GJL commented on a change in pull request #9663: [WIP][FLINK-12433][runtime]
Implement DefaultScheduler stub
URL: https://github.com/apache/flink/pull/9663#discussion_r327048907
##########
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 think this line should be removed. In legacy mode, only when we fail the
cancellation in `ExecutionGraph#failGlobal()`, the job should be transitioned
to `FAILED`.
----------------------------------------------------------------
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
