les: historical data garbage collection (#19570)

This change introduces garbage collection for the light client. Historical
chain data is deleted periodically. If you want to disable the GC, use
the --light.nopruning flag.

cmd/geth/main.go

@@ -98,6 +98,7 @@ var (
 		utils.LightEgressFlag,
 		utils.LightMaxPeersFlag,
 		utils.LegacyLightPeersFlag,
+		utils.LightNoPruneFlag,
 		utils.LightKDFFlag,
 		utils.UltraLightServersFlag,
 		utils.UltraLightFractionFlag,

cmd/geth/usage.go

@@ -96,6 +96,7 @@ var AppHelpFlagGroups = []flagGroup{
 			utils.UltraLightServersFlag,
 			utils.UltraLightFractionFlag,
 			utils.UltraLightOnlyAnnounceFlag,
+			utils.LightNoPruneFlag,
 		},
 	},
 	{

cmd/utils/flags.go

@@ -282,6 +282,10 @@ var (
 		Name:  "ulc.onlyannounce",
 		Usage: "Ultra light server sends announcements only",
 	}
+	LightNoPruneFlag = cli.BoolFlag{
+		Name:  "light.nopruning",
+		Usage: "Disable ancient light chain data pruning",
+	}
 	// Ethash settings
 	EthashCacheDirFlag = DirectoryFlag{
 		Name:  "ethash.cachedir",
@@ -1070,6 +1074,9 @@ func setLes(ctx *cli.Context, cfg *eth.Config) {
 	if ctx.GlobalIsSet(UltraLightOnlyAnnounceFlag.Name) {
 		cfg.UltraLightOnlyAnnounce = ctx.GlobalBool(UltraLightOnlyAnnounceFlag.Name)
 	}
+	if ctx.GlobalIsSet(LightNoPruneFlag.Name) {
+		cfg.LightNoPrune = ctx.GlobalBool(LightNoPruneFlag.Name)
+	}
 }
 
 // makeDatabaseHandles raises out the number of allowed file handles per process
@@ -1800,12 +1807,17 @@ func MakeChainDatabase(ctx *cli.Context, stack *node.Node) ethdb.Database {
 	var (
 		cache   = ctx.GlobalInt(CacheFlag.Name) * ctx.GlobalInt(CacheDatabaseFlag.Name) / 100
 		handles = makeDatabaseHandles()
+
+		err     error
+		chainDb ethdb.Database
 	)
-	name := "chaindata"
 	if ctx.GlobalString(SyncModeFlag.Name) == "light" {
-		name = "lightchaindata"
+		name := "lightchaindata"
+		chainDb, err = stack.OpenDatabase(name, cache, handles, "")
+	} else {
+		name := "chaindata"
+		chainDb, err = stack.OpenDatabaseWithFreezer(name, cache, handles, ctx.GlobalString(AncientFlag.Name), "")
 	}
-	chainDb, err := stack.OpenDatabaseWithFreezer(name, cache, handles, ctx.GlobalString(AncientFlag.Name), "")
 	if err != nil {
 		Fatalf("Could not open database: %v", err)
 	}

consensus/clique/clique.go

@@ -369,7 +369,7 @@ func (c *Clique) snapshot(chain consensus.ChainReader, number uint64, hash commo
 		// at a checkpoint block without a parent (light client CHT), or we have piled
 		// up more headers than allowed to be reorged (chain reinit from a freezer),
 		// consider the checkpoint trusted and snapshot it.
-		if number == 0 || (number%c.config.Epoch == 0 && (len(headers) > params.ImmutabilityThreshold || chain.GetHeaderByNumber(number-1) == nil)) {
+		if number == 0 || (number%c.config.Epoch == 0 && (len(headers) > params.FullImmutabilityThreshold || chain.GetHeaderByNumber(number-1) == nil)) {
 			checkpoint := chain.GetHeaderByNumber(number)
 			if checkpoint != nil {
 				hash := checkpoint.Hash()

core/blockchain.go

@@ -901,14 +901,14 @@ func (bc *BlockChain) Stop() {
 				recent := bc.GetBlockByNumber(number - offset)
 
 				log.Info("Writing cached state to disk", "block", recent.Number(), "hash", recent.Hash(), "root", recent.Root())
-				if err := triedb.Commit(recent.Root(), true); err != nil {
+				if err := triedb.Commit(recent.Root(), true, nil); err != nil {
 					log.Error("Failed to commit recent state trie", "err", err)
 				}
 			}
 		}
 		if snapBase != (common.Hash{}) {
 			log.Info("Writing snapshot state to disk", "root", snapBase)
-			if err := triedb.Commit(snapBase, true); err != nil {
+			if err := triedb.Commit(snapBase, true, nil); err != nil {
 				log.Error("Failed to commit recent state trie", "err", err)
 			}
 		}
@@ -1442,7 +1442,7 @@ func (bc *BlockChain) writeBlockWithState(block *types.Block, receipts []*types.
 
 	// If we're running an archive node, always flush
 	if bc.cacheConfig.TrieDirtyDisabled {
-		if err := triedb.Commit(root, false); err != nil {
+		if err := triedb.Commit(root, false, nil); err != nil {
 			return NonStatTy, err
 		}
 	} else {
@@ -1476,7 +1476,7 @@ func (bc *BlockChain) writeBlockWithState(block *types.Block, receipts []*types.
 						log.Info("State in memory for too long, committing", "time", bc.gcproc, "allowance", bc.cacheConfig.TrieTimeLimit, "optimum", float64(chosen-lastWrite)/TriesInMemory)
 					}
 					// Flush an entire trie and restart the counters
-					triedb.Commit(header.Root, true)
+					triedb.Commit(header.Root, true, nil)
 					lastWrite = chosen
 					bc.gcproc = 0
 				}

core/chain_indexer.go

@@ -46,6 +46,9 @@ type ChainIndexerBackend interface {
 
 	// Commit finalizes the section metadata and stores it into the database.
 	Commit() error
+
+	// Prune deletes the chain index older than the given threshold.
+	Prune(threshold uint64) error
 }
 
 // ChainIndexerChain interface is used for connecting the indexer to a blockchain
@@ -386,7 +389,6 @@ func (c *ChainIndexer) processSection(section uint64, lastHead common.Hash) (com
 	c.log.Trace("Processing new chain section", "section", section)
 
 	// Reset and partial processing
-
 	if err := c.backend.Reset(c.ctx, section, lastHead); err != nil {
 		c.setValidSections(0)
 		return common.Hash{}, err
@@ -459,6 +461,11 @@ func (c *ChainIndexer) AddChildIndexer(indexer *ChainIndexer) {
 	}
 }
 
+// Prune deletes all chain data older than given threshold.
+func (c *ChainIndexer) Prune(threshold uint64) error {
+	return c.backend.Prune(threshold)
+}
+
 // loadValidSections reads the number of valid sections from the index database
 // and caches is into the local state.
 func (c *ChainIndexer) loadValidSections() {

core/chain_indexer_test.go

@@ -236,3 +236,7 @@ func (b *testChainIndexBackend) Commit() error {
 	}
 	return nil
 }
+
+func (b *testChainIndexBackend) Prune(threshold uint64) error {
+	return nil
+}

core/chain_makers.go

@@ -220,7 +220,7 @@ func GenerateChain(config *params.ChainConfig, parent *types.Block, engine conse
 			if err != nil {
 				panic(fmt.Sprintf("state write error: %v", err))
 			}
-			if err := statedb.Database().TrieDB().Commit(root, false); err != nil {
+			if err := statedb.Database().TrieDB().Commit(root, false, nil); err != nil {
 				panic(fmt.Sprintf("trie write error: %v", err))
 			}
 			return block, b.receipts

core/dao_test.go

@@ -79,7 +79,7 @@ func TestDAOForkRangeExtradata(t *testing.T) {
 		if _, err := bc.InsertChain(blocks); err != nil {
 			t.Fatalf("failed to import contra-fork chain for expansion: %v", err)
 		}
-		if err := bc.stateCache.TrieDB().Commit(bc.CurrentHeader().Root, true); err != nil {
+		if err := bc.stateCache.TrieDB().Commit(bc.CurrentHeader().Root, true, nil); err != nil {
 			t.Fatalf("failed to commit contra-fork head for expansion: %v", err)
 		}
 		blocks, _ = GenerateChain(&proConf, conBc.CurrentBlock(), ethash.NewFaker(), db, 1, func(i int, gen *BlockGen) {})
@@ -104,7 +104,7 @@ func TestDAOForkRangeExtradata(t *testing.T) {
 		if _, err := bc.InsertChain(blocks); err != nil {
 			t.Fatalf("failed to import pro-fork chain for expansion: %v", err)
 		}
-		if err := bc.stateCache.TrieDB().Commit(bc.CurrentHeader().Root, true); err != nil {
+		if err := bc.stateCache.TrieDB().Commit(bc.CurrentHeader().Root, true, nil); err != nil {
 			t.Fatalf("failed to commit pro-fork head for expansion: %v", err)
 		}
 		blocks, _ = GenerateChain(&conConf, proBc.CurrentBlock(), ethash.NewFaker(), db, 1, func(i int, gen *BlockGen) {})
@@ -130,7 +130,7 @@ func TestDAOForkRangeExtradata(t *testing.T) {
 	if _, err := bc.InsertChain(blocks); err != nil {
 		t.Fatalf("failed to import contra-fork chain for expansion: %v", err)
 	}
-	if err := bc.stateCache.TrieDB().Commit(bc.CurrentHeader().Root, true); err != nil {
+	if err := bc.stateCache.TrieDB().Commit(bc.CurrentHeader().Root, true, nil); err != nil {
 		t.Fatalf("failed to commit contra-fork head for expansion: %v", err)
 	}
 	blocks, _ = GenerateChain(&proConf, conBc.CurrentBlock(), ethash.NewFaker(), db, 1, func(i int, gen *BlockGen) {})
@@ -150,7 +150,7 @@ func TestDAOForkRangeExtradata(t *testing.T) {
 	if _, err := bc.InsertChain(blocks); err != nil {
 		t.Fatalf("failed to import pro-fork chain for expansion: %v", err)
 	}
-	if err := bc.stateCache.TrieDB().Commit(bc.CurrentHeader().Root, true); err != nil {
+	if err := bc.stateCache.TrieDB().Commit(bc.CurrentHeader().Root, true, nil); err != nil {
 		t.Fatalf("failed to commit pro-fork head for expansion: %v", err)
 	}
 	blocks, _ = GenerateChain(&conConf, proBc.CurrentBlock(), ethash.NewFaker(), db, 1, func(i int, gen *BlockGen) {})

core/genesis.go

@@ -285,7 +285,7 @@ func (g *Genesis) ToBlock(db ethdb.Database) *types.Block {
 		head.Difficulty = params.GenesisDifficulty
 	}
 	statedb.Commit(false)
-	statedb.Database().TrieDB().Commit(root, true)
+	statedb.Database().TrieDB().Commit(root, true, nil)
 
 	return types.NewBlock(head, nil, nil, nil)
 }

core/rawdb/accessors_chain.go

@@ -80,6 +80,39 @@ func ReadAllHashes(db ethdb.Iteratee, number uint64) []common.Hash {
 	return hashes
 }
 
+// ReadAllCanonicalHashes retrieves all canonical number and hash mappings at the
+// certain chain range. If the accumulated entries reaches the given threshold,
+// abort the iteration and return the semi-finish result.
+func ReadAllCanonicalHashes(db ethdb.Iteratee, from uint64, to uint64, limit int) ([]uint64, []common.Hash) {
+	// Short circuit if the limit is 0.
+	if limit == 0 {
+		return nil, nil
+	}
+	var (
+		numbers []uint64
+		hashes  []common.Hash
+	)
+	// Construct the key prefix of start point.
+	start, end := headerHashKey(from), headerHashKey(to)
+	it := db.NewIterator(nil, start)
+	defer it.Release()
+
+	for it.Next() {
+		if bytes.Compare(it.Key(), end) >= 0 {
+			break
+		}
+		if key := it.Key(); len(key) == len(headerPrefix)+8+1 && bytes.Equal(key[len(key)-1:], headerHashSuffix) {
+			numbers = append(numbers, binary.BigEndian.Uint64(key[len(headerPrefix):len(headerPrefix)+8]))
+			hashes = append(hashes, common.BytesToHash(it.Value()))
+			// If the accumulated entries reaches the limit threshold, return.
+			if len(numbers) >= limit {
+				break
+			}
+		}
+	}
+	return numbers, hashes
+}
+
 // ReadHeaderNumber returns the header number assigned to a hash.
 func ReadHeaderNumber(db ethdb.KeyValueReader, hash common.Hash) *uint64 {
 	data, _ := db.Get(headerNumberKey(hash))

core/rawdb/accessors_chain_test.go

@@ -23,6 +23,7 @@ import (
 	"io/ioutil"
 	"math/big"
 	"os"
+	"reflect"
 	"testing"
 
 	"github.com/ethereum/go-ethereum/common"
@@ -424,3 +425,35 @@ func TestAncientStorage(t *testing.T) {
 		t.Fatalf("invalid td returned")
 	}
 }
+
+func TestCanonicalHashIteration(t *testing.T) {
+	var cases = []struct {
+		from, to uint64
+		limit    int
+		expect   []uint64
+	}{
+		{1, 8, 0, nil},
+		{1, 8, 1, []uint64{1}},
+		{1, 8, 10, []uint64{1, 2, 3, 4, 5, 6, 7}},
+		{1, 9, 10, []uint64{1, 2, 3, 4, 5, 6, 7, 8}},
+		{2, 9, 10, []uint64{2, 3, 4, 5, 6, 7, 8}},
+		{9, 10, 10, nil},
+	}
+	// Test empty db iteration
+	db := NewMemoryDatabase()
+	numbers, _ := ReadAllCanonicalHashes(db, 0, 10, 10)
+	if len(numbers) != 0 {
+		t.Fatalf("No entry should be returned to iterate an empty db")
+	}
+	// Fill database with testing data.
+	for i := uint64(1); i <= 8; i++ {
+		WriteCanonicalHash(db, common.Hash{}, i)
+		WriteTd(db, common.Hash{}, i, big.NewInt(10)) // Write some interferential data
+	}
+	for i, c := range cases {
+		numbers, _ := ReadAllCanonicalHashes(db, c.from, c.to, c.limit)
+		if !reflect.DeepEqual(numbers, c.expect) {
+			t.Fatalf("Case %d failed, want %v, got %v", i, c.expect, numbers)
+		}
+	}
+}

core/rawdb/accessors_indexes.go

@@ -17,6 +17,7 @@
 package rawdb
 
 import (
+	"bytes"
 	"math/big"
 
 	"github.com/ethereum/go-ethereum/common"
@@ -151,3 +152,24 @@ func WriteBloomBits(db ethdb.KeyValueWriter, bit uint, section uint64, head comm
 		log.Crit("Failed to store bloom bits", "err", err)
 	}
 }
+
+// DeleteBloombits removes all compressed bloom bits vector belonging to the
+// given section range and bit index.
+func DeleteBloombits(db ethdb.Database, bit uint, from uint64, to uint64) {
+	start, end := bloomBitsKey(bit, from, common.Hash{}), bloomBitsKey(bit, to, common.Hash{})
+	it := db.NewIterator(nil, start)
+	defer it.Release()
+
+	for it.Next() {
+		if bytes.Compare(it.Key(), end) >= 0 {
+			break
+		}
+		if len(it.Key()) != len(bloomBitsPrefix)+2+8+32 {
+			continue
+		}
+		db.Delete(it.Key())
+	}
+	if it.Error() != nil {
+		log.Crit("Failed to delete bloom bits", "err", it.Error())
+	}
+}

core/rawdb/accessors_indexes_test.go

@@ -17,12 +17,14 @@
 package rawdb
 
 import (
+	"bytes"
 	"math/big"
 	"testing"
 
 	"github.com/ethereum/go-ethereum/common"
 	"github.com/ethereum/go-ethereum/core/types"
 	"github.com/ethereum/go-ethereum/ethdb"
+	"github.com/ethereum/go-ethereum/params"
 	"github.com/ethereum/go-ethereum/rlp"
 )
 
@@ -106,3 +108,46 @@ func TestLookupStorage(t *testing.T) {
 		})
 	}
 }
+
+func TestDeleteBloomBits(t *testing.T) {
+	// Prepare testing data
+	db := NewMemoryDatabase()
+	for i := uint(0); i < 2; i++ {
+		for s := uint64(0); s < 2; s++ {
+			WriteBloomBits(db, i, s, params.MainnetGenesisHash, []byte{0x01, 0x02})
+			WriteBloomBits(db, i, s, params.RinkebyGenesisHash, []byte{0x01, 0x02})
+		}
+	}
+	check := func(bit uint, section uint64, head common.Hash, exist bool) {
+		bits, _ := ReadBloomBits(db, bit, section, head)
+		if exist && !bytes.Equal(bits, []byte{0x01, 0x02}) {
+			t.Fatalf("Bloombits mismatch")
+		}
+		if !exist && len(bits) > 0 {
+			t.Fatalf("Bloombits should be removed")
+		}
+	}
+	// Check the existence of written data.
+	check(0, 0, params.MainnetGenesisHash, true)
+	check(0, 0, params.RinkebyGenesisHash, true)
+
+	// Check the existence of deleted data.
+	DeleteBloombits(db, 0, 0, 1)
+	check(0, 0, params.MainnetGenesisHash, false)
+	check(0, 0, params.RinkebyGenesisHash, false)
+	check(0, 1, params.MainnetGenesisHash, true)
+	check(0, 1, params.RinkebyGenesisHash, true)
+
+	// Check the existence of deleted data.
+	DeleteBloombits(db, 0, 0, 2)
+	check(0, 0, params.MainnetGenesisHash, false)
+	check(0, 0, params.RinkebyGenesisHash, false)
+	check(0, 1, params.MainnetGenesisHash, false)
+	check(0, 1, params.RinkebyGenesisHash, false)
+
+	// Bit1 shouldn't be affect.
+	check(1, 0, params.MainnetGenesisHash, true)
+	check(1, 0, params.RinkebyGenesisHash, true)
+	check(1, 1, params.MainnetGenesisHash, true)
+	check(1, 1, params.RinkebyGenesisHash, true)
+}

core/rawdb/freezer.go

@@ -287,12 +287,12 @@ func (f *freezer) freeze(db ethdb.KeyValueStore) {
 			backoff = true
 			continue
 
-		case *number < params.ImmutabilityThreshold:
+		case *number < params.FullImmutabilityThreshold:
-			log.Debug("Current full block not old enough", "number", *number, "hash", hash, "delay", params.ImmutabilityThreshold)
+			log.Debug("Current full block not old enough", "number", *number, "hash", hash, "delay", params.FullImmutabilityThreshold)
 			backoff = true
 			continue
 
-		case *number-params.ImmutabilityThreshold <= f.frozen:
+		case *number-params.FullImmutabilityThreshold <= f.frozen:
 			log.Debug("Ancient blocks frozen already", "number", *number, "hash", hash, "frozen", f.frozen)
 			backoff = true
 			continue
@@ -304,7 +304,7 @@ func (f *freezer) freeze(db ethdb.KeyValueStore) {
 			continue
 		}
 		// Seems we have data ready to be frozen, process in usable batches
-		limit := *number - params.ImmutabilityThreshold
+		limit := *number - params.FullImmutabilityThreshold
 		if limit-f.frozen > freezerBatchLimit {
 			limit = f.frozen + freezerBatchLimit
 		}

core/state/statedb_test.go

@@ -55,7 +55,7 @@ func TestUpdateLeaks(t *testing.T) {
 	}
 
 	root := state.IntermediateRoot(false)
-	if err := state.Database().TrieDB().Commit(root, false); err != nil {
+	if err := state.Database().TrieDB().Commit(root, false, nil); err != nil {
 		t.Errorf("can not commit trie %v to persistent database", root.Hex())
 	}
 
@@ -106,7 +106,7 @@ func TestIntermediateLeaks(t *testing.T) {
 	if err != nil {
 		t.Fatalf("failed to commit transition state: %v", err)
 	}
-	if err = transState.Database().TrieDB().Commit(transRoot, false); err != nil {
+	if err = transState.Database().TrieDB().Commit(transRoot, false, nil); err != nil {
 		t.Errorf("can not commit trie %v to persistent database", transRoot.Hex())
 	}
 
@@ -114,7 +114,7 @@ func TestIntermediateLeaks(t *testing.T) {
 	if err != nil {
 		t.Fatalf("failed to commit final state: %v", err)
 	}
-	if err = finalState.Database().TrieDB().Commit(finalRoot, false); err != nil {
+	if err = finalState.Database().TrieDB().Commit(finalRoot, false, nil); err != nil {
 		t.Errorf("can not commit trie %v to persistent database", finalRoot.Hex())
 	}
 

eth/api_backend.go

@@ -185,8 +185,8 @@ func (b *EthAPIBackend) GetLogs(ctx context.Context, hash common.Hash) ([][]*typ
 	return logs, nil
 }
 
-func (b *EthAPIBackend) GetTd(blockHash common.Hash) *big.Int {
+func (b *EthAPIBackend) GetTd(ctx context.Context, hash common.Hash) *big.Int {
-	return b.eth.blockchain.GetTdByHash(blockHash)
+	return b.eth.blockchain.GetTdByHash(hash)
 }
 
 func (b *EthAPIBackend) GetEVM(ctx context.Context, msg core.Message, state *state.StateDB, header *types.Header) (*vm.EVM, func() error, error) {

eth/bloombits.go

@@ -136,3 +136,8 @@ func (b *BloomIndexer) Commit() error {
 	}
 	return batch.Write()
 }
+
+// PruneSections returns an empty error since we don't support pruning here.
+func (b *BloomIndexer) Prune(threshold uint64) error {
+	return nil
+}

eth/config.go

@@ -126,6 +126,7 @@ type Config struct {
 	LightIngress int  `toml:",omitempty"` // Incoming bandwidth limit for light servers
 	LightEgress  int  `toml:",omitempty"` // Outgoing bandwidth limit for light servers
 	LightPeers   int  `toml:",omitempty"` // Maximum number of LES client peers
+	LightNoPrune bool `toml:",omitempty"` // Whether to disable light chain pruning
 
 	// Ultra Light client options
 	UltraLightServers      []string `toml:",omitempty"` // List of trusted ultra light servers

eth/downloader/downloader.go

@@ -42,7 +42,6 @@ var (
 	MaxBlockFetch   = 128 // Amount of blocks to be fetched per retrieval request
 	MaxHeaderFetch  = 192 // Amount of block headers to be fetched per retrieval request
 	MaxSkeletonSize = 128 // Number of header fetches to need for a skeleton assembly
-	MaxBodyFetch    = 128 // Amount of block bodies to be fetched per retrieval request
 	MaxReceiptFetch = 256 // Amount of transaction receipts to allow fetching per request
 	MaxStateFetch   = 384 // Amount of node state values to allow fetching per request
 
@@ -59,7 +58,8 @@ var (
 	maxQueuedHeaders            = 32 * 1024                         // [eth/62] Maximum number of headers to queue for import (DOS protection)
 	maxHeadersProcess           = 2048                              // Number of header download results to import at once into the chain
 	maxResultsProcess           = 2048                              // Number of content download results to import at once into the chain
-	maxForkAncestry   uint64 = params.ImmutabilityThreshold // Maximum chain reorganisation (locally redeclared so tests can reduce it)
+	fullMaxForkAncestry  uint64 = params.FullImmutabilityThreshold  // Maximum chain reorganisation (locally redeclared so tests can reduce it)
+	lightMaxForkAncestry uint64 = params.LightImmutabilityThreshold // Maximum chain reorganisation (locally redeclared so tests can reduce it)
 
 	reorgProtThreshold   = 48 // Threshold number of recent blocks to disable mini reorg protection
 	reorgProtHeaderDelay = 2  // Number of headers to delay delivering to cover mini reorgs
@@ -490,10 +490,10 @@ func (d *Downloader) syncWithPeer(p *peerConnection, hash common.Hash, td *big.I
 		// The peer would start to feed us valid blocks until head, resulting in all of
 		// the blocks might be written into the ancient store. A following mini-reorg
 		// could cause issues.
-		if d.checkpoint != 0 && d.checkpoint > maxForkAncestry+1 {
+		if d.checkpoint != 0 && d.checkpoint > fullMaxForkAncestry+1 {
 			d.ancientLimit = d.checkpoint
-		} else if height > maxForkAncestry+1 {
+		} else if height > fullMaxForkAncestry+1 {
-			d.ancientLimit = height - maxForkAncestry - 1
+			d.ancientLimit = height - fullMaxForkAncestry - 1
 		}
 		frozen, _ := d.stateDB.Ancients() // Ignore the error here since light client can also hit here.
 		// If a part of blockchain data has already been written into active store,
@@ -727,6 +727,10 @@ func (d *Downloader) findAncestor(p *peerConnection, remoteHeader *types.Header)
 	p.log.Debug("Looking for common ancestor", "local", localHeight, "remote", remoteHeight)
 
 	// Recap floor value for binary search
+	maxForkAncestry := fullMaxForkAncestry
+	if d.getMode() == LightSync {
+		maxForkAncestry = lightMaxForkAncestry
+	}
 	if localHeight >= maxForkAncestry {
 		// We're above the max reorg threshold, find the earliest fork point
 		floor = int64(localHeight - maxForkAncestry)

eth/downloader/downloader_test.go

@@ -37,7 +37,8 @@ import (
 
 // Reduce some of the parameters to make the tester faster.
 func init() {
-	maxForkAncestry = 10000
+	fullMaxForkAncestry = 10000
+	lightMaxForkAncestry = 10000
 	blockCacheItems = 1024
 	fsHeaderContCheck = 500 * time.Millisecond
 }

eth/downloader/testchain_test.go

@@ -45,7 +45,7 @@ var testChainBase = newTestChain(blockCacheItems+200, testGenesis)
 var testChainForkLightA, testChainForkLightB, testChainForkHeavy *testChain
 
 func init() {
-	var forkLen = int(maxForkAncestry + 50)
+	var forkLen = int(fullMaxForkAncestry + 50)
 	var wg sync.WaitGroup
 	wg.Add(3)
 	go func() { testChainForkLightA = testChainBase.makeFork(forkLen, false, 1); wg.Done() }()

eth/gen_config.go

@@ -29,6 +29,7 @@ func (c Config) MarshalTOML() (interface{}, error) {
 		LightIngress            int                    `toml:",omitempty"`
 		LightEgress             int                    `toml:",omitempty"`
 		LightPeers              int                    `toml:",omitempty"`
+		LightNoPrune            bool                   `toml:",omitempty"`
 		UltraLightServers       []string               `toml:",omitempty"`
 		UltraLightFraction      int                    `toml:",omitempty"`
 		UltraLightOnlyAnnounce  bool                   `toml:",omitempty"`
@@ -66,6 +67,7 @@ func (c Config) MarshalTOML() (interface{}, error) {
 	enc.LightIngress = c.LightIngress
 	enc.LightEgress = c.LightEgress
 	enc.LightPeers = c.LightPeers
+	enc.LightNoPrune = c.LightNoPrune
 	enc.UltraLightServers = c.UltraLightServers
 	enc.UltraLightFraction = c.UltraLightFraction
 	enc.UltraLightOnlyAnnounce = c.UltraLightOnlyAnnounce
@@ -107,6 +109,7 @@ func (c *Config) UnmarshalTOML(unmarshal func(interface{}) error) error {
 		LightIngress            *int                   `toml:",omitempty"`
 		LightEgress             *int                   `toml:",omitempty"`
 		LightPeers              *int                   `toml:",omitempty"`
+		LightNoPrune            *bool                  `toml:",omitempty"`
 		UltraLightServers       []string               `toml:",omitempty"`
 		UltraLightFraction      *int                   `toml:",omitempty"`
 		UltraLightOnlyAnnounce  *bool                  `toml:",omitempty"`
@@ -171,6 +174,9 @@ func (c *Config) UnmarshalTOML(unmarshal func(interface{}) error) error {
 	if dec.LightPeers != nil {
 		c.LightPeers = *dec.LightPeers
 	}
+	if dec.LightNoPrune != nil {
+		c.LightNoPrune = *dec.LightNoPrune
+	}
 	if dec.UltraLightServers != nil {
 		c.UltraLightServers = dec.UltraLightServers
 	}

graphql/graphql.go

@@ -584,7 +584,7 @@ func (b *Block) TotalDifficulty(ctx context.Context) (hexutil.Big, error) {
 		}
 		h = header.Hash()
 	}
-	return hexutil.Big(*b.backend.GetTd(h)), nil
+	return hexutil.Big(*b.backend.GetTd(ctx, h)), nil
 }
 
 // BlockNumberArgs encapsulates arguments to accessors that specify a block number.

internal/ethapi/api.go

@@ -36,7 +36,6 @@ import (
 	"github.com/ethereum/go-ethereum/consensus/clique"
 	"github.com/ethereum/go-ethereum/consensus/ethash"
 	"github.com/ethereum/go-ethereum/core"
-	"github.com/ethereum/go-ethereum/core/rawdb"
 	"github.com/ethereum/go-ethereum/core/types"
 	"github.com/ethereum/go-ethereum/core/vm"
 	"github.com/ethereum/go-ethereum/crypto"
@@ -625,7 +624,7 @@ func (s *PublicBlockChainAPI) GetProof(ctx context.Context, address common.Addre
 func (s *PublicBlockChainAPI) GetHeaderByNumber(ctx context.Context, number rpc.BlockNumber) (map[string]interface{}, error) {
 	header, err := s.b.HeaderByNumber(ctx, number)
 	if header != nil && err == nil {
-		response := s.rpcMarshalHeader(header)
+		response := s.rpcMarshalHeader(ctx, header)
 		if number == rpc.PendingBlockNumber {
 			// Pending header need to nil out a few fields
 			for _, field := range []string{"hash", "nonce", "miner"} {
@@ -641,7 +640,7 @@ func (s *PublicBlockChainAPI) GetHeaderByNumber(ctx context.Context, number rpc.
 func (s *PublicBlockChainAPI) GetHeaderByHash(ctx context.Context, hash common.Hash) map[string]interface{} {
 	header, _ := s.b.HeaderByHash(ctx, hash)
 	if header != nil {
-		return s.rpcMarshalHeader(header)
+		return s.rpcMarshalHeader(ctx, header)
 	}
 	return nil
 }
@@ -654,7 +653,7 @@ func (s *PublicBlockChainAPI) GetHeaderByHash(ctx context.Context, hash common.H
 func (s *PublicBlockChainAPI) GetBlockByNumber(ctx context.Context, number rpc.BlockNumber, fullTx bool) (map[string]interface{}, error) {
 	block, err := s.b.BlockByNumber(ctx, number)
 	if block != nil && err == nil {
-		response, err := s.rpcMarshalBlock(block, true, fullTx)
+		response, err := s.rpcMarshalBlock(ctx, block, true, fullTx)
 		if err == nil && number == rpc.PendingBlockNumber {
 			// Pending blocks need to nil out a few fields
 			for _, field := range []string{"hash", "nonce", "miner"} {
@@ -671,7 +670,7 @@ func (s *PublicBlockChainAPI) GetBlockByNumber(ctx context.Context, number rpc.B
 func (s *PublicBlockChainAPI) GetBlockByHash(ctx context.Context, hash common.Hash, fullTx bool) (map[string]interface{}, error) {
 	block, err := s.b.BlockByHash(ctx, hash)
 	if block != nil {
-		return s.rpcMarshalBlock(block, true, fullTx)
+		return s.rpcMarshalBlock(ctx, block, true, fullTx)
 	}
 	return nil, err
 }
@@ -687,7 +686,7 @@ func (s *PublicBlockChainAPI) GetUncleByBlockNumberAndIndex(ctx context.Context,
 			return nil, nil
 		}
 		block = types.NewBlockWithHeader(uncles[index])
-		return s.rpcMarshalBlock(block, false, false)
+		return s.rpcMarshalBlock(ctx, block, false, false)
 	}
 	return nil, err
 }
@@ -703,7 +702,7 @@ func (s *PublicBlockChainAPI) GetUncleByBlockHashAndIndex(ctx context.Context, b
 			return nil, nil
 		}
 		block = types.NewBlockWithHeader(uncles[index])
-		return s.rpcMarshalBlock(block, false, false)
+		return s.rpcMarshalBlock(ctx, block, false, false)
 	}
 	return nil, err
 }
@@ -1173,21 +1172,21 @@ func RPCMarshalBlock(block *types.Block, inclTx bool, fullTx bool) (map[string]i
 
 // rpcMarshalHeader uses the generalized output filler, then adds the total difficulty field, which requires
 // a `PublicBlockchainAPI`.
-func (s *PublicBlockChainAPI) rpcMarshalHeader(header *types.Header) map[string]interface{} {
+func (s *PublicBlockChainAPI) rpcMarshalHeader(ctx context.Context, header *types.Header) map[string]interface{} {
 	fields := RPCMarshalHeader(header)
-	fields["totalDifficulty"] = (*hexutil.Big)(s.b.GetTd(header.Hash()))
+	fields["totalDifficulty"] = (*hexutil.Big)(s.b.GetTd(ctx, header.Hash()))
 	return fields
 }
 
 // rpcMarshalBlock uses the generalized output filler, then adds the total difficulty field, which requires
 // a `PublicBlockchainAPI`.
-func (s *PublicBlockChainAPI) rpcMarshalBlock(b *types.Block, inclTx bool, fullTx bool) (map[string]interface{}, error) {
+func (s *PublicBlockChainAPI) rpcMarshalBlock(ctx context.Context, b *types.Block, inclTx bool, fullTx bool) (map[string]interface{}, error) {
 	fields, err := RPCMarshalBlock(b, inclTx, fullTx)
 	if err != nil {
 		return nil, err
 	}
 	if inclTx {
-		fields["totalDifficulty"] = (*hexutil.Big)(s.b.GetTd(b.Hash()))
+		fields["totalDifficulty"] = (*hexutil.Big)(s.b.GetTd(ctx, b.Hash()))
 	}
 	return fields, err
 }
@@ -1393,8 +1392,8 @@ func (s *PublicTransactionPoolAPI) GetRawTransactionByHash(ctx context.Context,
 
 // GetTransactionReceipt returns the transaction receipt for the given transaction hash.
 func (s *PublicTransactionPoolAPI) GetTransactionReceipt(ctx context.Context, hash common.Hash) (map[string]interface{}, error) {
-	tx, blockHash, blockNumber, index := rawdb.ReadTransaction(s.b.ChainDb(), hash)
+	tx, blockHash, blockNumber, index, err := s.b.GetTransaction(ctx, hash)
-	if tx == nil {
+	if err != nil {
 		return nil, nil
 	}
 	receipts, err := s.b.GetReceipts(ctx, blockHash)

internal/ethapi/backend.go

@@ -59,7 +59,7 @@ type Backend interface {
 	StateAndHeaderByNumber(ctx context.Context, number rpc.BlockNumber) (*state.StateDB, *types.Header, error)
 	StateAndHeaderByNumberOrHash(ctx context.Context, blockNrOrHash rpc.BlockNumberOrHash) (*state.StateDB, *types.Header, error)
 	GetReceipts(ctx context.Context, hash common.Hash) (types.Receipts, error)
-	GetTd(hash common.Hash) *big.Int
+	GetTd(ctx context.Context, hash common.Hash) *big.Int
 	GetEVM(ctx context.Context, msg core.Message, state *state.StateDB, header *types.Header) (*vm.EVM, func() error, error)
 	SubscribeChainEvent(ch chan<- core.ChainEvent) event.Subscription
 	SubscribeChainHeadEvent(ch chan<- core.ChainHeadEvent) event.Subscription

les/api_backend.go

@@ -162,8 +162,11 @@ func (b *LesApiBackend) GetLogs(ctx context.Context, hash common.Hash) ([][]*typ
 	return nil, nil
 }
 
-func (b *LesApiBackend) GetTd(hash common.Hash) *big.Int {
+func (b *LesApiBackend) GetTd(ctx context.Context, hash common.Hash) *big.Int {
-	return b.eth.blockchain.GetTdByHash(hash)
+	if number := rawdb.ReadHeaderNumber(b.eth.chainDb, hash); number != nil {
+		return b.eth.blockchain.GetTdOdr(ctx, hash, *number)
+	}
+	return nil
 }
 
 func (b *LesApiBackend) GetEVM(ctx context.Context, msg core.Message, state *state.StateDB, header *types.Header) (*vm.EVM, func() error, error) {

les/client.go

@@ -62,6 +62,7 @@ type LightEthereum struct {
 	serverPool     *serverPool
 	valueTracker   *lpc.ValueTracker
 	dialCandidates enode.Iterator
+	pruner         *pruner
 
 	bloomRequests chan chan *bloombits.Retrieval // Channel receiving bloom data retrieval requests
 	bloomIndexer  *core.ChainIndexer             // Bloom indexer operating during block imports
@@ -121,8 +122,8 @@ func New(ctx *node.ServiceContext, config *eth.Config) (*LightEthereum, error) {
 	leth.relay = newLesTxRelay(peers, leth.retriever)
 
 	leth.odr = NewLesOdr(chainDb, light.DefaultClientIndexerConfig, leth.retriever)
-	leth.chtIndexer = light.NewChtIndexer(chainDb, leth.odr, params.CHTFrequency, params.HelperTrieConfirmations)
+	leth.chtIndexer = light.NewChtIndexer(chainDb, leth.odr, params.CHTFrequency, params.HelperTrieConfirmations, config.LightNoPrune)
-	leth.bloomTrieIndexer = light.NewBloomTrieIndexer(chainDb, leth.odr, params.BloomBitsBlocksClient, params.BloomTrieFrequency)
+	leth.bloomTrieIndexer = light.NewBloomTrieIndexer(chainDb, leth.odr, params.BloomBitsBlocksClient, params.BloomTrieFrequency, config.LightNoPrune)
 	leth.odr.SetIndexers(leth.chtIndexer, leth.bloomTrieIndexer, leth.bloomIndexer)
 
 	checkpoint := config.Checkpoint
@@ -149,6 +150,9 @@ func New(ctx *node.ServiceContext, config *eth.Config) (*LightEthereum, error) {
 	leth.chtIndexer.Start(leth.blockchain)
 	leth.bloomIndexer.Start(leth.blockchain)
 
+	// Start a light chain pruner to delete useless historical data.
+	leth.pruner = newPruner(chainDb, leth.chtIndexer, leth.bloomTrieIndexer)
+
 	// Rewind the chain in case of an incompatible config upgrade.
 	if compat, ok := genesisErr.(*params.ConfigCompatError); ok {
 		log.Warn("Rewinding chain to upgrade configuration", "err", compat)
@@ -302,6 +306,7 @@ func (s *LightEthereum) Stop() error {
 	s.handler.stop()
 	s.txPool.Stop()
 	s.engine.Close()
+	s.pruner.close()
 	s.eventMux.Stop()
 	s.chainDb.Close()
 	s.wg.Wait()

les/odr_requests.go

@@ -110,14 +110,16 @@ func (r *BlockRequest) Validate(db ethdb.Database, msg *Msg) error {
 	body := bodies[0]
 
 	// Retrieve our stored header and validate block content against it
-	header := rawdb.ReadHeader(db, r.Hash, r.Number)
+	if r.Header == nil {
-	if header == nil {
+		r.Header = rawdb.ReadHeader(db, r.Hash, r.Number)
+	}
+	if r.Header == nil {
 		return errHeaderUnavailable
 	}
-	if header.TxHash != types.DeriveSha(types.Transactions(body.Transactions)) {
+	if r.Header.TxHash != types.DeriveSha(types.Transactions(body.Transactions)) {
 		return errTxHashMismatch
 	}
-	if header.UncleHash != types.CalcUncleHash(body.Uncles) {
+	if r.Header.UncleHash != types.CalcUncleHash(body.Uncles) {
 		return errUncleHashMismatch
 	}
 	// Validations passed, encode and store RLP

les/odr_test.go

@@ -183,7 +183,7 @@ func odrTxStatus(ctx context.Context, db ethdb.Database, config *params.ChainCon
 // testOdr tests odr requests whose validation guaranteed by block headers.
 func testOdr(t *testing.T, protocol int, expFail uint64, checkCached bool, fn odrTestFn) {
 	// Assemble the test environment
-	server, client, tearDown := newClientServerEnv(t, 4, protocol, nil, nil, 0, false, true)
+	server, client, tearDown := newClientServerEnv(t, 4, protocol, nil, nil, 0, false, true, true)
 	defer tearDown()
 
 	// Ensure the client has synced all necessary data.

les/pruner.go

@@ -0,0 +1,98 @@
+// Copyright 2019 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/>.
+
+package les
+
+import (
+	"sync"
+	"time"
+
+	"github.com/ethereum/go-ethereum/common/math"
+	"github.com/ethereum/go-ethereum/core"
+	"github.com/ethereum/go-ethereum/ethdb"
+	"github.com/ethereum/go-ethereum/log"
+)
+
+// pruner is responsible for pruning historical light chain data.
+type pruner struct {
+	db       ethdb.Database
+	indexers []*core.ChainIndexer
+	closeCh  chan struct{}
+	wg       sync.WaitGroup
+}
+
+// newPruner returns a light chain pruner instance.
+func newPruner(db ethdb.Database, indexers ...*core.ChainIndexer) *pruner {
+	pruner := &pruner{
+		db:       db,
+		indexers: indexers,
+		closeCh:  make(chan struct{}),
+	}
+	pruner.wg.Add(1)
+	go pruner.loop()
+	return pruner
+}
+
+// close notifies all background goroutines belonging to pruner to exit.
+func (p *pruner) close() {
+	close(p.closeCh)
+	p.wg.Wait()
+}
+
+// loop periodically queries the status of chain indexers and prunes useless
+// historical chain data. Notably, whenever Geth restarts, it will iterate
+// all historical sections even they don't exist at all(below checkpoint) so
+// that light client can prune cached chain data that was ODRed after pruning
+// that section.
+func (p *pruner) loop() {
+	defer p.wg.Done()
+
+	// cleanTicker is the ticker used to trigger a history clean 2 times a day.
+	var cleanTicker = time.NewTicker(12 * time.Hour)
+
+	// pruning finds the sections that have been processed by all indexers
+	// and deletes all historical chain data.
+	// Note, if some indexers don't support pruning(e.g. eth.BloomIndexer),
+	// pruning operations can be silently ignored.
+	pruning := func() {
+		min := uint64(math.MaxUint64)
+		for _, indexer := range p.indexers {
+			sections, _, _ := indexer.Sections()
+			if sections < min {
+				min = sections
+			}
+		}
+		// Always keep the latest section data in database.
+		if min < 2 || len(p.indexers) == 0 {
+			return
+		}
+		for _, indexer := range p.indexers {
+			if err := indexer.Prune(min - 2); err != nil {
+				log.Debug("Failed to prune historical data", "err", err)
+				return
+			}
+		}
+		p.db.Compact(nil, nil) // Compact entire database, ensure all removed data are deleted.
+	}
+	for {
+		pruning()
+		select {
+		case <-cleanTicker.C:
+		case <-p.closeCh:
+			return
+		}
+	}
+}

les/pruner_test.go

@@ -0,0 +1,197 @@
+// Copyright 2019 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/>.
+
+package les
+
+import (
+	"bytes"
+	"context"
+	"encoding/binary"
+	"testing"
+	"time"
+
+	"github.com/ethereum/go-ethereum/core"
+	"github.com/ethereum/go-ethereum/light"
+)
+
+func TestLightPruner(t *testing.T) {
+	config := light.TestClientIndexerConfig
+
+	waitIndexers := func(cIndexer, bIndexer, btIndexer *core.ChainIndexer) {
+		for {
+			cs, _, _ := cIndexer.Sections()
+			bts, _, _ := btIndexer.Sections()
+			if cs >= 3 && bts >= 3 {
+				break
+			}
+			time.Sleep(10 * time.Millisecond)
+		}
+	}
+	server, client, tearDown := newClientServerEnv(t, int(3*config.ChtSize+config.ChtConfirms), 2, waitIndexers, nil, 0, false, true, false)
+	defer tearDown()
+
+	// checkDB iterates the chain with given prefix, resolves the block number
+	// with given callback and ensures this entry should exist or not.
+	checkDB := func(from, to uint64, prefix []byte, resolve func(key, value []byte) *uint64, exist bool) bool {
+		it := client.db.NewIterator(prefix, nil)
+		defer it.Release()
+
+		var next = from
+		for it.Next() {
+			number := resolve(it.Key(), it.Value())
+			if number == nil || *number < from {
+				continue
+			} else if *number > to {
+				return true
+			}
+			if exist {
+				if *number != next {
+					return false
+				}
+				next++
+			} else {
+				return false
+			}
+		}
+		return true
+	}
+	// checkPruned checks and ensures the stale chain data has been pruned.
+	checkPruned := func(from, to uint64) {
+		// Iterate canonical hash
+		if !checkDB(from, to, []byte("h"), func(key, value []byte) *uint64 {
+			if len(key) == 1+8+1 && bytes.Equal(key[9:10], []byte("n")) {
+				n := binary.BigEndian.Uint64(key[1:9])
+				return &n
+			}
+			return nil
+		}, false) {
+			t.Fatalf("canonical hash mappings are not properly pruned")
+		}
+		// Iterate header
+		if !checkDB(from, to, []byte("h"), func(key, value []byte) *uint64 {
+			if len(key) == 1+8+32 {
+				n := binary.BigEndian.Uint64(key[1:9])
+				return &n
+			}
+			return nil
+		}, false) {
+			t.Fatalf("headers are not properly pruned")
+		}
+		// Iterate body
+		if !checkDB(from, to, []byte("b"), func(key, value []byte) *uint64 {
+			if len(key) == 1+8+32 {
+				n := binary.BigEndian.Uint64(key[1:9])
+				return &n
+			}
+			return nil
+		}, false) {
+			t.Fatalf("block bodies are not properly pruned")
+		}
+		// Iterate receipts
+		if !checkDB(from, to, []byte("r"), func(key, value []byte) *uint64 {
+			if len(key) == 1+8+32 {
+				n := binary.BigEndian.Uint64(key[1:9])
+				return &n
+			}
+			return nil
+		}, false) {
+			t.Fatalf("receipts are not properly pruned")
+		}
+		// Iterate td
+		if !checkDB(from, to, []byte("h"), func(key, value []byte) *uint64 {
+			if len(key) == 1+8+32+1 && bytes.Equal(key[41:42], []byte("t")) {
+				n := binary.BigEndian.Uint64(key[1:9])
+				return &n
+			}
+			return nil
+		}, false) {
+			t.Fatalf("tds are not properly pruned")
+		}
+	}
+	// Start light pruner.
+	time.Sleep(1500 * time.Millisecond) // Ensure light client has finished the syncing and indexing
+	newPruner(client.db, client.chtIndexer, client.bloomTrieIndexer)
+
+	time.Sleep(1500 * time.Millisecond) // Ensure pruner have enough time to prune data.
+	checkPruned(1, config.ChtSize-1)
+
+	// Ensure all APIs still work after pruning.
+	var cases = []struct {
+		from, to   uint64
+		methodName string
+		method     func(uint64) bool
+	}{
+		{
+			1, 10, "GetHeaderByNumber",
+			func(n uint64) bool {
+				_, err := light.GetHeaderByNumber(context.Background(), client.handler.backend.odr, n)
+				return err == nil
+			},
+		},
+		{
+			11, 20, "GetCanonicalHash",
+			func(n uint64) bool {
+				_, err := light.GetCanonicalHash(context.Background(), client.handler.backend.odr, n)
+				return err == nil
+			},
+		},
+		{
+			21, 30, "GetTd",
+			func(n uint64) bool {
+				_, err := light.GetTd(context.Background(), client.handler.backend.odr, server.handler.blockchain.GetHeaderByNumber(n).Hash(), n)
+				return err == nil
+			},
+		},
+		{
+			31, 40, "GetBodyRLP",
+			func(n uint64) bool {
+				_, err := light.GetBodyRLP(context.Background(), client.handler.backend.odr, server.handler.blockchain.GetHeaderByNumber(n).Hash(), n)
+				return err == nil
+			},
+		},
+		{
+			41, 50, "GetBlock",
+			func(n uint64) bool {
+				_, err := light.GetBlock(context.Background(), client.handler.backend.odr, server.handler.blockchain.GetHeaderByNumber(n).Hash(), n)
+				return err == nil
+			},
+		},
+		{
+			51, 60, "GetBlockReceipts",
+			func(n uint64) bool {
+				_, err := light.GetBlockReceipts(context.Background(), client.handler.backend.odr, server.handler.blockchain.GetHeaderByNumber(n).Hash(), n)
+				return err == nil
+			},
+		},
+	}
+	for _, c := range cases {
+		for i := c.from; i <= c.to; i++ {
+			if !c.method(i) {
+				t.Fatalf("rpc method %s failed, number %d", c.methodName, i)
+			}
+		}
+	}
+	// Check GetBloombits
+	_, err := light.GetBloomBits(context.Background(), client.handler.backend.odr, 0, []uint64{0})
+	if err != nil {
+		t.Fatalf("Failed to retrieve bloombits of pruned section: %v", err)
+	}
+
+	// Ensure the ODR cached data can be cleaned by pruner.
+	newPruner(client.db, client.chtIndexer, client.bloomTrieIndexer)
+	time.Sleep(50 * time.Millisecond) // Ensure pruner have enough time to prune data.
+	checkPruned(1, config.ChtSize-1)  // Ensure all cached data(by odr) is cleaned.
+}

les/request_test.go

@@ -79,7 +79,7 @@ func tfCodeAccess(db ethdb.Database, bhash common.Hash, num uint64) light.OdrReq
 
 func testAccess(t *testing.T, protocol int, fn accessTestFn) {
 	// Assemble the test environment
-	server, client, tearDown := newClientServerEnv(t, 4, protocol, nil, nil, 0, false, true)
+	server, client, tearDown := newClientServerEnv(t, 4, protocol, nil, nil, 0, false, true, true)
 	defer tearDown()
 
 	// Ensure the client has synced all necessary data.

les/server.go

@@ -77,8 +77,8 @@ func NewLesServer(e *eth.Ethereum, config *eth.Config) (*LesServer, error) {
 			iConfig:          light.DefaultServerIndexerConfig,
 			chainDb:          e.ChainDb(),
 			chainReader:      e.BlockChain(),
-			chtIndexer:       light.NewChtIndexer(e.ChainDb(), nil, params.CHTFrequency, params.HelperTrieProcessConfirmations),
+			chtIndexer:       light.NewChtIndexer(e.ChainDb(), nil, params.CHTFrequency, params.HelperTrieProcessConfirmations, true),
-			bloomTrieIndexer: light.NewBloomTrieIndexer(e.ChainDb(), nil, params.BloomBitsBlocks, params.BloomTrieFrequency),
+			bloomTrieIndexer: light.NewBloomTrieIndexer(e.ChainDb(), nil, params.BloomBitsBlocks, params.BloomTrieFrequency, true),
 			closeCh:          make(chan struct{}),
 		},
 		archiveMode:  e.ArchiveMode(),

les/sync_test.go

@@ -54,7 +54,7 @@ func testCheckpointSyncing(t *testing.T, protocol int, syncMode int) {
 		}
 	}
 	// Generate 512+4 blocks (totally 1 CHT sections)
-	server, client, tearDown := newClientServerEnv(t, int(config.ChtSize+config.ChtConfirms), protocol, waitIndexers, nil, 0, false, false)
+	server, client, tearDown := newClientServerEnv(t, int(config.ChtSize+config.ChtConfirms), protocol, waitIndexers, nil, 0, false, false, true)
 	defer tearDown()
 
 	expected := config.ChtSize + config.ChtConfirms
@@ -144,7 +144,7 @@ func testMissOracleBackend(t *testing.T, hasCheckpoint bool) {
 		}
 	}
 	// Generate 512+4 blocks (totally 1 CHT sections)
-	server, client, tearDown := newClientServerEnv(t, int(config.ChtSize+config.ChtConfirms), 3, waitIndexers, nil, 0, false, false)
+	server, client, tearDown := newClientServerEnv(t, int(config.ChtSize+config.ChtConfirms), 3, waitIndexers, nil, 0, false, false, true)
 	defer tearDown()
 
 	expected := config.ChtSize + config.ChtConfirms

les/test_helper.go

@@ -158,11 +158,11 @@ func prepare(n int, backend *backends.SimulatedBackend) {
 }
 
 // testIndexers creates a set of indexers with specified params for testing purpose.
-func testIndexers(db ethdb.Database, odr light.OdrBackend, config *light.IndexerConfig) []*core.ChainIndexer {
+func testIndexers(db ethdb.Database, odr light.OdrBackend, config *light.IndexerConfig, disablePruning bool) []*core.ChainIndexer {
 	var indexers [3]*core.ChainIndexer
-	indexers[0] = light.NewChtIndexer(db, odr, config.ChtSize, config.ChtConfirms)
+	indexers[0] = light.NewChtIndexer(db, odr, config.ChtSize, config.ChtConfirms, disablePruning)
 	indexers[1] = eth.NewBloomIndexer(db, config.BloomSize, config.BloomConfirms)
-	indexers[2] = light.NewBloomTrieIndexer(db, odr, config.BloomSize, config.BloomTrieSize)
+	indexers[2] = light.NewBloomTrieIndexer(db, odr, config.BloomSize, config.BloomTrieSize, disablePruning)
 	// make bloomTrieIndexer as a child indexer of bloom indexer.
 	indexers[1].AddChildIndexer(indexers[2])
 	return indexers[:]
@@ -456,7 +456,7 @@ type testServer struct {
 
 func newServerEnv(t *testing.T, blocks int, protocol int, callback indexerCallback, simClock bool, newPeer bool, testCost uint64) (*testServer, func()) {
 	db := rawdb.NewMemoryDatabase()
-	indexers := testIndexers(db, nil, light.TestServerIndexerConfig)
+	indexers := testIndexers(db, nil, light.TestServerIndexerConfig, true)
 
 	var clock mclock.Clock = &mclock.System{}
 	if simClock {
@@ -499,7 +499,7 @@ func newServerEnv(t *testing.T, blocks int, protocol int, callback indexerCallba
 	return server, teardown
 }
 
-func newClientServerEnv(t *testing.T, blocks int, protocol int, callback indexerCallback, ulcServers []string, ulcFraction int, simClock bool, connect bool) (*testServer, *testClient, func()) {
+func newClientServerEnv(t *testing.T, blocks int, protocol int, callback indexerCallback, ulcServers []string, ulcFraction int, simClock bool, connect bool, disablePruning bool) (*testServer, *testClient, func()) {
 	sdb, cdb := rawdb.NewMemoryDatabase(), rawdb.NewMemoryDatabase()
 	speers, cpeers := newServerPeerSet(), newClientPeerSet()
 
@@ -511,8 +511,8 @@ func newClientServerEnv(t *testing.T, blocks int, protocol int, callback indexer
 	rm := newRetrieveManager(speers, dist, func() time.Duration { return time.Millisecond * 500 })
 	odr := NewLesOdr(cdb, light.TestClientIndexerConfig, rm)
 
-	sindexers := testIndexers(sdb, nil, light.TestServerIndexerConfig)
+	sindexers := testIndexers(sdb, nil, light.TestServerIndexerConfig, true)
-	cIndexers := testIndexers(cdb, odr, light.TestClientIndexerConfig)
+	cIndexers := testIndexers(cdb, odr, light.TestClientIndexerConfig, disablePruning)
 
 	scIndexer, sbIndexer, sbtIndexer := sindexers[0], sindexers[1], sindexers[2]
 	ccIndexer, cbIndexer, cbtIndexer := cIndexers[0], cIndexers[1], cIndexers[2]
@@ -542,7 +542,7 @@ func newClientServerEnv(t *testing.T, blocks int, protocol int, callback indexer
 		}
 		select {
 		case <-done:
-		case <-time.After(3 * time.Second):
+		case <-time.After(10 * time.Second):
 			t.Fatal("test peer did not connect and sync within 3s")
 		}
 	}

les/ulc_test.go

@@ -138,6 +138,6 @@ func newTestServerPeer(t *testing.T, blocks int, protocol int) (*testServer, *en
 
 // newTestLightPeer creates node with light sync mode
 func newTestLightPeer(t *testing.T, protocol int, ulcServers []string, ulcFraction int) (*testClient, func()) {
-	_, c, teardown := newClientServerEnv(t, 0, protocol, nil, ulcServers, ulcFraction, false, false)
+	_, c, teardown := newClientServerEnv(t, 0, protocol, nil, ulcServers, ulcFraction, false, false, true)
 	return c, teardown
 }

light/lightchain.go

@@ -112,7 +112,7 @@ func NewLightChain(odr OdrBackend, config *params.ChainConfig, engine consensus.
 		if header := bc.GetHeaderByHash(hash); header != nil {
 			log.Error("Found bad hash, rewinding chain", "number", header.Number, "hash", header.ParentHash)
 			bc.SetHead(header.Number.Uint64() - 1)
-			log.Error("Chain rewind was successful, resuming normal operation")
+			log.Info("Chain rewind was successful, resuming normal operation")
 		}
 	}
 	return bc, nil
@@ -155,7 +155,11 @@ func (lc *LightChain) loadLastState() error {
 		// Corrupt or empty database, init from scratch
 		lc.Reset()
 	} else {
-		if header := lc.GetHeaderByHash(head); header != nil {
+		header := lc.GetHeaderByHash(head)
+		if header == nil {
+			// Corrupt or empty database, init from scratch
+			lc.Reset()
+		} else {
 			lc.hc.SetCurrentHeader(header)
 		}
 	}
@@ -163,7 +167,6 @@ func (lc *LightChain) loadLastState() error {
 	header := lc.hc.CurrentHeader()
 	headerTd := lc.GetTd(header.Hash(), header.Number.Uint64())
 	log.Info("Loaded most recent local header", "number", header.Number, "hash", header.Hash(), "td", headerTd, "age", common.PrettyAge(time.Unix(int64(header.Time), 0)))
-
 	return nil
 }
 
@@ -431,6 +434,17 @@ func (lc *LightChain) GetTdByHash(hash common.Hash) *big.Int {
 	return lc.hc.GetTdByHash(hash)
 }
 
+// GetHeaderByNumberOdr retrieves the total difficult from the database or
+// network by hash and number, caching it (associated with its hash) if found.
+func (lc *LightChain) GetTdOdr(ctx context.Context, hash common.Hash, number uint64) *big.Int {
+	td := lc.GetTd(hash, number)
+	if td != nil {
+		return td
+	}
+	td, _ = GetTd(ctx, lc.odr, hash, number)
+	return td
+}
+
 // GetHeader retrieves a block header from the database by hash and number,
 // caching it if found.
 func (lc *LightChain) GetHeader(hash common.Hash, number uint64) *types.Header {

light/odr.go

@@ -82,7 +82,6 @@ func StorageTrieID(state *TrieID, addrHash, root common.Hash) *TrieID {
 
 // TrieRequest is the ODR request type for state/storage trie entries
 type TrieRequest struct {
-	OdrRequest
 	Id    *TrieID
 	Key   []byte
 	Proof *NodeSet
@@ -95,7 +94,6 @@ func (req *TrieRequest) StoreResult(db ethdb.Database) {
 
 // CodeRequest is the ODR request type for retrieving contract code
 type CodeRequest struct {
-	OdrRequest
 	Id   *TrieID // references storage trie of the account
 	Hash common.Hash
 	Data []byte
@@ -108,9 +106,9 @@ func (req *CodeRequest) StoreResult(db ethdb.Database) {
 
 // BlockRequest is the ODR request type for retrieving block bodies
 type BlockRequest struct {
-	OdrRequest
 	Hash   common.Hash
 	Number uint64
+	Header *types.Header
 	Rlp    []byte
 }
 
@@ -119,9 +117,8 @@ func (req *BlockRequest) StoreResult(db ethdb.Database) {
 	rawdb.WriteBodyRLP(db, req.Hash, req.Number, req.Rlp)
 }
 
-// ReceiptsRequest is the ODR request type for retrieving block bodies
+// ReceiptsRequest is the ODR request type for retrieving receipts.
 type ReceiptsRequest struct {
-	OdrRequest
 	Untrusted bool // Indicator whether the result retrieved is trusted or not
 	Hash      common.Hash
 	Number    uint64
@@ -138,7 +135,6 @@ func (req *ReceiptsRequest) StoreResult(db ethdb.Database) {
 
 // ChtRequest is the ODR request type for state/storage trie entries
 type ChtRequest struct {
-	OdrRequest
 	Untrusted        bool   // Indicator whether the result retrieved is trusted or not
 	PeerId           string // The specified peer id from which to retrieve data.
 	Config           *IndexerConfig
@@ -193,7 +189,6 @@ type TxStatus struct {
 
 // TxStatusRequest is the ODR request type for retrieving transaction status
 type TxStatusRequest struct {
-	OdrRequest
 	Hashes []common.Hash
 	Status []TxStatus
 }

light/odr_util.go

@@ -19,6 +19,7 @@ package light
 import (
 	"bytes"
 	"context"
+	"math/big"
 
 	"github.com/ethereum/go-ethereum/common"
 	"github.com/ethereum/go-ethereum/core"
@@ -30,65 +31,83 @@ import (
 
 var sha3Nil = crypto.Keccak256Hash(nil)
 
+// GetHeaderByNumber retrieves the canonical block header corresponding to the
+// given number.
 func GetHeaderByNumber(ctx context.Context, odr OdrBackend, number uint64) (*types.Header, error) {
+	// Try to find it in the local database first.
 	db := odr.Database()
 	hash := rawdb.ReadCanonicalHash(db, number)
+
+	// If there is a canonical hash, there should have a header too.
+	// But if it's pruned, re-fetch from network again.
 	if (hash != common.Hash{}) {
-		// if there is a canonical hash, there is a header too
+		if header := rawdb.ReadHeader(db, hash, number); header != nil {
-		header := rawdb.ReadHeader(db, hash, number)
-		if header == nil {
-			panic("Canonical hash present but header not found")
-		}
 			return header, nil
 		}
-
-	var (
-		chtCount, sectionHeadNum uint64
-		sectionHead              common.Hash
-	)
-	if odr.ChtIndexer() != nil {
-		chtCount, sectionHeadNum, sectionHead = odr.ChtIndexer().Sections()
-		canonicalHash := rawdb.ReadCanonicalHash(db, sectionHeadNum)
-		// if the CHT was injected as a trusted checkpoint, we have no canonical hash yet so we accept zero hash too
-		for chtCount > 0 && canonicalHash != sectionHead && canonicalHash != (common.Hash{}) {
-			chtCount--
-			if chtCount > 0 {
-				sectionHeadNum = chtCount*odr.IndexerConfig().ChtSize - 1
-				sectionHead = odr.ChtIndexer().SectionHead(chtCount - 1)
-				canonicalHash = rawdb.ReadCanonicalHash(db, sectionHeadNum)
 	}
-		}
+	// Retrieve the header via ODR, ensure the requested header is covered
-	}
+	// by local trusted CHT.
-	if number >= chtCount*odr.IndexerConfig().ChtSize {
+	chts, _, chtHead := odr.ChtIndexer().Sections()
+	if number >= chts*odr.IndexerConfig().ChtSize {
 		return nil, errNoTrustedCht
 	}
-	r := &ChtRequest{ChtRoot: GetChtRoot(db, chtCount-1, sectionHead), ChtNum: chtCount - 1, BlockNum: number, Config: odr.IndexerConfig()}
+	r := &ChtRequest{
+		ChtRoot:  GetChtRoot(db, chts-1, chtHead),
+		ChtNum:   chts - 1,
+		BlockNum: number,
+		Config:   odr.IndexerConfig(),
+	}
 	if err := odr.Retrieve(ctx, r); err != nil {
 		return nil, err
 	}
 	return r.Header, nil
 }
 
-// GetUntrustedHeaderByNumber fetches specified block header without correctness checking.
+// GetUntrustedHeaderByNumber retrieves specified block header without
-// Note this function should only be used in light client checkpoint syncing.
+// correctness checking. Note this function should only be used in light
+// client checkpoint syncing.
 func GetUntrustedHeaderByNumber(ctx context.Context, odr OdrBackend, number uint64, peerId string) (*types.Header, error) {
-	r := &ChtRequest{BlockNum: number, ChtNum: number / odr.IndexerConfig().ChtSize, Untrusted: true, PeerId: peerId, Config: odr.IndexerConfig()}
+	// todo(rjl493456442) it's a hack to retrieve headers which is not covered
+	// by CHT. Fix it in LES4
+	r := &ChtRequest{
+		BlockNum:  number,
+		ChtNum:    number / odr.IndexerConfig().ChtSize,
+		Untrusted: true,
+		PeerId:    peerId,
+		Config:    odr.IndexerConfig(),
+	}
 	if err := odr.Retrieve(ctx, r); err != nil {
 		return nil, err
 	}
 	return r.Header, nil
 }
 
+// GetCanonicalHash retrieves the canonical block hash corresponding to the number.
 func GetCanonicalHash(ctx context.Context, odr OdrBackend, number uint64) (common.Hash, error) {
 	hash := rawdb.ReadCanonicalHash(odr.Database(), number)
-	if (hash != common.Hash{}) {
+	if hash != (common.Hash{}) {
 		return hash, nil
 	}
 	header, err := GetHeaderByNumber(ctx, odr, number)
-	if header != nil {
+	if err != nil {
+		return common.Hash{}, err
+	}
+	// number -> canonical mapping already be stored in db, get it.
 	return header.Hash(), nil
 }
-	return common.Hash{}, err
+
+// GetTd retrieves the total difficulty corresponding to the number and hash.
+func GetTd(ctx context.Context, odr OdrBackend, hash common.Hash, number uint64) (*big.Int, error) {
+	td := rawdb.ReadTd(odr.Database(), hash, number)
+	if td != nil {
+		return td, nil
+	}
+	_, err := GetHeaderByNumber(ctx, odr, number)
+	if err != nil {
+		return nil, err
+	}
+	// <hash, number> -> td mapping already be stored in db, get it.
+	return rawdb.ReadTd(odr.Database(), hash, number), nil
 }
 
 // GetBodyRLP retrieves the block body (transactions and uncles) in RLP encoding.
@@ -96,15 +115,19 @@ func GetBodyRLP(ctx context.Context, odr OdrBackend, hash common.Hash, number ui
 	if data := rawdb.ReadBodyRLP(odr.Database(), hash, number); data != nil {
 		return data, nil
 	}
-	r := &BlockRequest{Hash: hash, Number: number}
+	// Retrieve the block header first and pass it for verification.
+	header, err := GetHeaderByNumber(ctx, odr, number)
+	if err != nil {
+		return nil, errNoHeader
+	}
+	r := &BlockRequest{Hash: hash, Number: number, Header: header}
 	if err := odr.Retrieve(ctx, r); err != nil {
 		return nil, err
-	} else {
+	}
 	return r.Rlp, nil
 }
-}
 
-// GetBody retrieves the block body (transactons, uncles) corresponding to the
+// GetBody retrieves the block body (transactions, uncles) corresponding to the
 // hash.
 func GetBody(ctx context.Context, odr OdrBackend, hash common.Hash, number uint64) (*types.Body, error) {
 	data, err := GetBodyRLP(ctx, odr, hash, number)
@@ -122,8 +145,8 @@ func GetBody(ctx context.Context, odr OdrBackend, hash common.Hash, number uint6
 // back from the stored header and body.
 func GetBlock(ctx context.Context, odr OdrBackend, hash common.Hash, number uint64) (*types.Block, error) {
 	// Retrieve the block header and body contents
-	header := rawdb.ReadHeader(odr.Database(), hash, number)
+	header, err := GetHeaderByNumber(ctx, odr, number)
-	if header == nil {
+	if err != nil {
 		return nil, errNoHeader
 	}
 	body, err := GetBody(ctx, odr, hash, number)
@@ -140,7 +163,11 @@ func GetBlockReceipts(ctx context.Context, odr OdrBackend, hash common.Hash, num
 	// Assume receipts are already stored locally and attempt to retrieve.
 	receipts := rawdb.ReadRawReceipts(odr.Database(), hash, number)
 	if receipts == nil {
-		r := &ReceiptsRequest{Hash: hash, Number: number}
+		header, err := GetHeaderByNumber(ctx, odr, number)
+		if err != nil {
+			return nil, errNoHeader
+		}
+		r := &ReceiptsRequest{Hash: hash, Number: number, Header: header}
 		if err := odr.Retrieve(ctx, r); err != nil {
 			return nil, err
 		}
@@ -171,7 +198,6 @@ func GetBlockLogs(ctx context.Context, odr OdrBackend, hash common.Hash, number
 	if err != nil {
 		return nil, err
 	}
-	// Return the logs without deriving any computed fields on the receipts
 	logs := make([][]*types.Log, len(receipts))
 	for i, receipt := range receipts {
 		logs[i] = receipt.Logs
@@ -203,82 +229,68 @@ func GetUntrustedBlockLogs(ctx context.Context, odr OdrBackend, header *types.He
 	return logs, nil
 }
 
-// GetBloomBits retrieves a batch of compressed bloomBits vectors belonging to the given bit index and section indexes
+// GetBloomBits retrieves a batch of compressed bloomBits vectors belonging to
-func GetBloomBits(ctx context.Context, odr OdrBackend, bitIdx uint, sectionIdxList []uint64) ([][]byte, error) {
+// the given bit index and section indexes.
+func GetBloomBits(ctx context.Context, odr OdrBackend, bit uint, sections []uint64) ([][]byte, error) {
 	var (
+		reqIndex    []int
+		reqSections []uint64
 		db          = odr.Database()
-		result  = make([][]byte, len(sectionIdxList))
+		result      = make([][]byte, len(sections))
-		reqList []uint64
-		reqIdx  []int
 	)
-
+	blooms, _, sectionHead := odr.BloomTrieIndexer().Sections()
-	var (
+	for i, section := range sections {
-		bloomTrieCount, sectionHeadNum uint64
+		sectionHead := rawdb.ReadCanonicalHash(db, (section+1)*odr.IndexerConfig().BloomSize-1)
-		sectionHead                    common.Hash
+		// If we don't have the canonical hash stored for this section head number,
-	)
+		// we'll still look for an entry with a zero sectionHead (we store it with
-	if odr.BloomTrieIndexer() != nil {
+		// zero section head too if we don't know it at the time of the retrieval)
-		bloomTrieCount, sectionHeadNum, sectionHead = odr.BloomTrieIndexer().Sections()
+		if bloomBits, _ := rawdb.ReadBloomBits(db, bit, section, sectionHead); len(bloomBits) != 0 {
-		canonicalHash := rawdb.ReadCanonicalHash(db, sectionHeadNum)
-		// if the BloomTrie was injected as a trusted checkpoint, we have no canonical hash yet so we accept zero hash too
-		for bloomTrieCount > 0 && canonicalHash != sectionHead && canonicalHash != (common.Hash{}) {
-			bloomTrieCount--
-			if bloomTrieCount > 0 {
-				sectionHeadNum = bloomTrieCount*odr.IndexerConfig().BloomTrieSize - 1
-				sectionHead = odr.BloomTrieIndexer().SectionHead(bloomTrieCount - 1)
-				canonicalHash = rawdb.ReadCanonicalHash(db, sectionHeadNum)
-			}
-		}
-	}
-
-	for i, sectionIdx := range sectionIdxList {
-		sectionHead := rawdb.ReadCanonicalHash(db, (sectionIdx+1)*odr.IndexerConfig().BloomSize-1)
-		// if we don't have the canonical hash stored for this section head number, we'll still look for
-		// an entry with a zero sectionHead (we store it with zero section head too if we don't know it
-		// at the time of the retrieval)
-		bloomBits, err := rawdb.ReadBloomBits(db, bitIdx, sectionIdx, sectionHead)
-		if err == nil {
 			result[i] = bloomBits
-		} else {
+			continue
+		}
 		// TODO(rjl493456442) Convert sectionIndex to BloomTrie relative index
-			if sectionIdx >= bloomTrieCount {
+		if section >= blooms {
 			return nil, errNoTrustedBloomTrie
 		}
-			reqList = append(reqList, sectionIdx)
+		reqSections = append(reqSections, section)
-			reqIdx = append(reqIdx, i)
+		reqIndex = append(reqIndex, i)
 	}
-	}
+	// Find all bloombits in database, nothing to query via odr, return.
-	if reqList == nil {
+	if reqSections == nil {
 		return result, nil
 	}
-
+	// Send odr request to retrieve missing bloombits.
-	r := &BloomRequest{BloomTrieRoot: GetBloomTrieRoot(db, bloomTrieCount-1, sectionHead), BloomTrieNum: bloomTrieCount - 1,
+	r := &BloomRequest{
-		BitIdx: bitIdx, SectionIndexList: reqList, Config: odr.IndexerConfig()}
+		BloomTrieRoot:    GetBloomTrieRoot(db, blooms-1, sectionHead),
+		BloomTrieNum:     blooms - 1,
+		BitIdx:           bit,
+		SectionIndexList: reqSections,
+		Config:           odr.IndexerConfig(),
+	}
 	if err := odr.Retrieve(ctx, r); err != nil {
 		return nil, err
-	} else {
+	}
-		for i, idx := range reqIdx {
+	for i, idx := range reqIndex {
 		result[idx] = r.BloomBits[i]
 	}
 	return result, nil
 }
-}
 
 // GetTransaction retrieves a canonical transaction by hash and also returns its position in the chain
 func GetTransaction(ctx context.Context, odr OdrBackend, txHash common.Hash) (*types.Transaction, common.Hash, uint64, uint64, error) {
 	r := &TxStatusRequest{Hashes: []common.Hash{txHash}}
 	if err := odr.Retrieve(ctx, r); err != nil || r.Status[0].Status != core.TxStatusIncluded {
 		return nil, common.Hash{}, 0, 0, err
-	} else {
+	}
 	pos := r.Status[0].Lookup
 	// first ensure that we have the header, otherwise block body retrieval will fail
 	// also verify if this is a canonical block by getting the header by number and checking its hash
 	if header, err := GetHeaderByNumber(ctx, odr, pos.BlockIndex); err != nil || header.Hash() != pos.BlockHash {
 		return nil, common.Hash{}, 0, 0, err
 	}
-		if body, err := GetBody(ctx, odr, pos.BlockHash, pos.BlockIndex); err != nil || uint64(len(body.Transactions)) <= pos.Index || body.Transactions[pos.Index].Hash() != txHash {
+	body, err := GetBody(ctx, odr, pos.BlockHash, pos.BlockIndex)
+	if err != nil || uint64(len(body.Transactions)) <= pos.Index || body.Transactions[pos.Index].Hash() != txHash {
 		return nil, common.Hash{}, 0, 0, err
-		} else {
+	}
 	return body.Transactions[pos.Index], pos.BlockHash, pos.BlockIndex, pos.Index, nil
 }
-	}
-}

light/postprocess.go

@@ -17,6 +17,7 @@
 package light
 
 import (
+	"bytes"
 	"context"
 	"encoding/binary"
 	"errors"
@@ -24,6 +25,7 @@ import (
 	"math/big"
 	"time"
 
+	"github.com/deckarep/golang-set"
 	"github.com/ethereum/go-ethereum/common"
 	"github.com/ethereum/go-ethereum/common/bitutil"
 	"github.com/ethereum/go-ethereum/core"
@@ -128,23 +130,27 @@ func StoreChtRoot(db ethdb.Database, sectionIdx uint64, sectionHead, root common
 
 // ChtIndexerBackend implements core.ChainIndexerBackend.
 type ChtIndexerBackend struct {
+	disablePruning       bool
 	diskdb, trieTable    ethdb.Database
 	odr                  OdrBackend
 	triedb               *trie.Database
+	trieset              mapset.Set
 	section, sectionSize uint64
 	lastHash             common.Hash
 	trie                 *trie.Trie
 }
 
 // NewChtIndexer creates a Cht chain indexer
-func NewChtIndexer(db ethdb.Database, odr OdrBackend, size, confirms uint64) *core.ChainIndexer {
+func NewChtIndexer(db ethdb.Database, odr OdrBackend, size, confirms uint64, disablePruning bool) *core.ChainIndexer {
 	trieTable := rawdb.NewTable(db, ChtTablePrefix)
 	backend := &ChtIndexerBackend{
 		diskdb:         db,
 		odr:            odr,
 		trieTable:      trieTable,
 		triedb:         trie.NewDatabaseWithCache(trieTable, 1), // Use a tiny cache only to keep memory down
+		trieset:        mapset.NewSet(),
 		sectionSize:    size,
+		disablePruning: disablePruning,
 	}
 	return core.NewChainIndexer(db, rawdb.NewTable(db, "chtIndexV2-"), backend, size, confirms, time.Millisecond*100, "cht")
 }
@@ -189,7 +195,6 @@ func (c *ChtIndexerBackend) Reset(ctx context.Context, section uint64, lastSecti
 			c.trie, err = trie.New(root, c.triedb)
 		}
 	}
-
 	c.section = section
 	return err
 }
@@ -216,13 +221,83 @@ func (c *ChtIndexerBackend) Commit() error {
 	if err != nil {
 		return err
 	}
-	c.triedb.Commit(root, false)
+	// Pruning historical trie nodes if necessary.
+	if !c.disablePruning {
+		// Flush the triedb and track the latest trie nodes.
+		c.trieset.Clear()
+		c.triedb.Commit(root, false, func(hash common.Hash) { c.trieset.Add(hash) })
 
+		it := c.trieTable.NewIterator(nil, nil)
+		defer it.Release()
+
+		var (
+			deleted   int
+			remaining int
+			t         = time.Now()
+		)
+		for it.Next() {
+			trimmed := bytes.TrimPrefix(it.Key(), []byte(ChtTablePrefix))
+			if !c.trieset.Contains(common.BytesToHash(trimmed)) {
+				c.trieTable.Delete(trimmed)
+				deleted += 1
+			} else {
+				remaining += 1
+			}
+		}
+		log.Debug("Prune historical CHT trie nodes", "deleted", deleted, "remaining", remaining, "elapsed", common.PrettyDuration(time.Since(t)))
+	} else {
+		c.triedb.Commit(root, false, nil)
+	}
 	log.Info("Storing CHT", "section", c.section, "head", fmt.Sprintf("%064x", c.lastHash), "root", fmt.Sprintf("%064x", root))
 	StoreChtRoot(c.diskdb, c.section, c.lastHash, root)
 	return nil
 }
 
+// PruneSections implements core.ChainIndexerBackend which deletes all
+// chain data(except hash<->number mappings) older than the specified
+// threshold.
+func (c *ChtIndexerBackend) Prune(threshold uint64) error {
+	// Short circuit if the light pruning is disabled.
+	if c.disablePruning {
+		return nil
+	}
+	t := time.Now()
+	// Always keep genesis header in database.
+	start, end := uint64(1), (threshold+1)*c.sectionSize
+
+	var batch = c.diskdb.NewBatch()
+	for {
+		numbers, hashes := rawdb.ReadAllCanonicalHashes(c.diskdb, start, end, 10240)
+		if len(numbers) == 0 {
+			break
+		}
+		for i := 0; i < len(numbers); i++ {
+			// Keep hash<->number mapping in database otherwise the hash based
+			// API(e.g. GetReceipt, GetLogs) will be broken.
+			//
+			// Storage size wise, the size of a mapping is ~41bytes. For one
+			// section is about 1.3MB which is acceptable.
+			//
+			// In order to totally get rid of this index, we need an additional
+			// flag to specify how many historical data light client can serve.
+			rawdb.DeleteCanonicalHash(batch, numbers[i])
+			rawdb.DeleteBlockWithoutNumber(batch, hashes[i], numbers[i])
+		}
+		if batch.ValueSize() > ethdb.IdealBatchSize {
+			if err := batch.Write(); err != nil {
+				return err
+			}
+			batch.Reset()
+		}
+		start = numbers[len(numbers)-1] + 1
+	}
+	if err := batch.Write(); err != nil {
+		return err
+	}
+	log.Debug("Prune history headers", "threshold", threshold, "elapsed", common.PrettyDuration(time.Since(t)))
+	return nil
+}
+
 var (
 	bloomTriePrefix      = []byte("bltRoot-") // bloomTriePrefix + bloomTrieNum (uint64 big endian) -> trie root hash
 	BloomTrieTablePrefix = "blt-"
@@ -245,8 +320,10 @@ func StoreBloomTrieRoot(db ethdb.Database, sectionIdx uint64, sectionHead, root
 
 // BloomTrieIndexerBackend implements core.ChainIndexerBackend
 type BloomTrieIndexerBackend struct {
+	disablePruning    bool
 	diskdb, trieTable ethdb.Database
 	triedb            *trie.Database
+	trieset           mapset.Set
 	odr               OdrBackend
 	section           uint64
 	parentSize        uint64
@@ -257,15 +334,17 @@ type BloomTrieIndexerBackend struct {
 }
 
 // NewBloomTrieIndexer creates a BloomTrie chain indexer
-func NewBloomTrieIndexer(db ethdb.Database, odr OdrBackend, parentSize, size uint64) *core.ChainIndexer {
+func NewBloomTrieIndexer(db ethdb.Database, odr OdrBackend, parentSize, size uint64, disablePruning bool) *core.ChainIndexer {
 	trieTable := rawdb.NewTable(db, BloomTrieTablePrefix)
 	backend := &BloomTrieIndexerBackend{
 		diskdb:         db,
 		odr:            odr,
 		trieTable:      trieTable,
 		triedb:         trie.NewDatabaseWithCache(trieTable, 1), // Use a tiny cache only to keep memory down
+		trieset:        mapset.NewSet(),
 		parentSize:     parentSize,
 		size:           size,
+		disablePruning: disablePruning,
 	}
 	backend.bloomTrieRatio = size / parentSize
 	backend.sectionHeads = make([]common.Hash, backend.bloomTrieRatio)
@@ -303,7 +382,6 @@ func (b *BloomTrieIndexerBackend) fetchMissingNodes(ctx context.Context, section
 			}
 		}()
 	}
-
 	for i := uint(0); i < types.BloomBitLength; i++ {
 		indexCh <- i
 	}
@@ -380,10 +458,51 @@ func (b *BloomTrieIndexerBackend) Commit() error {
 	if err != nil {
 		return err
 	}
-	b.triedb.Commit(root, false)
+	// Pruning historical trie nodes if necessary.
+	if !b.disablePruning {
+		// Flush the triedb and track the latest trie nodes.
+		b.trieset.Clear()
+		b.triedb.Commit(root, false, func(hash common.Hash) { b.trieset.Add(hash) })
 
+		it := b.trieTable.NewIterator(nil, nil)
+		defer it.Release()
+
+		var (
+			deleted   int
+			remaining int
+			t         = time.Now()
+		)
+		for it.Next() {
+			trimmed := bytes.TrimPrefix(it.Key(), []byte(BloomTrieTablePrefix))
+			if !b.trieset.Contains(common.BytesToHash(trimmed)) {
+				b.trieTable.Delete(trimmed)
+				deleted += 1
+			} else {
+				remaining += 1
+			}
+		}
+		log.Debug("Prune historical bloom trie nodes", "deleted", deleted, "remaining", remaining, "elapsed", common.PrettyDuration(time.Since(t)))
+	} else {
+		b.triedb.Commit(root, false, nil)
+	}
 	sectionHead := b.sectionHeads[b.bloomTrieRatio-1]
-	log.Info("Storing bloom trie", "section", b.section, "head", fmt.Sprintf("%064x", sectionHead), "root", fmt.Sprintf("%064x", root), "compression", float64(compSize)/float64(decompSize))
 	StoreBloomTrieRoot(b.diskdb, b.section, sectionHead, root)
+	log.Info("Storing bloom trie", "section", b.section, "head", fmt.Sprintf("%064x", sectionHead), "root", fmt.Sprintf("%064x", root), "compression", float64(compSize)/float64(decompSize))
+
+	return nil
+}
+
+// Prune implements core.ChainIndexerBackend which deletes all
+// bloombits which older than the specified threshold.
+func (b *BloomTrieIndexerBackend) Prune(threshold uint64) error {
+	// Short circuit if the light pruning is disabled.
+	if b.disablePruning {
+		return nil
+	}
+	start := time.Now()
+	for i := uint(0); i < types.BloomBitLength; i++ {
+		rawdb.DeleteBloombits(b.diskdb, i, 0, threshold*b.bloomTrieRatio+b.bloomTrieRatio)
+	}
+	log.Debug("Prune history bloombits", "threshold", threshold, "elapsed", common.PrettyDuration(time.Since(start)))
 	return nil
 }

params/network_params.go

@@ -53,9 +53,15 @@ const (
 	// CheckpointProcessConfirmations is the number before a checkpoint is generated
 	CheckpointProcessConfirmations = 256
 
-	// ImmutabilityThreshold is the number of blocks after which a chain segment is
+	// FullImmutabilityThreshold is the number of blocks after which a chain segment is
 	// considered immutable (i.e. soft finality). It is used by the downloader as a
 	// hard limit against deep ancestors, by the blockchain against deep reorgs, by
 	// the freezer as the cutoff threshold and by clique as the snapshot trust limit.
-	ImmutabilityThreshold = 90000
+	FullImmutabilityThreshold = 90000
+
+	// LightImmutabilityThreshold is the number of blocks after which a header chain
+	// segment is considered immutable for light client(i.e. soft finality). It is used by
+	// the downloader as a hard limit against deep ancestors, by the blockchain against deep
+	// reorgs, by the light pruner as the pruning validity guarantee.
+	LightImmutabilityThreshold = 30000
 )

trie/database.go

@@ -693,7 +693,7 @@ func (db *Database) Cap(limit common.StorageSize) error {
 //
 // Note, this method is a non-synchronized mutator. It is unsafe to call this
 // concurrently with other mutators.
-func (db *Database) Commit(node common.Hash, report bool) error {
+func (db *Database) Commit(node common.Hash, report bool, callback func(common.Hash)) error {
 	// Create a database batch to flush persistent data out. It is important that
 	// outside code doesn't see an inconsistent state (referenced data removed from
 	// memory cache during commit but not yet in persistent storage). This is ensured
@@ -732,7 +732,7 @@ func (db *Database) Commit(node common.Hash, report bool) error {
 	nodes, storage := len(db.dirties), db.dirtiesSize
 
 	uncacher := &cleaner{db}
-	if err := db.commit(node, batch, uncacher); err != nil {
+	if err := db.commit(node, batch, uncacher, callback); err != nil {
 		log.Error("Failed to commit trie from trie database", "err", err)
 		return err
 	}
@@ -771,7 +771,7 @@ func (db *Database) Commit(node common.Hash, report bool) error {
 }
 
 // commit is the private locked version of Commit.
-func (db *Database) commit(hash common.Hash, batch ethdb.Batch, uncacher *cleaner) error {
+func (db *Database) commit(hash common.Hash, batch ethdb.Batch, uncacher *cleaner, callback func(common.Hash)) error {
 	// If the node does not exist, it's a previously committed node
 	node, ok := db.dirties[hash]
 	if !ok {
@@ -780,7 +780,7 @@ func (db *Database) commit(hash common.Hash, batch ethdb.Batch, uncacher *cleane
 	var err error
 	node.forChilds(func(child common.Hash) {
 		if err == nil {
-			err = db.commit(child, batch, uncacher)
+			err = db.commit(child, batch, uncacher, callback)
 		}
 	})
 	if err != nil {
@@ -789,6 +789,9 @@ func (db *Database) commit(hash common.Hash, batch ethdb.Batch, uncacher *cleane
 	if err := batch.Put(hash[:], node.rlp()); err != nil {
 		return err
 	}
+	if callback != nil {
+		callback(hash)
+	}
 	// If we've reached an optimal batch size, commit and start over
 	if batch.ValueSize() >= ethdb.IdealBatchSize {
 		if err := batch.Write(); err != nil {

trie/iterator_test.go

@@ -301,7 +301,7 @@ func testIteratorContinueAfterError(t *testing.T, memonly bool) {
 	}
 	tr.Commit(nil)
 	if !memonly {
-		triedb.Commit(tr.Hash(), true)
+		triedb.Commit(tr.Hash(), true, nil)
 	}
 	wantNodeCount := checkIteratorNoDups(t, tr.NodeIterator(nil), nil)
 
@@ -392,7 +392,7 @@ func testIteratorContinueAfterSeekError(t *testing.T, memonly bool) {
 	}
 	root, _ := ctr.Commit(nil)
 	if !memonly {
-		triedb.Commit(root, true)
+		triedb.Commit(root, true, nil)
 	}
 	barNodeHash := common.HexToHash("05041990364eb72fcb1127652ce40d8bab765f2bfe53225b1170d276cc101c2e")
 	var (

trie/trie_test.go

@@ -88,7 +88,7 @@ func testMissingNode(t *testing.T, memonly bool) {
 	updateString(trie, "123456", "asdfasdfasdfasdfasdfasdfasdfasdf")
 	root, _ := trie.Commit(nil)
 	if !memonly {
-		triedb.Commit(root, true)
+		triedb.Commit(root, true, nil)
 	}
 
 	trie, _ = New(root, triedb)