core: place a cap on reorglogs (ethereum#25711)

This PR makes the event-sending for deleted and new logs happen in batches, to prevent OOM situation due to large reorgs. 

Co-authored-by: Felix Lange <fjl@twurst.com>

Author: holiman

core/blockchain.go

@@ -2000,21 +2000,6 @@ func (bc *BlockChain) collectLogs(hash common.Hash, removed bool) []*types.Log {
 	return logs
 }
 
-// mergeLogs returns a merged log slice with specified sort order.
-func mergeLogs(logs [][]*types.Log, reverse bool) []*types.Log {
-	var ret []*types.Log
-	if reverse {
-		for i := len(logs) - 1; i >= 0; i-- {
-			ret = append(ret, logs[i]...)
-		}
-	} else {
-		for i := 0; i < len(logs); i++ {
-			ret = append(ret, logs[i]...)
-		}
-	}
-	return ret
-}
-
 // reorg takes two blocks, an old chain and a new chain and will reconstruct the
 // blocks and inserts them to be part of the new canonical chain and accumulates
 // potential missing transactions and post an event about them.
@@ -2028,9 +2013,6 @@ func (bc *BlockChain) reorg(oldBlock, newBlock *types.Block) error {
 
 		deletedTxs []common.Hash
 		addedTxs   []common.Hash
-
-		deletedLogs [][]*types.Log
-		rebirthLogs [][]*types.Log
 	)
 	// Reduce the longer chain to the same number as the shorter one
 	if oldBlock.NumberU64() > newBlock.NumberU64() {
@@ -2040,12 +2022,6 @@ func (bc *BlockChain) reorg(oldBlock, newBlock *types.Block) error {
 			for _, tx := range oldBlock.Transactions() {
 				deletedTxs = append(deletedTxs, tx.Hash())
 			}
-
-			// Collect deleted logs for notification
-			logs := bc.collectLogs(oldBlock.Hash(), true)
-			if len(logs) > 0 {
-				deletedLogs = append(deletedLogs, logs)
-			}
 		}
 	} else {
 		// New chain is longer, stash all blocks away for subsequent insertion
@@ -2072,12 +2048,6 @@ func (bc *BlockChain) reorg(oldBlock, newBlock *types.Block) error {
 		for _, tx := range oldBlock.Transactions() {
 			deletedTxs = append(deletedTxs, tx.Hash())
 		}
-
-		// Collect deleted logs for notification
-		logs := bc.collectLogs(oldBlock.Hash(), true)
-		if len(logs) > 0 {
-			deletedLogs = append(deletedLogs, logs)
-		}
 		newChain = append(newChain, newBlock)
 
 		// Step back with both chains
@@ -2151,28 +2121,42 @@ func (bc *BlockChain) reorg(oldBlock, newBlock *types.Block) error {
 		log.Crit("Failed to delete useless indexes", "err", err)
 	}
 
-	// Collect the logs
-	for i := len(newChain) - 1; i >= 1; i-- {
-		// Collect reborn logs due to chain reorg
-		logs := bc.collectLogs(newChain[i].Hash(), false)
-		if len(logs) > 0 {
-			rebirthLogs = append(rebirthLogs, logs)
+	// Send out events for logs from the old canon chain, and 'reborn'
+	// logs from the new canon chain. The number of logs can be very
+	// high, so the events are sent in batches of size around 512.
+
+	// Deleted logs + blocks:
+	var deletedLogs []*types.Log
+	for i := len(oldChain) - 1; i >= 0; i-- {
+		// Also send event for blocks removed from the canon chain.
+		bc.chainSideFeed.Send(ChainSideEvent{Block: oldChain[i]})
+
+		// Collect deleted logs for notification
+		if logs := bc.collectLogs(oldChain[i].Hash(), true); len(logs) > 0 {
+			deletedLogs = append(deletedLogs, logs...)
+		}
+		if len(deletedLogs) > 512 {
+			bc.rmLogsFeed.Send(RemovedLogsEvent{deletedLogs})
+			deletedLogs = nil
 		}
 	}
-	// If any logs need to be fired, do it now. In theory we could avoid creating
-	// this goroutine if there are no events to fire, but realistcally that only
-	// ever happens if we're reorging empty blocks, which will only happen on idle
-	// networks where performance is not an issue either way.
 	if len(deletedLogs) > 0 {
-		bc.rmLogsFeed.Send(RemovedLogsEvent{mergeLogs(deletedLogs, true)})
+		bc.rmLogsFeed.Send(RemovedLogsEvent{deletedLogs})
 	}
-	if len(rebirthLogs) > 0 {
-		bc.logsFeed.Send(mergeLogs(rebirthLogs, false))
-	}
-	if len(oldChain) > 0 {
-		for i := len(oldChain) - 1; i >= 0; i-- {
-			bc.chainSideFeed.Send(ChainSideEvent{Block: oldChain[i]})
+
+	// New logs:
+	var rebirthLogs []*types.Log
+	for i := len(newChain) - 1; i >= 1; i-- {
+		if logs := bc.collectLogs(newChain[i].Hash(), false); len(logs) > 0 {
+			rebirthLogs = append(rebirthLogs, logs...)
 		}
+		if len(rebirthLogs) > 512 {
+			bc.logsFeed.Send(rebirthLogs)
+			rebirthLogs = nil
+		}
+	}
+	if len(rebirthLogs) > 0 {
+		bc.logsFeed.Send(rebirthLogs)
 	}
 	return nil
 }

core/blockchain_test.go

@@ -1158,37 +1158,53 @@ func TestLogRebirth(t *testing.T) {
 	blockchain.SubscribeLogsEvent(newLogCh)
 	blockchain.SubscribeRemovedLogsEvent(rmLogsCh)
 
-	// This chain contains a single log.
-	genDb, chain, _ := GenerateChainWithGenesis(gspec, engine, 2, func(i int, gen *BlockGen) {
-		if i == 1 {
-			tx, err := types.SignTx(types.NewContractCreation(gen.TxNonce(addr1), new(big.Int), 1000000, gen.header.BaseFee, logCode), signer, key1)
-			if err != nil {
-				t.Fatalf("failed to create tx: %v", err)
+	// This chain contains 10 logs.
+	genDb, chain, _ := GenerateChainWithGenesis(gspec, engine, 3, func(i int, gen *BlockGen) {
+		if i < 2 {
+			for ii := 0; ii < 5; ii++ {
+				tx, err := types.SignNewTx(key1, signer, &types.LegacyTx{
+					Nonce:    gen.TxNonce(addr1),
+					GasPrice: gen.header.BaseFee,
+					Gas:      uint64(1000001),
+					Data:     logCode,
+				})
+				if err != nil {
+					t.Fatalf("failed to create tx: %v", err)
+				}
+				gen.AddTx(tx)
 			}
-			gen.AddTx(tx)
 		}
 	})
 	if _, err := blockchain.InsertChain(chain); err != nil {
 		t.Fatalf("failed to insert chain: %v", err)
 	}
-	checkLogEvents(t, newLogCh, rmLogsCh, 1, 0)
+	checkLogEvents(t, newLogCh, rmLogsCh, 10, 0)
 
-	// Generate long reorg chain containing another log. Inserting the
-	// chain removes one log and adds one.
-	_, forkChain, _ := GenerateChainWithGenesis(gspec, engine, 2, func(i int, gen *BlockGen) {
-		if i == 1 {
-			tx, err := types.SignTx(types.NewContractCreation(gen.TxNonce(addr1), new(big.Int), 1000000, gen.header.BaseFee, logCode), signer, key1)
+	// Generate long reorg chain containing more logs. Inserting the
+	// chain removes one log and adds four.
+	_, forkChain, _ := GenerateChainWithGenesis(gspec, engine, 3, func(i int, gen *BlockGen) {
+		if i == 2 {
+			// The last (head) block is not part of the reorg-chain, we can ignore it
+			return
+		}
+		for ii := 0; ii < 5; ii++ {
+			tx, err := types.SignNewTx(key1, signer, &types.LegacyTx{
+				Nonce:    gen.TxNonce(addr1),
+				GasPrice: gen.header.BaseFee,
+				Gas:      uint64(1000000),
+				Data:     logCode,
+			})
 			if err != nil {
 				t.Fatalf("failed to create tx: %v", err)
 			}
 			gen.AddTx(tx)
-			gen.OffsetTime(-9) // higher block difficulty
 		}
+		gen.OffsetTime(-9) // higher block difficulty
 	})
 	if _, err := blockchain.InsertChain(forkChain); err != nil {
 		t.Fatalf("failed to insert forked chain: %v", err)
 	}
-	checkLogEvents(t, newLogCh, rmLogsCh, 1, 1)
+	checkLogEvents(t, newLogCh, rmLogsCh, 10, 10)
 
 	// This chain segment is rooted in the original chain, but doesn't contain any logs.
 	// When inserting it, the canonical chain switches away from forkChain and re-emits
@@ -1197,7 +1213,7 @@ func TestLogRebirth(t *testing.T) {
 	if _, err := blockchain.InsertChain(newBlocks); err != nil {
 		t.Fatalf("failed to insert forked chain: %v", err)
 	}
-	checkLogEvents(t, newLogCh, rmLogsCh, 1, 1)
+	checkLogEvents(t, newLogCh, rmLogsCh, 10, 10)
 }
 
 // This test is a variation of TestLogRebirth. It verifies that log events are emitted
@@ -1252,19 +1268,43 @@ func TestSideLogRebirth(t *testing.T) {
 
 func checkLogEvents(t *testing.T, logsCh <-chan []*types.Log, rmLogsCh <-chan RemovedLogsEvent, wantNew, wantRemoved int) {
 	t.Helper()
-
-	if len(logsCh) != wantNew {
-		t.Fatalf("wrong number of log events: got %d, want %d", len(logsCh), wantNew)
-	}
-	if len(rmLogsCh) != wantRemoved {
-		t.Fatalf("wrong number of removed log events: got %d, want %d", len(rmLogsCh), wantRemoved)
-	}
+	var (
+		countNew int
+		countRm  int
+		prev     int
+	)
 	// Drain events.
-	for i := 0; i < len(logsCh); i++ {
-		<-logsCh
+	for len(logsCh) > 0 {
+		x := <-logsCh
+		countNew += len(x)
+		for _, log := range x {
+			// We expect added logs to be in ascending order: 0:0, 0:1, 1:0 ...
+			have := 100*int(log.BlockNumber) + int(log.TxIndex)
+			if have < prev {
+				t.Fatalf("Expected new logs to arrive in ascending order (%d < %d)", have, prev)
+			}
+			prev = have
+		}
+	}
+	prev = 0
+	for len(rmLogsCh) > 0 {
+		x := <-rmLogsCh
+		countRm += len(x.Logs)
+		for _, log := range x.Logs {
+			// We expect removed logs to be in ascending order: 0:0, 0:1, 1:0 ...
+			have := 100*int(log.BlockNumber) + int(log.TxIndex)
+			if have < prev {
+				t.Fatalf("Expected removed logs to arrive in ascending order (%d < %d)", have, prev)
+			}
+			prev = have
+		}
+	}
+
+	if countNew != wantNew {
+		t.Fatalf("wrong number of log events: got %d, want %d", countNew, wantNew)
 	}
-	for i := 0; i < len(rmLogsCh); i++ {
-		<-rmLogsCh
+	if countRm != wantRemoved {
+		t.Fatalf("wrong number of removed log events: got %d, want %d", countRm, wantRemoved)
 	}
 }