Search in sources :

Example 1 with RootExceptionHistoryEntry

use of org.apache.flink.runtime.scheduler.exceptionhistory.RootExceptionHistoryEntry in project flink by apache.

the class DefaultSchedulerTest method testExceptionHistoryConcurrentRestart.

@Test
public void testExceptionHistoryConcurrentRestart() throws Exception {
    final JobGraph jobGraph = singleJobVertexJobGraph(2);
    final TaskManagerLocation taskManagerLocation = new LocalTaskManagerLocation();
    final TestingLogicalSlotBuilder logicalSlotBuilder = new TestingLogicalSlotBuilder();
    logicalSlotBuilder.setTaskManagerLocation(taskManagerLocation);
    executionSlotAllocatorFactory = new TestExecutionSlotAllocatorFactory(logicalSlotBuilder);
    final ReorganizableManuallyTriggeredScheduledExecutor delayExecutor = new ReorganizableManuallyTriggeredScheduledExecutor();
    final TestFailoverStrategyFactory failoverStrategyFactory = new TestFailoverStrategyFactory();
    final DefaultScheduler scheduler = createScheduler(jobGraph, ComponentMainThreadExecutorServiceAdapter.forMainThread(), new PipelinedRegionSchedulingStrategy.Factory(), failoverStrategyFactory, delayExecutor);
    scheduler.startScheduling();
    final ExecutionVertex executionVertex0 = Iterables.get(scheduler.getExecutionGraph().getAllExecutionVertices(), 0);
    final ExecutionVertex executionVertex1 = Iterables.get(scheduler.getExecutionGraph().getAllExecutionVertices(), 1);
    // single-ExecutionVertex failure
    final RuntimeException exception0 = new RuntimeException("failure #0");
    failoverStrategyFactory.setTasksToRestart(executionVertex0.getID());
    final long updateStateTriggeringRestartTimestamp0 = initiateFailure(scheduler, executionVertex0.getCurrentExecutionAttempt().getAttemptId(), exception0);
    // multi-ExecutionVertex failure
    final RuntimeException exception1 = new RuntimeException("failure #1");
    failoverStrategyFactory.setTasksToRestart(executionVertex1.getID(), executionVertex0.getID());
    final long updateStateTriggeringRestartTimestamp1 = initiateFailure(scheduler, executionVertex1.getCurrentExecutionAttempt().getAttemptId(), exception1);
    // there might be a race condition with the delayExecutor if the tasks are scheduled quite
    // close to each other which we want to simulate here
    Collections.reverse(delayExecutor.getCollectedScheduledTasks());
    delayExecutor.triggerNonPeriodicScheduledTasks();
    assertThat(scheduler.getExceptionHistory(), IsIterableWithSize.iterableWithSize(2));
    final Iterator<RootExceptionHistoryEntry> actualExceptionHistory = scheduler.getExceptionHistory().iterator();
    final RootExceptionHistoryEntry entry0 = actualExceptionHistory.next();
    assertThat(entry0, is(ExceptionHistoryEntryMatcher.matchesFailure(exception0, updateStateTriggeringRestartTimestamp0, executionVertex0.getTaskNameWithSubtaskIndex(), executionVertex0.getCurrentAssignedResourceLocation())));
    assertThat(entry0.getConcurrentExceptions(), IsIterableContainingInOrder.contains(ExceptionHistoryEntryMatcher.matchesFailure(exception1, updateStateTriggeringRestartTimestamp1, executionVertex1.getTaskNameWithSubtaskIndex(), executionVertex1.getCurrentAssignedResourceLocation())));
    final RootExceptionHistoryEntry entry1 = actualExceptionHistory.next();
    assertThat(entry1, is(ExceptionHistoryEntryMatcher.matchesFailure(exception1, updateStateTriggeringRestartTimestamp1, executionVertex1.getTaskNameWithSubtaskIndex(), executionVertex1.getCurrentAssignedResourceLocation())));
    assertThat(entry1.getConcurrentExceptions(), IsEmptyIterable.emptyIterable());
}
Also used : RootExceptionHistoryEntry(org.apache.flink.runtime.scheduler.exceptionhistory.RootExceptionHistoryEntry) TestFailoverStrategyFactory(org.apache.flink.runtime.executiongraph.utils.TestFailoverStrategyFactory) LocalTaskManagerLocation(org.apache.flink.runtime.taskmanager.LocalTaskManagerLocation) TaskManagerLocation(org.apache.flink.runtime.taskmanager.TaskManagerLocation) ArchivedExecutionVertex(org.apache.flink.runtime.executiongraph.ArchivedExecutionVertex) ExecutionVertex(org.apache.flink.runtime.executiongraph.ExecutionVertex) SchedulingExecutionVertex(org.apache.flink.runtime.scheduler.strategy.SchedulingExecutionVertex) JobGraph(org.apache.flink.runtime.jobgraph.JobGraph) LocalTaskManagerLocation(org.apache.flink.runtime.taskmanager.LocalTaskManagerLocation) TestingLogicalSlotBuilder(org.apache.flink.runtime.jobmaster.TestingLogicalSlotBuilder) PipelinedRegionSchedulingStrategy(org.apache.flink.runtime.scheduler.strategy.PipelinedRegionSchedulingStrategy) AdaptiveSchedulerTest(org.apache.flink.runtime.scheduler.adaptive.AdaptiveSchedulerTest) Test(org.junit.Test)

Example 2 with RootExceptionHistoryEntry

use of org.apache.flink.runtime.scheduler.exceptionhistory.RootExceptionHistoryEntry in project flink by apache.

the class AdaptiveSchedulerTest method testExceptionHistoryWithTaskFailure.

@Test
public void testExceptionHistoryWithTaskFailure() throws Exception {
    final Exception expectedException = new Exception("Expected Local Exception");
    BiConsumer<AdaptiveScheduler, List<ExecutionAttemptID>> testLogic = (scheduler, attemptIds) -> {
        final ExecutionAttemptID attemptId = attemptIds.get(1);
        scheduler.updateTaskExecutionState(new TaskExecutionStateTransition(new TaskExecutionState(attemptId, ExecutionState.FAILED, expectedException)));
    };
    final Iterable<RootExceptionHistoryEntry> actualExceptionHistory = runExceptionHistoryTests(testLogic);
    assertThat(actualExceptionHistory).hasSize(1);
    final RootExceptionHistoryEntry failure = actualExceptionHistory.iterator().next();
    assertThat(failure.getException().deserializeError(classLoader)).isEqualTo(expectedException);
}
Also used : Arrays(java.util.Arrays) ResourceRequirement(org.apache.flink.runtime.slots.ResourceRequirement) TaskNotRunningException(org.apache.flink.runtime.operators.coordination.TaskNotRunningException) ArchivedExecution(org.apache.flink.runtime.executiongraph.ArchivedExecution) TestingSlotAllocator(org.apache.flink.runtime.scheduler.adaptive.allocator.TestingSlotAllocator) CheckpointException(org.apache.flink.runtime.checkpoint.CheckpointException) ResultPartitionID(org.apache.flink.runtime.io.network.partition.ResultPartitionID) TestingFatalErrorHandler(org.apache.flink.runtime.util.TestingFatalErrorHandler) MetricRegistry(org.apache.flink.runtime.metrics.MetricRegistry) Duration(java.time.Duration) ClassRule(org.junit.ClassRule) TestingCheckpointRecoveryFactory(org.apache.flink.runtime.checkpoint.TestingCheckpointRecoveryFactory) ManuallyTriggeredComponentMainThreadExecutor(org.apache.flink.runtime.concurrent.ManuallyTriggeredComponentMainThreadExecutor) BlockingQueue(java.util.concurrent.BlockingQueue) MetricOptions(org.apache.flink.configuration.MetricOptions) JobManagerOptions(org.apache.flink.configuration.JobManagerOptions) Executors(java.util.concurrent.Executors) MetricNames(org.apache.flink.runtime.metrics.MetricNames) ArrayBlockingQueue(java.util.concurrent.ArrayBlockingQueue) VertexParallelismStore(org.apache.flink.runtime.scheduler.VertexParallelismStore) Time(org.apache.flink.api.common.time.Time) RootExceptionHistoryEntry(org.apache.flink.runtime.scheduler.exceptionhistory.RootExceptionHistoryEntry) FlinkException(org.apache.flink.util.FlinkException) ComponentMainThreadExecutor(org.apache.flink.runtime.concurrent.ComponentMainThreadExecutor) UpTimeGauge(org.apache.flink.runtime.executiongraph.metrics.UpTimeGauge) ExceptionHistoryEntry(org.apache.flink.runtime.scheduler.exceptionhistory.ExceptionHistoryEntry) LocalTaskManagerLocation(org.apache.flink.runtime.taskmanager.LocalTaskManagerLocation) ResourceCounter(org.apache.flink.runtime.util.ResourceCounter) JobStatus(org.apache.flink.api.common.JobStatus) DefaultAllocatedSlotPool(org.apache.flink.runtime.jobmaster.slotpool.DefaultAllocatedSlotPool) ArrayList(java.util.ArrayList) SchedulerNG(org.apache.flink.runtime.scheduler.SchedulerNG) DownTimeGauge(org.apache.flink.runtime.executiongraph.metrics.DownTimeGauge) PartitionProducerDisposedException(org.apache.flink.runtime.jobmanager.PartitionProducerDisposedException) Gauge(org.apache.flink.metrics.Gauge) ScheduledExecutorService(java.util.concurrent.ScheduledExecutorService) BiConsumer(java.util.function.BiConsumer) FixedDelayRestartBackoffTimeStrategy(org.apache.flink.runtime.executiongraph.failover.flip1.FixedDelayRestartBackoffTimeStrategy) Nullable(javax.annotation.Nullable) ArchivedExecutionVertex(org.apache.flink.runtime.executiongraph.ArchivedExecutionVertex) TestExecutorResource(org.apache.flink.testutils.executor.TestExecutorResource) ExecutionState(org.apache.flink.runtime.execution.ExecutionState) CheckpointsCleaner(org.apache.flink.runtime.checkpoint.CheckpointsCleaner) TestOperatorEvent(org.apache.flink.runtime.operators.coordination.TestOperatorEvent) Test(org.junit.Test) IOException(java.io.IOException) IterableUtils(org.apache.flink.util.IterableUtils) SimpleAckingTaskManagerGateway(org.apache.flink.runtime.executiongraph.utils.SimpleAckingTaskManagerGateway) ExecutionException(java.util.concurrent.ExecutionException) TaskExecutionStateTransition(org.apache.flink.runtime.executiongraph.TaskExecutionStateTransition) JobID(org.apache.flink.api.common.JobID) NoRestartBackoffTimeStrategy(org.apache.flink.runtime.executiongraph.failover.flip1.NoRestartBackoffTimeStrategy) SlotPoolTestUtils.offerSlots(org.apache.flink.runtime.jobmaster.slotpool.SlotPoolTestUtils.offerSlots) JobManagerMetricGroup(org.apache.flink.runtime.metrics.groups.JobManagerMetricGroup) ComponentMainThreadExecutorServiceAdapter(org.apache.flink.runtime.concurrent.ComponentMainThreadExecutorServiceAdapter) Assertions.assertThat(org.assertj.core.api.Assertions.assertThat) JobGraph(org.apache.flink.runtime.jobgraph.JobGraph) CheckpointCoordinatorConfiguration(org.apache.flink.runtime.jobgraph.tasks.CheckpointCoordinatorConfiguration) SuppressRestartsException(org.apache.flink.runtime.execution.SuppressRestartsException) ArchivedExecutionGraphTest(org.apache.flink.runtime.executiongraph.ArchivedExecutionGraphTest) AtomicInteger(java.util.concurrent.atomic.AtomicInteger) TestLogger(org.apache.flink.util.TestLogger) JobCheckpointingSettings(org.apache.flink.runtime.jobgraph.tasks.JobCheckpointingSettings) CheckpointIDCounter(org.apache.flink.runtime.checkpoint.CheckpointIDCounter) ArchivedExecutionJobVertex(org.apache.flink.runtime.executiongraph.ArchivedExecutionJobVertex) TestingCompletedCheckpointStore(org.apache.flink.runtime.checkpoint.TestingCompletedCheckpointStore) TestRestartBackoffTimeStrategy(org.apache.flink.runtime.executiongraph.failover.flip1.TestRestartBackoffTimeStrategy) IntermediateDataSetID(org.apache.flink.runtime.jobgraph.IntermediateDataSetID) Collectors(java.util.stream.Collectors) ResourceProfile(org.apache.flink.runtime.clusterframework.types.ResourceProfile) DefaultDeclarativeSlotPool(org.apache.flink.runtime.jobmaster.slotpool.DefaultDeclarativeSlotPool) List(java.util.List) StandaloneCheckpointIDCounter(org.apache.flink.runtime.checkpoint.StandaloneCheckpointIDCounter) CoordinationRequest(org.apache.flink.runtime.operators.coordination.CoordinationRequest) OperatorID(org.apache.flink.runtime.jobgraph.OperatorID) Optional(java.util.Optional) SchedulerExecutionMode(org.apache.flink.configuration.SchedulerExecutionMode) KeyGroupRangeAssignment(org.apache.flink.runtime.state.KeyGroupRangeAssignment) OneShotLatch(org.apache.flink.core.testutils.OneShotLatch) SchedulerBase(org.apache.flink.runtime.scheduler.SchedulerBase) SavepointFormatType(org.apache.flink.core.execution.SavepointFormatType) JobVertex(org.apache.flink.runtime.jobgraph.JobVertex) AtomicBoolean(java.util.concurrent.atomic.AtomicBoolean) CompletableFuture(java.util.concurrent.CompletableFuture) VertexParallelismInformation(org.apache.flink.runtime.scheduler.VertexParallelismInformation) TaskDeploymentDescriptor(org.apache.flink.runtime.deployment.TaskDeploymentDescriptor) TestingCheckpointIDCounter(org.apache.flink.runtime.checkpoint.TestingCheckpointIDCounter) TestingMetricRegistry(org.apache.flink.runtime.metrics.util.TestingMetricRegistry) Nonnull(javax.annotation.Nonnull) StandaloneCompletedCheckpointStore(org.apache.flink.runtime.checkpoint.StandaloneCompletedCheckpointStore) DefaultSchedulerTest(org.apache.flink.runtime.scheduler.DefaultSchedulerTest) JobGraphTestUtils.streamingJobGraph(org.apache.flink.runtime.jobgraph.JobGraphTestUtils.streamingJobGraph) ArchivedExecutionGraph(org.apache.flink.runtime.executiongraph.ArchivedExecutionGraph) ArchivedExecutionGraphBuilder(org.apache.flink.runtime.rest.handler.legacy.utils.ArchivedExecutionGraphBuilder) Logger(org.slf4j.Logger) Configuration(org.apache.flink.configuration.Configuration) CompletedCheckpointStore(org.apache.flink.runtime.checkpoint.CompletedCheckpointStore) JobStatusListener(org.apache.flink.runtime.executiongraph.JobStatusListener) DefaultDeclarativeSlotPoolTest.createSlotOffersForResourceRequirements(org.apache.flink.runtime.jobmaster.slotpool.DefaultDeclarativeSlotPoolTest.createSlotOffersForResourceRequirements) TimeUnit(java.util.concurrent.TimeUnit) Consumer(java.util.function.Consumer) ExecutionAttemptID(org.apache.flink.runtime.executiongraph.ExecutionAttemptID) ExecutionGraphTestUtils.createNoOpVertex(org.apache.flink.runtime.executiongraph.ExecutionGraphTestUtils.createNoOpVertex) SchedulerTestingUtils.enableCheckpointing(org.apache.flink.runtime.scheduler.SchedulerTestingUtils.enableCheckpointing) TaskExecutionState(org.apache.flink.runtime.taskmanager.TaskExecutionState) TemporaryFolder(org.junit.rules.TemporaryFolder) ExecutionAttemptID(org.apache.flink.runtime.executiongraph.ExecutionAttemptID) TaskExecutionStateTransition(org.apache.flink.runtime.executiongraph.TaskExecutionStateTransition) RootExceptionHistoryEntry(org.apache.flink.runtime.scheduler.exceptionhistory.RootExceptionHistoryEntry) ArrayList(java.util.ArrayList) List(java.util.List) TaskNotRunningException(org.apache.flink.runtime.operators.coordination.TaskNotRunningException) CheckpointException(org.apache.flink.runtime.checkpoint.CheckpointException) FlinkException(org.apache.flink.util.FlinkException) PartitionProducerDisposedException(org.apache.flink.runtime.jobmanager.PartitionProducerDisposedException) IOException(java.io.IOException) ExecutionException(java.util.concurrent.ExecutionException) SuppressRestartsException(org.apache.flink.runtime.execution.SuppressRestartsException) TaskExecutionState(org.apache.flink.runtime.taskmanager.TaskExecutionState) Test(org.junit.Test) ArchivedExecutionGraphTest(org.apache.flink.runtime.executiongraph.ArchivedExecutionGraphTest) DefaultSchedulerTest(org.apache.flink.runtime.scheduler.DefaultSchedulerTest)

Example 3 with RootExceptionHistoryEntry

use of org.apache.flink.runtime.scheduler.exceptionhistory.RootExceptionHistoryEntry in project flink by apache.

the class AdaptiveSchedulerTest method testExceptionHistoryWithTaskConcurrentGlobalFailure.

@Test
public void testExceptionHistoryWithTaskConcurrentGlobalFailure() throws Exception {
    final Exception expectedException1 = new Exception("Expected Global Exception 1");
    final Exception expectedException2 = new Exception("Expected Global Exception 2");
    BiConsumer<AdaptiveScheduler, List<ExecutionAttemptID>> testLogic = (scheduler, attemptIds) -> {
        scheduler.handleGlobalFailure(expectedException1);
        scheduler.handleGlobalFailure(expectedException2);
    };
    final Iterable<RootExceptionHistoryEntry> entries = runExceptionHistoryTests(testLogic);
    assertThat(entries).hasSize(1);
    final RootExceptionHistoryEntry failure = entries.iterator().next();
    assertThat(failure.getException().deserializeError(classLoader)).isEqualTo(expectedException1);
    final Iterable<ExceptionHistoryEntry> concurrentExceptions = failure.getConcurrentExceptions();
    final List<Throwable> foundExceptions = IterableUtils.toStream(concurrentExceptions).map(ExceptionHistoryEntry::getException).map(exception -> exception.deserializeError(classLoader)).collect(Collectors.toList());
    assertThat(foundExceptions).containsExactly(expectedException2);
}
Also used : Arrays(java.util.Arrays) ResourceRequirement(org.apache.flink.runtime.slots.ResourceRequirement) TaskNotRunningException(org.apache.flink.runtime.operators.coordination.TaskNotRunningException) ArchivedExecution(org.apache.flink.runtime.executiongraph.ArchivedExecution) TestingSlotAllocator(org.apache.flink.runtime.scheduler.adaptive.allocator.TestingSlotAllocator) CheckpointException(org.apache.flink.runtime.checkpoint.CheckpointException) ResultPartitionID(org.apache.flink.runtime.io.network.partition.ResultPartitionID) TestingFatalErrorHandler(org.apache.flink.runtime.util.TestingFatalErrorHandler) MetricRegistry(org.apache.flink.runtime.metrics.MetricRegistry) Duration(java.time.Duration) ClassRule(org.junit.ClassRule) TestingCheckpointRecoveryFactory(org.apache.flink.runtime.checkpoint.TestingCheckpointRecoveryFactory) ManuallyTriggeredComponentMainThreadExecutor(org.apache.flink.runtime.concurrent.ManuallyTriggeredComponentMainThreadExecutor) BlockingQueue(java.util.concurrent.BlockingQueue) MetricOptions(org.apache.flink.configuration.MetricOptions) JobManagerOptions(org.apache.flink.configuration.JobManagerOptions) Executors(java.util.concurrent.Executors) MetricNames(org.apache.flink.runtime.metrics.MetricNames) ArrayBlockingQueue(java.util.concurrent.ArrayBlockingQueue) VertexParallelismStore(org.apache.flink.runtime.scheduler.VertexParallelismStore) Time(org.apache.flink.api.common.time.Time) RootExceptionHistoryEntry(org.apache.flink.runtime.scheduler.exceptionhistory.RootExceptionHistoryEntry) FlinkException(org.apache.flink.util.FlinkException) ComponentMainThreadExecutor(org.apache.flink.runtime.concurrent.ComponentMainThreadExecutor) UpTimeGauge(org.apache.flink.runtime.executiongraph.metrics.UpTimeGauge) ExceptionHistoryEntry(org.apache.flink.runtime.scheduler.exceptionhistory.ExceptionHistoryEntry) LocalTaskManagerLocation(org.apache.flink.runtime.taskmanager.LocalTaskManagerLocation) ResourceCounter(org.apache.flink.runtime.util.ResourceCounter) JobStatus(org.apache.flink.api.common.JobStatus) DefaultAllocatedSlotPool(org.apache.flink.runtime.jobmaster.slotpool.DefaultAllocatedSlotPool) ArrayList(java.util.ArrayList) SchedulerNG(org.apache.flink.runtime.scheduler.SchedulerNG) DownTimeGauge(org.apache.flink.runtime.executiongraph.metrics.DownTimeGauge) PartitionProducerDisposedException(org.apache.flink.runtime.jobmanager.PartitionProducerDisposedException) Gauge(org.apache.flink.metrics.Gauge) ScheduledExecutorService(java.util.concurrent.ScheduledExecutorService) BiConsumer(java.util.function.BiConsumer) FixedDelayRestartBackoffTimeStrategy(org.apache.flink.runtime.executiongraph.failover.flip1.FixedDelayRestartBackoffTimeStrategy) Nullable(javax.annotation.Nullable) ArchivedExecutionVertex(org.apache.flink.runtime.executiongraph.ArchivedExecutionVertex) TestExecutorResource(org.apache.flink.testutils.executor.TestExecutorResource) ExecutionState(org.apache.flink.runtime.execution.ExecutionState) CheckpointsCleaner(org.apache.flink.runtime.checkpoint.CheckpointsCleaner) TestOperatorEvent(org.apache.flink.runtime.operators.coordination.TestOperatorEvent) Test(org.junit.Test) IOException(java.io.IOException) IterableUtils(org.apache.flink.util.IterableUtils) SimpleAckingTaskManagerGateway(org.apache.flink.runtime.executiongraph.utils.SimpleAckingTaskManagerGateway) ExecutionException(java.util.concurrent.ExecutionException) TaskExecutionStateTransition(org.apache.flink.runtime.executiongraph.TaskExecutionStateTransition) JobID(org.apache.flink.api.common.JobID) NoRestartBackoffTimeStrategy(org.apache.flink.runtime.executiongraph.failover.flip1.NoRestartBackoffTimeStrategy) SlotPoolTestUtils.offerSlots(org.apache.flink.runtime.jobmaster.slotpool.SlotPoolTestUtils.offerSlots) JobManagerMetricGroup(org.apache.flink.runtime.metrics.groups.JobManagerMetricGroup) ComponentMainThreadExecutorServiceAdapter(org.apache.flink.runtime.concurrent.ComponentMainThreadExecutorServiceAdapter) Assertions.assertThat(org.assertj.core.api.Assertions.assertThat) JobGraph(org.apache.flink.runtime.jobgraph.JobGraph) CheckpointCoordinatorConfiguration(org.apache.flink.runtime.jobgraph.tasks.CheckpointCoordinatorConfiguration) SuppressRestartsException(org.apache.flink.runtime.execution.SuppressRestartsException) ArchivedExecutionGraphTest(org.apache.flink.runtime.executiongraph.ArchivedExecutionGraphTest) AtomicInteger(java.util.concurrent.atomic.AtomicInteger) TestLogger(org.apache.flink.util.TestLogger) JobCheckpointingSettings(org.apache.flink.runtime.jobgraph.tasks.JobCheckpointingSettings) CheckpointIDCounter(org.apache.flink.runtime.checkpoint.CheckpointIDCounter) ArchivedExecutionJobVertex(org.apache.flink.runtime.executiongraph.ArchivedExecutionJobVertex) TestingCompletedCheckpointStore(org.apache.flink.runtime.checkpoint.TestingCompletedCheckpointStore) TestRestartBackoffTimeStrategy(org.apache.flink.runtime.executiongraph.failover.flip1.TestRestartBackoffTimeStrategy) IntermediateDataSetID(org.apache.flink.runtime.jobgraph.IntermediateDataSetID) Collectors(java.util.stream.Collectors) ResourceProfile(org.apache.flink.runtime.clusterframework.types.ResourceProfile) DefaultDeclarativeSlotPool(org.apache.flink.runtime.jobmaster.slotpool.DefaultDeclarativeSlotPool) List(java.util.List) StandaloneCheckpointIDCounter(org.apache.flink.runtime.checkpoint.StandaloneCheckpointIDCounter) CoordinationRequest(org.apache.flink.runtime.operators.coordination.CoordinationRequest) OperatorID(org.apache.flink.runtime.jobgraph.OperatorID) Optional(java.util.Optional) SchedulerExecutionMode(org.apache.flink.configuration.SchedulerExecutionMode) KeyGroupRangeAssignment(org.apache.flink.runtime.state.KeyGroupRangeAssignment) OneShotLatch(org.apache.flink.core.testutils.OneShotLatch) SchedulerBase(org.apache.flink.runtime.scheduler.SchedulerBase) SavepointFormatType(org.apache.flink.core.execution.SavepointFormatType) JobVertex(org.apache.flink.runtime.jobgraph.JobVertex) AtomicBoolean(java.util.concurrent.atomic.AtomicBoolean) CompletableFuture(java.util.concurrent.CompletableFuture) VertexParallelismInformation(org.apache.flink.runtime.scheduler.VertexParallelismInformation) TaskDeploymentDescriptor(org.apache.flink.runtime.deployment.TaskDeploymentDescriptor) TestingCheckpointIDCounter(org.apache.flink.runtime.checkpoint.TestingCheckpointIDCounter) TestingMetricRegistry(org.apache.flink.runtime.metrics.util.TestingMetricRegistry) Nonnull(javax.annotation.Nonnull) StandaloneCompletedCheckpointStore(org.apache.flink.runtime.checkpoint.StandaloneCompletedCheckpointStore) DefaultSchedulerTest(org.apache.flink.runtime.scheduler.DefaultSchedulerTest) JobGraphTestUtils.streamingJobGraph(org.apache.flink.runtime.jobgraph.JobGraphTestUtils.streamingJobGraph) ArchivedExecutionGraph(org.apache.flink.runtime.executiongraph.ArchivedExecutionGraph) ArchivedExecutionGraphBuilder(org.apache.flink.runtime.rest.handler.legacy.utils.ArchivedExecutionGraphBuilder) Logger(org.slf4j.Logger) Configuration(org.apache.flink.configuration.Configuration) CompletedCheckpointStore(org.apache.flink.runtime.checkpoint.CompletedCheckpointStore) JobStatusListener(org.apache.flink.runtime.executiongraph.JobStatusListener) DefaultDeclarativeSlotPoolTest.createSlotOffersForResourceRequirements(org.apache.flink.runtime.jobmaster.slotpool.DefaultDeclarativeSlotPoolTest.createSlotOffersForResourceRequirements) TimeUnit(java.util.concurrent.TimeUnit) Consumer(java.util.function.Consumer) ExecutionAttemptID(org.apache.flink.runtime.executiongraph.ExecutionAttemptID) ExecutionGraphTestUtils.createNoOpVertex(org.apache.flink.runtime.executiongraph.ExecutionGraphTestUtils.createNoOpVertex) SchedulerTestingUtils.enableCheckpointing(org.apache.flink.runtime.scheduler.SchedulerTestingUtils.enableCheckpointing) TaskExecutionState(org.apache.flink.runtime.taskmanager.TaskExecutionState) TemporaryFolder(org.junit.rules.TemporaryFolder) RootExceptionHistoryEntry(org.apache.flink.runtime.scheduler.exceptionhistory.RootExceptionHistoryEntry) ExceptionHistoryEntry(org.apache.flink.runtime.scheduler.exceptionhistory.ExceptionHistoryEntry) RootExceptionHistoryEntry(org.apache.flink.runtime.scheduler.exceptionhistory.RootExceptionHistoryEntry) ArrayList(java.util.ArrayList) List(java.util.List) TaskNotRunningException(org.apache.flink.runtime.operators.coordination.TaskNotRunningException) CheckpointException(org.apache.flink.runtime.checkpoint.CheckpointException) FlinkException(org.apache.flink.util.FlinkException) PartitionProducerDisposedException(org.apache.flink.runtime.jobmanager.PartitionProducerDisposedException) IOException(java.io.IOException) ExecutionException(java.util.concurrent.ExecutionException) SuppressRestartsException(org.apache.flink.runtime.execution.SuppressRestartsException) Test(org.junit.Test) ArchivedExecutionGraphTest(org.apache.flink.runtime.executiongraph.ArchivedExecutionGraphTest) DefaultSchedulerTest(org.apache.flink.runtime.scheduler.DefaultSchedulerTest)

Example 4 with RootExceptionHistoryEntry

use of org.apache.flink.runtime.scheduler.exceptionhistory.RootExceptionHistoryEntry in project flink by apache.

the class AdaptiveSchedulerTest method testExceptionHistoryWithTaskFailureFromStopWithSavepoint.

@Test
public void testExceptionHistoryWithTaskFailureFromStopWithSavepoint() throws Exception {
    final Exception expectedException = new Exception("Expected Local Exception");
    Consumer<JobGraph> setupJobGraph = jobGraph -> jobGraph.setSnapshotSettings(new JobCheckpointingSettings(CheckpointCoordinatorConfiguration.builder().build(), null));
    final CompletedCheckpointStore completedCheckpointStore = new StandaloneCompletedCheckpointStore(1);
    final CheckpointIDCounter checkpointIDCounter = new StandaloneCheckpointIDCounter();
    final CheckpointsCleaner checkpointCleaner = new CheckpointsCleaner();
    TestingCheckpointRecoveryFactory checkpointRecoveryFactory = new TestingCheckpointRecoveryFactory(completedCheckpointStore, checkpointIDCounter);
    Consumer<AdaptiveSchedulerBuilder> setupScheduler = builder -> builder.setCheckpointRecoveryFactory(checkpointRecoveryFactory).setCheckpointCleaner(checkpointCleaner);
    BiConsumer<AdaptiveScheduler, List<ExecutionAttemptID>> testLogic = (scheduler, attemptIds) -> {
        final ExecutionAttemptID attemptId = attemptIds.get(1);
        scheduler.stopWithSavepoint("file:///tmp/target", true, SavepointFormatType.CANONICAL);
        scheduler.updateTaskExecutionState(new TaskExecutionStateTransition(new TaskExecutionState(attemptId, ExecutionState.FAILED, expectedException)));
    };
    final Iterable<RootExceptionHistoryEntry> actualExceptionHistory = runExceptionHistoryTests(testLogic, setupScheduler, setupJobGraph);
    assertThat(actualExceptionHistory).hasSize(1);
    final RootExceptionHistoryEntry failure = actualExceptionHistory.iterator().next();
    assertThat(failure.getException().deserializeError(classLoader)).isEqualTo(expectedException);
}
Also used : Arrays(java.util.Arrays) ResourceRequirement(org.apache.flink.runtime.slots.ResourceRequirement) TaskNotRunningException(org.apache.flink.runtime.operators.coordination.TaskNotRunningException) ArchivedExecution(org.apache.flink.runtime.executiongraph.ArchivedExecution) TestingSlotAllocator(org.apache.flink.runtime.scheduler.adaptive.allocator.TestingSlotAllocator) CheckpointException(org.apache.flink.runtime.checkpoint.CheckpointException) ResultPartitionID(org.apache.flink.runtime.io.network.partition.ResultPartitionID) TestingFatalErrorHandler(org.apache.flink.runtime.util.TestingFatalErrorHandler) MetricRegistry(org.apache.flink.runtime.metrics.MetricRegistry) Duration(java.time.Duration) ClassRule(org.junit.ClassRule) TestingCheckpointRecoveryFactory(org.apache.flink.runtime.checkpoint.TestingCheckpointRecoveryFactory) ManuallyTriggeredComponentMainThreadExecutor(org.apache.flink.runtime.concurrent.ManuallyTriggeredComponentMainThreadExecutor) BlockingQueue(java.util.concurrent.BlockingQueue) MetricOptions(org.apache.flink.configuration.MetricOptions) JobManagerOptions(org.apache.flink.configuration.JobManagerOptions) Executors(java.util.concurrent.Executors) MetricNames(org.apache.flink.runtime.metrics.MetricNames) ArrayBlockingQueue(java.util.concurrent.ArrayBlockingQueue) VertexParallelismStore(org.apache.flink.runtime.scheduler.VertexParallelismStore) Time(org.apache.flink.api.common.time.Time) RootExceptionHistoryEntry(org.apache.flink.runtime.scheduler.exceptionhistory.RootExceptionHistoryEntry) FlinkException(org.apache.flink.util.FlinkException) ComponentMainThreadExecutor(org.apache.flink.runtime.concurrent.ComponentMainThreadExecutor) UpTimeGauge(org.apache.flink.runtime.executiongraph.metrics.UpTimeGauge) ExceptionHistoryEntry(org.apache.flink.runtime.scheduler.exceptionhistory.ExceptionHistoryEntry) LocalTaskManagerLocation(org.apache.flink.runtime.taskmanager.LocalTaskManagerLocation) ResourceCounter(org.apache.flink.runtime.util.ResourceCounter) JobStatus(org.apache.flink.api.common.JobStatus) DefaultAllocatedSlotPool(org.apache.flink.runtime.jobmaster.slotpool.DefaultAllocatedSlotPool) ArrayList(java.util.ArrayList) SchedulerNG(org.apache.flink.runtime.scheduler.SchedulerNG) DownTimeGauge(org.apache.flink.runtime.executiongraph.metrics.DownTimeGauge) PartitionProducerDisposedException(org.apache.flink.runtime.jobmanager.PartitionProducerDisposedException) Gauge(org.apache.flink.metrics.Gauge) ScheduledExecutorService(java.util.concurrent.ScheduledExecutorService) BiConsumer(java.util.function.BiConsumer) FixedDelayRestartBackoffTimeStrategy(org.apache.flink.runtime.executiongraph.failover.flip1.FixedDelayRestartBackoffTimeStrategy) Nullable(javax.annotation.Nullable) ArchivedExecutionVertex(org.apache.flink.runtime.executiongraph.ArchivedExecutionVertex) TestExecutorResource(org.apache.flink.testutils.executor.TestExecutorResource) ExecutionState(org.apache.flink.runtime.execution.ExecutionState) CheckpointsCleaner(org.apache.flink.runtime.checkpoint.CheckpointsCleaner) TestOperatorEvent(org.apache.flink.runtime.operators.coordination.TestOperatorEvent) Test(org.junit.Test) IOException(java.io.IOException) IterableUtils(org.apache.flink.util.IterableUtils) SimpleAckingTaskManagerGateway(org.apache.flink.runtime.executiongraph.utils.SimpleAckingTaskManagerGateway) ExecutionException(java.util.concurrent.ExecutionException) TaskExecutionStateTransition(org.apache.flink.runtime.executiongraph.TaskExecutionStateTransition) JobID(org.apache.flink.api.common.JobID) NoRestartBackoffTimeStrategy(org.apache.flink.runtime.executiongraph.failover.flip1.NoRestartBackoffTimeStrategy) SlotPoolTestUtils.offerSlots(org.apache.flink.runtime.jobmaster.slotpool.SlotPoolTestUtils.offerSlots) JobManagerMetricGroup(org.apache.flink.runtime.metrics.groups.JobManagerMetricGroup) ComponentMainThreadExecutorServiceAdapter(org.apache.flink.runtime.concurrent.ComponentMainThreadExecutorServiceAdapter) Assertions.assertThat(org.assertj.core.api.Assertions.assertThat) JobGraph(org.apache.flink.runtime.jobgraph.JobGraph) CheckpointCoordinatorConfiguration(org.apache.flink.runtime.jobgraph.tasks.CheckpointCoordinatorConfiguration) SuppressRestartsException(org.apache.flink.runtime.execution.SuppressRestartsException) ArchivedExecutionGraphTest(org.apache.flink.runtime.executiongraph.ArchivedExecutionGraphTest) AtomicInteger(java.util.concurrent.atomic.AtomicInteger) TestLogger(org.apache.flink.util.TestLogger) JobCheckpointingSettings(org.apache.flink.runtime.jobgraph.tasks.JobCheckpointingSettings) CheckpointIDCounter(org.apache.flink.runtime.checkpoint.CheckpointIDCounter) ArchivedExecutionJobVertex(org.apache.flink.runtime.executiongraph.ArchivedExecutionJobVertex) TestingCompletedCheckpointStore(org.apache.flink.runtime.checkpoint.TestingCompletedCheckpointStore) TestRestartBackoffTimeStrategy(org.apache.flink.runtime.executiongraph.failover.flip1.TestRestartBackoffTimeStrategy) IntermediateDataSetID(org.apache.flink.runtime.jobgraph.IntermediateDataSetID) Collectors(java.util.stream.Collectors) ResourceProfile(org.apache.flink.runtime.clusterframework.types.ResourceProfile) DefaultDeclarativeSlotPool(org.apache.flink.runtime.jobmaster.slotpool.DefaultDeclarativeSlotPool) List(java.util.List) StandaloneCheckpointIDCounter(org.apache.flink.runtime.checkpoint.StandaloneCheckpointIDCounter) CoordinationRequest(org.apache.flink.runtime.operators.coordination.CoordinationRequest) OperatorID(org.apache.flink.runtime.jobgraph.OperatorID) Optional(java.util.Optional) SchedulerExecutionMode(org.apache.flink.configuration.SchedulerExecutionMode) KeyGroupRangeAssignment(org.apache.flink.runtime.state.KeyGroupRangeAssignment) OneShotLatch(org.apache.flink.core.testutils.OneShotLatch) SchedulerBase(org.apache.flink.runtime.scheduler.SchedulerBase) SavepointFormatType(org.apache.flink.core.execution.SavepointFormatType) JobVertex(org.apache.flink.runtime.jobgraph.JobVertex) AtomicBoolean(java.util.concurrent.atomic.AtomicBoolean) CompletableFuture(java.util.concurrent.CompletableFuture) VertexParallelismInformation(org.apache.flink.runtime.scheduler.VertexParallelismInformation) TaskDeploymentDescriptor(org.apache.flink.runtime.deployment.TaskDeploymentDescriptor) TestingCheckpointIDCounter(org.apache.flink.runtime.checkpoint.TestingCheckpointIDCounter) TestingMetricRegistry(org.apache.flink.runtime.metrics.util.TestingMetricRegistry) Nonnull(javax.annotation.Nonnull) StandaloneCompletedCheckpointStore(org.apache.flink.runtime.checkpoint.StandaloneCompletedCheckpointStore) DefaultSchedulerTest(org.apache.flink.runtime.scheduler.DefaultSchedulerTest) JobGraphTestUtils.streamingJobGraph(org.apache.flink.runtime.jobgraph.JobGraphTestUtils.streamingJobGraph) ArchivedExecutionGraph(org.apache.flink.runtime.executiongraph.ArchivedExecutionGraph) ArchivedExecutionGraphBuilder(org.apache.flink.runtime.rest.handler.legacy.utils.ArchivedExecutionGraphBuilder) Logger(org.slf4j.Logger) Configuration(org.apache.flink.configuration.Configuration) CompletedCheckpointStore(org.apache.flink.runtime.checkpoint.CompletedCheckpointStore) JobStatusListener(org.apache.flink.runtime.executiongraph.JobStatusListener) DefaultDeclarativeSlotPoolTest.createSlotOffersForResourceRequirements(org.apache.flink.runtime.jobmaster.slotpool.DefaultDeclarativeSlotPoolTest.createSlotOffersForResourceRequirements) TimeUnit(java.util.concurrent.TimeUnit) Consumer(java.util.function.Consumer) ExecutionAttemptID(org.apache.flink.runtime.executiongraph.ExecutionAttemptID) ExecutionGraphTestUtils.createNoOpVertex(org.apache.flink.runtime.executiongraph.ExecutionGraphTestUtils.createNoOpVertex) SchedulerTestingUtils.enableCheckpointing(org.apache.flink.runtime.scheduler.SchedulerTestingUtils.enableCheckpointing) TaskExecutionState(org.apache.flink.runtime.taskmanager.TaskExecutionState) TemporaryFolder(org.junit.rules.TemporaryFolder) ExecutionAttemptID(org.apache.flink.runtime.executiongraph.ExecutionAttemptID) TaskExecutionStateTransition(org.apache.flink.runtime.executiongraph.TaskExecutionStateTransition) RootExceptionHistoryEntry(org.apache.flink.runtime.scheduler.exceptionhistory.RootExceptionHistoryEntry) JobCheckpointingSettings(org.apache.flink.runtime.jobgraph.tasks.JobCheckpointingSettings) TaskNotRunningException(org.apache.flink.runtime.operators.coordination.TaskNotRunningException) CheckpointException(org.apache.flink.runtime.checkpoint.CheckpointException) FlinkException(org.apache.flink.util.FlinkException) PartitionProducerDisposedException(org.apache.flink.runtime.jobmanager.PartitionProducerDisposedException) IOException(java.io.IOException) ExecutionException(java.util.concurrent.ExecutionException) SuppressRestartsException(org.apache.flink.runtime.execution.SuppressRestartsException) TestingCheckpointRecoveryFactory(org.apache.flink.runtime.checkpoint.TestingCheckpointRecoveryFactory) TaskExecutionState(org.apache.flink.runtime.taskmanager.TaskExecutionState) JobGraph(org.apache.flink.runtime.jobgraph.JobGraph) JobGraphTestUtils.streamingJobGraph(org.apache.flink.runtime.jobgraph.JobGraphTestUtils.streamingJobGraph) StandaloneCompletedCheckpointStore(org.apache.flink.runtime.checkpoint.StandaloneCompletedCheckpointStore) CheckpointsCleaner(org.apache.flink.runtime.checkpoint.CheckpointsCleaner) ArrayList(java.util.ArrayList) List(java.util.List) CheckpointIDCounter(org.apache.flink.runtime.checkpoint.CheckpointIDCounter) StandaloneCheckpointIDCounter(org.apache.flink.runtime.checkpoint.StandaloneCheckpointIDCounter) TestingCheckpointIDCounter(org.apache.flink.runtime.checkpoint.TestingCheckpointIDCounter) TestingCompletedCheckpointStore(org.apache.flink.runtime.checkpoint.TestingCompletedCheckpointStore) StandaloneCompletedCheckpointStore(org.apache.flink.runtime.checkpoint.StandaloneCompletedCheckpointStore) CompletedCheckpointStore(org.apache.flink.runtime.checkpoint.CompletedCheckpointStore) StandaloneCheckpointIDCounter(org.apache.flink.runtime.checkpoint.StandaloneCheckpointIDCounter) Test(org.junit.Test) ArchivedExecutionGraphTest(org.apache.flink.runtime.executiongraph.ArchivedExecutionGraphTest) DefaultSchedulerTest(org.apache.flink.runtime.scheduler.DefaultSchedulerTest)

Example 5 with RootExceptionHistoryEntry

use of org.apache.flink.runtime.scheduler.exceptionhistory.RootExceptionHistoryEntry in project flink by apache.

the class DefaultJobMasterServiceProcessTest method testInitializationFailureSetsExceptionHistoryProperly.

@Test
public void testInitializationFailureSetsExceptionHistoryProperly() throws ExecutionException, InterruptedException {
    final CompletableFuture<JobMasterService> jobMasterServiceFuture = new CompletableFuture<>();
    DefaultJobMasterServiceProcess serviceProcess = createTestInstance(jobMasterServiceFuture);
    final RuntimeException originalCause = new RuntimeException("Expected RuntimeException");
    long beforeFailureTimestamp = System.currentTimeMillis();
    jobMasterServiceFuture.completeExceptionally(originalCause);
    long afterFailureTimestamp = System.currentTimeMillis();
    final RootExceptionHistoryEntry entry = Iterables.getOnlyElement(serviceProcess.getResultFuture().get().getExecutionGraphInfo().getExceptionHistory());
    assertInitializationException(entry.getException(), originalCause, entry.getTimestamp(), beforeFailureTimestamp, afterFailureTimestamp);
    assertThat(entry.isGlobal()).isTrue();
}
Also used : CompletableFuture(java.util.concurrent.CompletableFuture) RootExceptionHistoryEntry(org.apache.flink.runtime.scheduler.exceptionhistory.RootExceptionHistoryEntry) Test(org.junit.jupiter.api.Test)

Aggregations

RootExceptionHistoryEntry (org.apache.flink.runtime.scheduler.exceptionhistory.RootExceptionHistoryEntry)19 Test (org.junit.Test)14 LocalTaskManagerLocation (org.apache.flink.runtime.taskmanager.LocalTaskManagerLocation)12 ArchivedExecutionVertex (org.apache.flink.runtime.executiongraph.ArchivedExecutionVertex)10 JobGraph (org.apache.flink.runtime.jobgraph.JobGraph)10 ArrayList (java.util.ArrayList)8 CompletableFuture (java.util.concurrent.CompletableFuture)8 JobID (org.apache.flink.api.common.JobID)8 IOException (java.io.IOException)7 Duration (java.time.Duration)7 Arrays (java.util.Arrays)7 List (java.util.List)7 Optional (java.util.Optional)7 ArrayBlockingQueue (java.util.concurrent.ArrayBlockingQueue)7 BlockingQueue (java.util.concurrent.BlockingQueue)7 ExecutionException (java.util.concurrent.ExecutionException)7 Executors (java.util.concurrent.Executors)7 ScheduledExecutorService (java.util.concurrent.ScheduledExecutorService)7 TimeUnit (java.util.concurrent.TimeUnit)7 AtomicBoolean (java.util.concurrent.atomic.AtomicBoolean)7