public abstract class ContaineredJobManager extends JobManager
JobManager
with additional messages
to start/administer/stop the session.
param: flinkConfiguration Configuration object for the actor
param: futureExecutor Execution context which is used to execute concurrent tasks in the
ExecutionGraph
param: ioExecutor to execute blocking io operations
param: instanceManager Instance manager to manage the registered
TaskManager
param: scheduler Scheduler to schedule Flink jobs
param: libraryCacheManager Manager to manage uploaded jar files
param: archive Archive for finished Flink jobs
param: restartStrategyFactory Restart strategy to be used in case of a job recovery
param: timeout Timeout for futures
param: leaderElectionService LeaderElectionService to participate in the leader election
Constructor and Description |
---|
ContaineredJobManager(Configuration flinkConfiguration,
ScheduledExecutorService futureExecutor,
Executor ioExecutor,
InstanceManager instanceManager,
Scheduler scheduler,
BlobLibraryCacheManager libraryCacheManager,
akka.actor.ActorRef archive,
RestartStrategyFactory restartStrategyFactory,
scala.concurrent.duration.FiniteDuration timeout,
LeaderElectionService leaderElectionService,
SubmittedJobGraphStore submittedJobGraphs,
CheckpointRecoveryFactory checkpointRecoveryFactory,
scala.concurrent.duration.FiniteDuration jobRecoveryTimeout,
scala.Option<MetricRegistry> metricsRegistry) |
Modifier and Type | Method and Description |
---|---|
scala.PartialFunction<Object,scala.runtime.BoxedUnit> |
handleContainerMessage() |
scala.PartialFunction<Object,scala.runtime.BoxedUnit> |
handleMessage()
Central work method of the JobManager actor.
|
abstract scala.concurrent.duration.FiniteDuration |
jobPollingInterval() |
JobID |
stopWhenJobFinished() |
archive, aroundPostRestart, aroundPostStop, aroundPreRestart, aroundPreStart, aroundReceive, checkpointRecoveryFactory, context, createJobManagerComponents, currentJobs, currentResourceManager, currentResourceManagerConnectionId, decorateMessage, defaultSavepointDir, flinkConfiguration, futureExecutor, futuresToComplete, getAddress, getArchiveProps, getJobManagerProps, grantLeadership, haMode, handleError, instanceManager, ioExecutor, jobManagerMetricGroup, jobRecoveryTimeout, leaderElectionService, leaderSessionID, libraryCacheManager, log, LOG, main, metricsRegistry, onAddedJobGraph, onRemovedJobGraph, parseArgs, postRestart, postStop, preRestart, preStart, receive, restartStrategyFactory, retryOnBindException, revokeLeadership, runJobManager, runJobManager, RUNTIME_FAILURE_RETURN_CODE, scheduler, self, sender, shutdown, startActorSystem, startJobManagerActors, startJobManagerActors, startJobManagerActors, STARTUP_FAILURE_RETURN_CODE, submittedJobGraphs, supervisorStrategy, taskManagerMap, timeout, triggerResourceManagerReconnectInterval, unhandled, webMonitorPort
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
decorateMessage, handleDiscardedMessage, handleNoLeaderId, receive
receive
public ContaineredJobManager(Configuration flinkConfiguration, ScheduledExecutorService futureExecutor, Executor ioExecutor, InstanceManager instanceManager, Scheduler scheduler, BlobLibraryCacheManager libraryCacheManager, akka.actor.ActorRef archive, RestartStrategyFactory restartStrategyFactory, scala.concurrent.duration.FiniteDuration timeout, LeaderElectionService leaderElectionService, SubmittedJobGraphStore submittedJobGraphs, CheckpointRecoveryFactory checkpointRecoveryFactory, scala.concurrent.duration.FiniteDuration jobRecoveryTimeout, scala.Option<MetricRegistry> metricsRegistry)
public abstract scala.concurrent.duration.FiniteDuration jobPollingInterval()
public JobID stopWhenJobFinished()
public scala.PartialFunction<Object,scala.runtime.BoxedUnit> handleMessage()
JobManager
handleMessage
in interface FlinkActor
handleMessage
in class JobManager
public scala.PartialFunction<Object,scala.runtime.BoxedUnit> handleContainerMessage()
Copyright © 2014–2018 The Apache Software Foundation. All rights reserved.