Merge pull request #5573 from ethereum-optimism/inphi/fpp-empty

op-e2e: Add FPP test for empty blocks

op-e2e/system_fpp_test.go

@@ -27,6 +27,119 @@ func TestVerifyL2OutputRootDetached(t *testing.T) {
 	testVerifyL2OutputRoot(t, true)
 }
 
+func TestVerifyL2OutputRootEmptyBlock(t *testing.T) {
+	testVerifyL2OutputRootEmptyBlock(t, false)
+}
+
+func TestVerifyL2OutputRootEmptyBlockDetached(t *testing.T) {
+	testVerifyL2OutputRootEmptyBlock(t, true)
+}
+
+// TestVerifyL2OutputRootEmptyBlock asserts that the program can verify the output root of an empty block
+// induced by missing batches.
+// Setup is as follows:
+// - create initial conditions and agreed l2 state
+// - stop the batch submitter to induce empty blocks
+// - wait for the seq window to expire so we can observe empty blocks
+// - select an empty block as our claim
+// - reboot the batch submitter
+// - update the state root via a tx
+// - run program
+func testVerifyL2OutputRootEmptyBlock(t *testing.T, detached bool) {
+	InitParallel(t)
+	ctx := context.Background()
+
+	cfg := DefaultSystemConfig(t)
+	// We don't need a verifier - just the sequencer is enough
+	delete(cfg.Nodes, "verifier")
+	// Use a small sequencer window size to avoid test timeout while waiting for empty blocks
+	// But not too small to ensure that our claim and subsequent state change is published
+	cfg.DeployConfig.SequencerWindowSize = 16
+
+	sys, err := cfg.Start()
+	require.Nil(t, err, "Error starting up system")
+	defer sys.Close()
+
+	log := testlog.Logger(t, log.LvlInfo)
+	log.Info("genesis", "l2", sys.RollupConfig.Genesis.L2, "l1", sys.RollupConfig.Genesis.L1, "l2_time", sys.RollupConfig.Genesis.L2Time)
+
+	l1Client := sys.Clients["l1"]
+	l2Seq := sys.Clients["sequencer"]
+	rollupRPCClient, err := rpc.DialContext(context.Background(), sys.RollupNodes["sequencer"].HTTPEndpoint())
+	require.Nil(t, err)
+	rollupClient := sources.NewRollupClient(client.NewBaseRPCClient(rollupRPCClient))
+
+	// Avoids flaky test by avoiding reorgs at epoch 0
+	t.Log("Wait for safe head to advance once for setup")
+	ss, err := l2Seq.BlockByNumber(ctx, big.NewInt(int64(rpc.SafeBlockNumber)))
+	require.NoError(t, err)
+	require.NoError(t, waitForSafeHead(ctx, ss.NumberU64()+cfg.DeployConfig.SequencerWindowSize+1, rollupClient))
+
+	t.Log("Sending transactions to setup existing state, prior to challenged period")
+	aliceKey := cfg.Secrets.Alice
+	receipt := SendL2Tx(t, cfg, l2Seq, aliceKey, func(opts *TxOpts) {
+		opts.ToAddr = &cfg.Secrets.Addresses().Bob
+		opts.Value = big.NewInt(1_000)
+	})
+	require.NoError(t, waitForSafeHead(ctx, receipt.BlockNumber.Uint64(), rollupClient))
+
+	t.Logf("Capture current L2 head as agreed starting point. l2Head=%x l2BlockNumber=%v", receipt.BlockHash, receipt.BlockNumber)
+	l2Head := receipt.BlockHash
+
+	t.Log("=====Stopping batch submitter=====")
+	err = sys.BatchSubmitter.Stop(ctx)
+	require.NoError(t, err, "could not stop batch submitter")
+
+	// Wait for the sequencer to catch up with the current L1 head so we know all submitted batches are processed
+	t.Log("Wait for sequencer to catch up with last submitted batch")
+	l1HeadNum, err := l1Client.BlockNumber(ctx)
+	require.NoError(t, err)
+	_, err = waitForL1OriginOnL2(l1HeadNum, l2Seq, 30*time.Second)
+	require.NoError(t, err)
+
+	// Get the current safe head now that the batcher is stopped
+	safeBlock, err := l2Seq.BlockByNumber(ctx, big.NewInt(int64(rpc.SafeBlockNumber)))
+	require.NoError(t, err)
+
+	// Wait for safe head to start advancing again when the sequencing window elapses, for at least three blocks
+	t.Log("Wait for safe head to advance after sequencing window elapses")
+	require.NoError(t, waitForSafeHead(ctx, safeBlock.NumberU64()+3, rollupClient))
+
+	// Use the 2nd empty block as our L2 claim block
+	t.Log("Determine L2 claim")
+	l2ClaimBlock, err := l2Seq.BlockByNumber(ctx, big.NewInt(int64(safeBlock.NumberU64()+2)))
+	require.NoError(t, err, "get L2 claim block number")
+	l2ClaimBlockNumber := l2ClaimBlock.Number().Uint64()
+	l2Output, err := rollupClient.OutputAtBlock(ctx, l2ClaimBlockNumber)
+	require.NoError(t, err, "could not get expected output")
+	l2Claim := l2Output.OutputRoot
+
+	t.Log("=====Restarting batch submitter=====")
+	err = sys.BatchSubmitter.Start()
+	require.NoError(t, err, "could not start batch submitter")
+
+	t.Log("Add a transaction to the next batch after sequence of empty blocks")
+	receipt = SendL2Tx(t, cfg, l2Seq, aliceKey, func(opts *TxOpts) {
+		opts.ToAddr = &cfg.Secrets.Addresses().Bob
+		opts.Value = big.NewInt(1_000)
+		opts.Nonce = 1
+	})
+	require.NoError(t, waitForSafeHead(ctx, receipt.BlockNumber.Uint64(), rollupClient))
+
+	t.Log("Determine L1 head that includes batch after sequence of empty blocks")
+	l1HeadBlock, err := l1Client.BlockByNumber(ctx, nil)
+	require.NoError(t, err, "get l1 head block")
+	l1Head := l1HeadBlock.Hash()
+
+	testFaultProofProgramScenario(t, ctx, sys, &FaultProofProgramTestScenario{
+		L1Head:             l1Head,
+		L2Head:             l2Head,
+		L2Claim:            common.Hash(l2Claim),
+		L2ClaimBlockNumber: l2ClaimBlockNumber,
+		Detached:           detached,
+	})
+}
+
 func testVerifyL2OutputRoot(t *testing.T, detached bool) {
 	InitParallel(t)
 	ctx := context.Background()
@@ -96,19 +209,39 @@ func testVerifyL2OutputRoot(t *testing.T, detached bool) {
 	require.NoError(t, err, "get l1 head block")
 	l1Head := l1HeadBlock.Hash()
 
+	testFaultProofProgramScenario(t, ctx, sys, &FaultProofProgramTestScenario{
+		L1Head:             l1Head,
+		L2Head:             l2Head,
+		L2Claim:            common.Hash(l2Claim),
+		L2ClaimBlockNumber: l2ClaimBlockNumber,
+		Detached:           detached,
+	})
+}
+
+type FaultProofProgramTestScenario struct {
+	L1Head             common.Hash
+	L2Head             common.Hash
+	L2Claim            common.Hash
+	L2ClaimBlockNumber uint64
+	Detached           bool
+}
+
+// testFaultProofProgramScenario runs the fault proof program in several contexts, given a test scenario.
+func testFaultProofProgramScenario(t *testing.T, ctx context.Context, sys *System, s *FaultProofProgramTestScenario) {
 	preimageDir := t.TempDir()
-	fppConfig := oppconf.NewConfig(sys.RollupConfig, sys.L2GenesisCfg.Config, l1Head, l2Head, common.Hash(l2Claim), l2ClaimBlockNumber)
+	fppConfig := oppconf.NewConfig(sys.RollupConfig, sys.L2GenesisCfg.Config, s.L1Head, s.L2Head, common.Hash(s.L2Claim), s.L2ClaimBlockNumber)
 	fppConfig.L1URL = sys.NodeEndpoint("l1")
 	fppConfig.L2URL = sys.NodeEndpoint("sequencer")
 	fppConfig.DataDir = preimageDir
-	if detached {
+	if s.Detached {
 		// When running in detached mode we need to compile the client executable since it will be called directly.
 		fppConfig.ExecCmd = BuildOpProgramClient(t)
 	}
 
 	// Check the FPP confirms the expected output
 	t.Log("Running fault proof in fetching mode")
-	err = opp.FaultProofProgram(ctx, log, fppConfig)
+	log := testlog.Logger(t, log.LvlInfo)
+	err := opp.FaultProofProgram(ctx, log, fppConfig)
 	require.NoError(t, err)
 
 	t.Log("Shutting down network")
@@ -131,7 +264,7 @@ func testVerifyL2OutputRoot(t *testing.T, detached bool) {
 	t.Log("Running fault proof with invalid claim")
 	fppConfig.L2Claim = common.Hash{0xaa}
 	err = opp.FaultProofProgram(ctx, log, fppConfig)
-	if detached {
+	if s.Detached {
 		require.Error(t, err, "exit status 1")
 	} else {
 		require.ErrorIs(t, err, driver.ErrClaimNotValid)
@@ -139,7 +272,7 @@ func testVerifyL2OutputRoot(t *testing.T, detached bool) {
 }
 
 func waitForSafeHead(ctx context.Context, safeBlockNum uint64, rollupClient *sources.RollupClient) error {
-	ctx, cancel := context.WithTimeout(ctx, 30*time.Second)
+	ctx, cancel := context.WithTimeout(ctx, 60*time.Second)
 	defer cancel()
 	for {
 		seqStatus, err := rollupClient.SyncStatus(ctx)