal = file
+	} else {
+		journal = new(bytes.Buffer)
+	}
+
 	// Firstly write out the metadata of journal
-	journal := new(bytes.Buffer)
 	if err := rlp.Encode(journal, journalVersion); err != nil {
 		return err
 	}
@@ -334,11 +380,38 @@ func (db *Database) Journal(root common.Hash) error {
 	if err := l.journal(journal); err != nil {
 		return err
 	}
-	// Store the journal into the database and return
-	rawdb.WriteTrieJournal(db.diskdb, journal.Bytes())
 
+	// Store the journal into the database and return
+	if file == nil {
+		data := journal.(*bytes.Buffer)
+		size := data.Len()
+		rawdb.WriteTrieJournal(db.diskdb, data.Bytes())
+		log.Info("Persisted dirty state to disk", "size", common.StorageSize(size), "elapsed", common.PrettyDuration(time.Since(start)))
+	} else {
+		stat, err := file.Stat()
+		if err != nil {
+			return err
+		}
+		size := int(stat.Size())
+
+		// Close the temporary file and atomically rename it
+		if err := file.Sync(); err != nil {
+			return fmt.Errorf("failed to fsync the journal, %v", err)
+		}
+		if err := file.Close(); err != nil {
+			return fmt.Errorf("failed to close the journal: %v", err)
+		}
+		// Replace the live journal with the newly generated one
+		if err := os.Rename(journalPath+tempJournalSuffix, journalPath); err != nil {
+			return fmt.Errorf("failed to rename the journal: %v", err)
+		}
+		if err := syncDir(db.config.JournalDirectory); err != nil {
+			return fmt.Errorf("failed to fsync the dir: %v", err)
+		}
+		file = nil
+		log.Info("Persisted dirty state to file", "path", journalPath, "size", common.StorageSize(size), "elapsed", common.PrettyDuration(time.Since(start)))
+	}
 	// Set the db in read only mode to reject all following mutations
 	db.readOnly = true
-	log.Info("Persisted dirty state to disk", "size", common.StorageSize(journal.Len()), "elapsed", common.PrettyDuration(time.Since(start)))
 	return nil
 }