Java Code Examples for org.apache.flink.runtime.executiongraph.Execution#getState()

public static ShuffleDescriptor getConsumedPartitionShuffleDescriptor(
		ExecutionEdge edge,
		boolean allowUnknownPartitions) {
	IntermediateResultPartition consumedPartition = edge.getSource();
	Execution producer = consumedPartition.getProducer().getCurrentExecutionAttempt();

	ExecutionState producerState = producer.getState();
	Optional<ResultPartitionDeploymentDescriptor> consumedPartitionDescriptor =
		producer.getResultPartitionDeploymentDescriptor(consumedPartition.getPartitionId());

	ResultPartitionID consumedPartitionId = new ResultPartitionID(
		consumedPartition.getPartitionId(),
		producer.getAttemptId());

	return getConsumedPartitionShuffleDescriptor(
		consumedPartitionId,
		consumedPartition.getResultType(),
		consumedPartition.isConsumable(),
		producerState,
		allowUnknownPartitions,
		consumedPartitionDescriptor.orElse(null));
}
 

public static ShuffleDescriptor getConsumedPartitionShuffleDescriptor(
		ExecutionEdge edge,
		boolean allowUnknownPartitions) {
	IntermediateResultPartition consumedPartition = edge.getSource();
	Execution producer = consumedPartition.getProducer().getCurrentExecutionAttempt();

	ExecutionState producerState = producer.getState();
	Optional<ResultPartitionDeploymentDescriptor> consumedPartitionDescriptor =
		producer.getResultPartitionDeploymentDescriptor(consumedPartition.getPartitionId());

	ResultPartitionID consumedPartitionId = new ResultPartitionID(
		consumedPartition.getPartitionId(),
		producer.getAttemptId());

	return getConsumedPartitionShuffleDescriptor(
		consumedPartitionId,
		consumedPartition.getResultType(),
		consumedPartition.isConsumable(),
		producerState,
		allowUnknownPartitions,
		consumedPartitionDescriptor.orElse(null));
}
 

@Override
public ExecutionState requestPartitionState(
		final IntermediateDataSetID intermediateResultId,
		final ResultPartitionID resultPartitionId) throws PartitionProducerDisposedException {

	mainThreadExecutor.assertRunningInMainThread();

	final Execution execution = executionGraph.getRegisteredExecutions().get(resultPartitionId.getProducerId());
	if (execution != null) {
		return execution.getState();
	}
	else {
		final IntermediateResult intermediateResult =
			executionGraph.getAllIntermediateResults().get(intermediateResultId);

		if (intermediateResult != null) {
			// Try to find the producing execution
			Execution producerExecution = intermediateResult
				.getPartitionById(resultPartitionId.getPartitionId())
				.getProducer()
				.getCurrentExecutionAttempt();

			if (producerExecution.getAttemptId().equals(resultPartitionId.getProducerId())) {
				return producerExecution.getState();
			} else {
				throw new PartitionProducerDisposedException(resultPartitionId);
			}
		} else {
			throw new IllegalArgumentException("Intermediate data set with ID "
				+ intermediateResultId + " not found.");
		}
	}
}
 

/**
 * Check if all tasks that we need to trigger are running. If not, abort the checkpoint.
 *
 * @return the executions need to be triggered.
 * @throws CheckpointException the exception fails checking
 */
private Execution[] getTriggerExecutions() throws CheckpointException {
	Execution[] executions = new Execution[tasksToTrigger.length];
	for (int i = 0; i < tasksToTrigger.length; i++) {
		Execution ee = tasksToTrigger[i].getCurrentExecutionAttempt();
		if (ee == null) {
			LOG.info(
				"Checkpoint triggering task {} of job {} is not being executed at the moment. Aborting checkpoint.",
				tasksToTrigger[i].getTaskNameWithSubtaskIndex(),
				job);
			throw new CheckpointException(
				CheckpointFailureReason.NOT_ALL_REQUIRED_TASKS_RUNNING);
		} else if (ee.getState() == ExecutionState.RUNNING) {
			executions[i] = ee;
		} else {
			LOG.info(
				"Checkpoint triggering task {} of job {} is not in state {} but {} instead. Aborting checkpoint.",
				tasksToTrigger[i].getTaskNameWithSubtaskIndex(),
				job,
				ExecutionState.RUNNING,
				ee.getState());
			throw new CheckpointException(
				CheckpointFailureReason.NOT_ALL_REQUIRED_TASKS_RUNNING);
		}
	}
	return executions;
}
 

@Override
public ExecutionState requestPartitionState(
	final IntermediateDataSetID intermediateResultId,
	final ResultPartitionID resultPartitionId) throws PartitionProducerDisposedException {

	mainThreadExecutor.assertRunningInMainThread();

	final Execution execution = executionGraph.getRegisteredExecutions().get(resultPartitionId.getProducerId());
	if (execution != null) {
		return execution.getState();
	}
	else {
		final IntermediateResult intermediateResult =
			executionGraph.getAllIntermediateResults().get(intermediateResultId);

		if (intermediateResult != null) {
			// Try to find the producing execution
			Execution producerExecution = intermediateResult
				.getPartitionById(resultPartitionId.getPartitionId())
				.getProducer()
				.getCurrentExecutionAttempt();

			if (producerExecution.getAttemptId().equals(resultPartitionId.getProducerId())) {
				return producerExecution.getState();
			} else {
				throw new PartitionProducerDisposedException(resultPartitionId);
			}
		} else {
			throw new IllegalArgumentException("Intermediate data set with ID "
				+ intermediateResultId + " not found.");
		}
	}
}
 

@Override
public void deliverOperatorEventToCoordinator(
		final ExecutionAttemptID taskExecutionId,
		final OperatorID operatorId,
		final OperatorEvent evt) throws FlinkException {

	// Failure semantics (as per the javadocs of the method):
	// If the task manager sends an event for a non-running task or an non-existing operator
	// coordinator, then respond with an exception to the call. If task and coordinator exist,
	// then we assume that the call from the TaskManager was valid, and any bubbling exception
	// needs to cause a job failure.

	final Execution exec = executionGraph.getRegisteredExecutions().get(taskExecutionId);
	if (exec == null || exec.getState() != ExecutionState.RUNNING) {
		// This situation is common when cancellation happens, or when the task failed while the
		// event was just being dispatched asynchronously on the TM side.
		// It should be fine in those expected situations to just ignore this event, but, to be
		// on the safe, we notify the TM that the event could not be delivered.
		throw new TaskNotRunningException("Task is not known or in state running on the JobManager.");
	}

	final OperatorCoordinatorHolder coordinator = coordinatorMap.get(operatorId);
	if (coordinator == null) {
		throw new FlinkException("No coordinator registered for operator " + operatorId);
	}

	try {
		coordinator.handleEventFromOperator(exec.getParallelSubtaskIndex(), evt);
	} catch (Throwable t) {
		ExceptionUtils.rethrowIfFatalErrorOrOOM(t);
		handleGlobalFailure(t);
	}
}
 

/**
 * Triggers a task back pressure stats request to all tasks.
 *
 * @param tasks Tasks to request.
 * @return A future of the completed task back pressure stats.
 */
CompletableFuture<BackPressureStats> triggerBackPressureRequest(ExecutionVertex[] tasks) {
	checkNotNull(tasks, "Tasks to request must not be null.");
	checkArgument(tasks.length >= 1, "No tasks to request.");

	// Execution IDs of running tasks
	ExecutionAttemptID[] triggerIds = new ExecutionAttemptID[tasks.length];
	Execution[] executions = new Execution[tasks.length];

	// Check that all tasks are RUNNING before triggering anything. The
	// triggering can still fail.
	for (int i = 0; i < triggerIds.length; i++) {
		Execution execution = tasks[i].getCurrentExecutionAttempt();
		if (execution != null && execution.getState() == ExecutionState.RUNNING) {
			executions[i] = execution;
			triggerIds[i] = execution.getAttemptId();
		} else {
			return FutureUtils.completedExceptionally(new IllegalStateException("Task " + tasks[i]
				.getTaskNameWithSubtaskIndex() + " is not running."));
		}
	}

	synchronized (lock) {
		if (isShutDown) {
			return FutureUtils.completedExceptionally(new IllegalStateException("Shut down."));
		}

		int requestId = requestIdCounter++;

		LOG.debug("Triggering task back pressure request {}.", requestId);

		PendingBackPressureRequest pending = new PendingBackPressureRequest(requestId, triggerIds);

		// Add the pending request before scheduling the discard task to
		// prevent races with removing it again.
		pendingRequests.put(requestId, pending);

		requestBackPressure(executions, requestId);

		return pending.getBackPressureStatsFuture();
	}
}