sink(cdc): always handle sink failures for cases with sync-point enabled (#10132)

cdc/processor/sinkmanager/manager.go

@@ -530,13 +530,50 @@ func (m *SinkManager) generateSinkTasks(ctx context.Context) error {
 			slowestTableProgress := progs[i]
 			lowerBound := slowestTableProgress.nextLowerBoundPos
 			upperBound := m.getUpperBound(tableSink.getUpperBoundTs())
-			// The table has no available progress.
-			if lowerBound.Compare(upperBound) >= 0 {
+
+			if !tableSink.initTableSink() {
+				// The table hasn't been attached to a sink.
 				m.sinkProgressHeap.push(slowestTableProgress)
 				continue
 			}
-			// The table hasn't been attached to a sink.
-			if !tableSink.initTableSink() {
+
+			if sinkErr := tableSink.checkTableSinkHealth(); sinkErr != nil {
+				switch errors.Cause(sinkErr).(type) {
+				case tablesink.SinkInternalError:
+					tableSink.closeAndClearTableSink()
+					if restartErr := tableSink.restart(ctx); restartErr == nil {
+						// Restart the table sink based on the checkpoint position.
+						ckpt := tableSink.getCheckpointTs().ResolvedMark()
+						lastWrittenPos := engine.Position{StartTs: ckpt - 1, CommitTs: ckpt}
+						p := &progress{
+							span:              tableSink.span,
+							nextLowerBoundPos: lastWrittenPos.Next(),
+							version:           slowestTableProgress.version,
+						}
+						m.sinkProgressHeap.push(p)
+						log.Info("table sink has been restarted",
+							zap.String("namespace", m.changefeedID.Namespace),
+							zap.String("changefeed", m.changefeedID.ID),
+							zap.Stringer("span", &tableSink.span),
+							zap.Any("lastWrittenPos", lastWrittenPos),
+							zap.String("sinkError", sinkErr.Error()))
+					} else {
+						m.sinkProgressHeap.push(slowestTableProgress)
+						log.Warn("table sink restart fail",
+							zap.String("namespace", m.changefeedID.Namespace),
+							zap.String("changefeed", m.changefeedID.ID),
+							zap.Stringer("span", &tableSink.span),
+							zap.String("sinkError", sinkErr.Error()),
+							zap.Error(restartErr))
+					}
+				default:
+					return sinkErr
+				}
+				continue
+			}
+
+			// The table has no available progress.
+			if lowerBound.Compare(upperBound) >= 0 {
 				m.sinkProgressHeap.push(slowestTableProgress)
 				continue
 			}

cdc/processor/sinkmanager/manager_test.go

@@ -372,3 +372,45 @@ func TestSinkManagerNeedsStuckCheck(t *testing.T) {
 
 	require.False(t, manager.needsStuckCheck())
 }
+
+func TestSinkManagerRestartTableSinks(t *testing.T) {
+	failpoint.Enable("github.com/pingcap/tiflow/cdc/processor/sinkmanager/SinkWorkerTaskHandlePause", "return")
+	defer failpoint.Disable("github.com/pingcap/tiflow/cdc/processor/sinkmanager/SinkWorkerTaskHandlePause")
+
+	ctx, cancel := context.WithCancel(context.Background())
+	errCh := make(chan error, 16)
+	changefeedInfo := getChangefeedInfo()
+	manager, _, _ := CreateManagerWithMemEngine(t, ctx, model.ChangeFeedID{}, changefeedInfo, errCh)
+	defer func() {
+		cancel()
+		manager.Close()
+	}()
+
+	span := tablepb.Span{TableID: 1}
+	manager.AddTable(span, 1, 100)
+	require.Nil(t, manager.StartTable(span, 2))
+	table, exists := manager.tableSinks.Load(span)
+	require.True(t, exists)
+
+	table.(*tableSinkWrapper).updateReceivedSorterResolvedTs(4)
+	table.(*tableSinkWrapper).updateBarrierTs(4)
+	select {
+	case task := <-manager.sinkTaskChan:
+		require.Equal(t, engine.Position{StartTs: 0, CommitTs: 3}, task.lowerBound)
+		task.callback(engine.Position{StartTs: 3, CommitTs: 4})
+	case <-time.After(2 * time.Second):
+		panic("should always get a sink task")
+	}
+
+	// With the failpoint blackhole/WriteEventsFail enabled, sink manager should restarts
+	// the table sink at its checkpoint.
+	failpoint.Enable("github.com/pingcap/tiflow/cdc/sink/dmlsink/blackhole/WriteEventsFail", "1*return")
+	defer failpoint.Disable("github.com/pingcap/tiflow/cdc/sink/dmlsink/blackhole/WriteEventsFail")
+	select {
+	case task := <-manager.sinkTaskChan:
+		require.Equal(t, engine.Position{StartTs: 2, CommitTs: 2}, task.lowerBound)
+		task.callback(engine.Position{StartTs: 3, CommitTs: 4})
+	case <-time.After(2 * time.Second):
+		panic("should always get a sink task")
+	}
+}

cdc/processor/sinkmanager/table_sink_worker.go

@@ -96,6 +96,7 @@ func newSinkWorker(
 }
 
 func (w *sinkWorker) handleTasks(ctx context.Context, taskChan <-chan *sinkTask) error {
+	failpoint.Inject("SinkWorkerTaskHandlePause", func() { <-ctx.Done() })
 	for {
 		select {
 		case <-ctx.Done():
@@ -169,25 +170,11 @@ func (w *sinkWorker) handleTask(ctx context.Context, task *sinkTask) (finalErr e
 			// events have been reported. Then we can continue the table
 			// at the checkpoint position.
 			case tablesink.SinkInternalError:
-				task.tableSink.closeAndClearTableSink()
 				// After the table sink is cleared all pending events are sent out or dropped.
 				// So we can re-add the table into sinkMemQuota.
 				w.sinkMemQuota.ClearTable(task.tableSink.span)
-
-				// Restart the table sink based on the checkpoint position.
-				if err := task.tableSink.restart(ctx); err == nil {
-					checkpointTs := task.tableSink.getCheckpointTs()
-					ckpt := checkpointTs.ResolvedMark()
-					lastWrittenPos := engine.Position{StartTs: ckpt - 1, CommitTs: ckpt}
-					performCallback(lastWrittenPos)
-					log.Info("table sink has been restarted",
-						zap.String("namespace", w.changefeedID.Namespace),
-						zap.String("changefeed", w.changefeedID.ID),
-						zap.Stringer("span", &task.span),
-						zap.Any("lastWrittenPos", lastWrittenPos),
-						zap.String("sinkError", finalErr.Error()))
-					finalErr = err
-				}
+				performCallback(advancer.lastPos)
+				finalErr = nil
 			default:
 			}
 		}

cdc/processor/sinkmanager/table_sink_wrapper.go

@@ -166,6 +166,11 @@ func (t *tableSinkWrapper) start(ctx context.Context, startTs model.Ts) (err err
 			break
 		}
 	}
+	if model.NewResolvedTs(startTs).Greater(t.tableSink.checkpointTs) {
+		t.tableSink.checkpointTs = model.NewResolvedTs(startTs)
+		t.tableSink.resolvedTs = model.NewResolvedTs(startTs)
+		t.tableSink.advanced = time.Now()
+	}
 	t.state.Store(tablepb.TableStateReplicating)
 	return nil
 }
@@ -363,6 +368,15 @@ func (t *tableSinkWrapper) doTableSinkClear() {
 	t.tableSink.version = 0
 }
 
+func (t *tableSinkWrapper) checkTableSinkHealth() (err error) {
+	t.tableSink.RLock()
+	defer t.tableSink.RUnlock()
+	if t.tableSink.s != nil {
+		err = t.tableSink.s.CheckHealth()
+	}
+	return
+}
+
 // When the attached sink fail, there can be some events that have already been
 // committed at downstream but we don't know. So we need to update `replicateTs`
 // of the table so that we can re-send those events later.

cdc/sink/dmlsink/blackhole/black_hole_dml_sink.go

@@ -14,6 +14,8 @@
 package blackhole
 
 import (
+	"github.com/pingcap/errors"
+	"github.com/pingcap/failpoint"
 	"github.com/pingcap/log"
 	"github.com/pingcap/tiflow/cdc/model"
 	"github.com/pingcap/tiflow/cdc/sink/dmlsink"
@@ -33,14 +35,16 @@ func NewDMLSink() *DMLSink {
 }
 
 // WriteEvents log the events.
-func (s *DMLSink) WriteEvents(rows ...*dmlsink.CallbackableEvent[*model.RowChangedEvent]) error {
-	for _, row := range rows {
-		// NOTE: don't change the log, some tests depend on it.
-		log.Debug("BlackHoleSink: WriteEvents", zap.Any("row", row.Event))
-		row.Callback()
+func (s *DMLSink) WriteEvents(rows ...*dmlsink.CallbackableEvent[*model.RowChangedEvent]) (err error) {
+	failpoint.Inject("WriteEventsFail", func() { err = errors.New("InjectedErrorForWriteEventsFail") })
+	if err == nil {
+		for _, row := range rows {
+			// NOTE: don't change the log, some tests depend on it.
+			log.Debug("BlackHoleSink: WriteEvents", zap.Any("row", row.Event))
+			row.Callback()
+		}
 	}
-
-	return nil
+	return
 }
 
 // Close do nothing.

cdc/sink/tablesink/table_sink.go

@@ -37,6 +37,8 @@ type TableSink interface {
 	Close()
 	// AsyncClose closes the table sink asynchronously. Returns true if it's closed.
 	AsyncClose() bool
+	// CheckHealth checks whether the associated sink backend is healthy or not.
+	CheckHealth() error
 }
 
 // SinkInternalError means the error comes from sink internal.

cdc/sink/tablesink/table_sink_impl.go

@@ -158,6 +158,14 @@ func (e *EventTableSink[E, P]) AsyncClose() bool {
 	return false
 }
 
+// CheckHealth checks whether the associated sink backend is healthy or not.
+func (e *EventTableSink[E, P]) CheckHealth() error {
+	if err := e.backendSink.WriteEvents(); err != nil {
+		return SinkInternalError{err}
+	}
+	return nil
+}
+
 func (e *EventTableSink[E, P]) freeze() {
 	// Notice: We have to set the state to stopping first,
 	// otherwise the progressTracker may be advanced incorrectly.