zhuzhurk commented on a change in pull request #9663:
[WIP][FLINK-12433][runtime] Implement DefaultScheduler stub
URL: https://github.com/apache/flink/pull/9663#discussion_r327418704
##########
File path:
flink-runtime/src/main/java/org/apache/flink/runtime/scheduler/DefaultScheduler.java
##########
@@ -75,10 +128,293 @@ public DefaultScheduler(
slotRequestTimeout,
shuffleMaster,
partitionTracker);
+
+ this.log = log;
+
+ this.delayExecutor = checkNotNull(delayExecutor);
+ this.userCodeLoader = checkNotNull(userCodeLoader);
+ this.executionVertexOperations =
checkNotNull(executionVertexOperations);
+ this.executionVertexVersioner =
checkNotNull(executionVertexVersioner);
+
+ this.executionFailureHandler = new
ExecutionFailureHandler(failoverStrategyFactory.create(getFailoverTopology()),
restartBackoffTimeStrategy);
+ this.schedulingStrategy =
schedulingStrategyFactory.createInstance(this, getSchedulingTopology(),
getJobGraph());
+ this.executionSlotAllocator = new
DefaultExecutionSlotAllocator(slotProvider, getInputsLocationsRetriever(),
slotRequestTimeout);
+ setTaskFailureListener(new
UpdateSchedulerNgOnInternalTaskFailuresListener(this,
getJobGraph().getJobID()));
+ }
+
+ //
------------------------------------------------------------------------
+ // SchedulerNG
+ //
------------------------------------------------------------------------
+
+ @Override
+ protected void startSchedulingInternal() {
+ prepareExecutionGraphForScheduling();
+ schedulingStrategy.startScheduling();
+ }
+
+ @Override
+ public boolean updateTaskExecutionState(final TaskExecutionState
taskExecutionState) {
+ final Optional<ExecutionVertexID> executionVertexIdOptional =
getExecutionVertexId(taskExecutionState.getID());
+ if (executionVertexIdOptional.isPresent()) {
+ final ExecutionVertexID executionVertexId =
executionVertexIdOptional.get();
+ updateState(taskExecutionState);
+
schedulingStrategy.onExecutionStateChange(executionVertexId,
taskExecutionState.getExecutionState());
+ maybeHandleTaskFailure(taskExecutionState,
executionVertexId);
+ return true;
+ }
+
+ return false;
+ }
+
+ private void maybeHandleTaskFailure(final TaskExecutionState
taskExecutionState, final ExecutionVertexID executionVertexId) {
+ if (taskExecutionState.getExecutionState() ==
ExecutionState.FAILED) {
+ final Throwable error =
taskExecutionState.getError(userCodeLoader);
+ handleTaskFailure(executionVertexId, error);
+ }
+ }
+
+ private void handleTaskFailure(final ExecutionVertexID
executionVertexId, final Throwable error) {
+ final FailureHandlingResult failureHandlingResult =
executionFailureHandler.getFailureHandlingResult(executionVertexId, error);
+ maybeRestartTasks(failureHandlingResult);
+ }
+
+ private void maybeRestartTasks(final FailureHandlingResult
failureHandlingResult) {
+ if (failureHandlingResult.canRestart()) {
+ restartTasksWithDelay(failureHandlingResult);
+ } else {
+ failJob(failureHandlingResult.getError());
+ }
+ }
+
+ private void restartTasksWithDelay(final FailureHandlingResult
failureHandlingResult) {
+ final Set<ExecutionVertexID> verticesToRestart =
failureHandlingResult.getVerticesToRestart();
+
+ final Set<ExecutionVertexVersion> executionVertexVersions =
+ new
HashSet<>(executionVertexVersioner.recordVertexModifications(verticesToRestart).values());
+
+ final CompletableFuture<?> cancelFuture =
cancelTasksAsync(verticesToRestart);
+
+ delayExecutor.schedule(
+ () -> FutureUtils.assertNoException(
+
cancelFuture.handleAsync(restartTasksOrHandleError(executionVertexVersions),
getMainThreadExecutor())),
+ failureHandlingResult.getRestartDelayMS(),
+ TimeUnit.MILLISECONDS);
+ }
+
+ private BiFunction<Object, Throwable, Void>
restartTasksOrHandleError(final Set<ExecutionVertexVersion>
executionVertexVersions) {
+ return (Object ignored, Throwable throwable) -> {
+
+ if (throwable == null) {
+ final Set<ExecutionVertexID> verticesToRestart
=
executionVertexVersioner.getUnmodifiedExecutionVertices(executionVertexVersions);
+
schedulingStrategy.restartTasks(verticesToRestart);
+ } else {
+ failJob(throwable);
Review comment:
I tried to track the cancellation future, it it `Execution#releaseFuture`
and it is only completed exceptionally when `Execution#terminalStateFuture` is
completed exceptionally, which should not happen from the code.
The cancellation future will wait without a timeout. But usually it
completes with success cancellation, or the TM heartbeat timed out. There is a
`TaskCancelerWatchDog` at TM side, which will cause the TM to exit if it fails
to cancel tasks before `"task.cancellation.timeout"`.
So seems the cancellation failure handling should not matter, since it is
not expected to happen.
And agreed that transitioning the job to `FAILED` should be our last choice.
We should do it only when the restart strategy says so or we are sure that the
job cannot be recovered through a global/JM failover.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
With regards,
Apache Git Services
