diff options
Diffstat (limited to 'core/src/main/scala/org/apache/spark/deploy/master/Master.scala')
-rw-r--r-- | core/src/main/scala/org/apache/spark/deploy/master/Master.scala | 52 |
1 files changed, 18 insertions, 34 deletions
diff --git a/core/src/main/scala/org/apache/spark/deploy/master/Master.scala b/core/src/main/scala/org/apache/spark/deploy/master/Master.scala index 01901bbf85..b443e8f051 100644 --- a/core/src/main/scala/org/apache/spark/deploy/master/Master.scala +++ b/core/src/main/scala/org/apache/spark/deploy/master/Master.scala @@ -217,7 +217,7 @@ private[deploy] class Master( } override def receive: PartialFunction[Any, Unit] = { - case ElectedLeader => { + case ElectedLeader => val (storedApps, storedDrivers, storedWorkers) = persistenceEngine.readPersistedData(rpcEnv) state = if (storedApps.isEmpty && storedDrivers.isEmpty && storedWorkers.isEmpty) { RecoveryState.ALIVE @@ -233,16 +233,14 @@ private[deploy] class Master( } }, WORKER_TIMEOUT_MS, TimeUnit.MILLISECONDS) } - } case CompleteRecovery => completeRecovery() - case RevokedLeadership => { + case RevokedLeadership => logError("Leadership has been revoked -- master shutting down.") System.exit(0) - } - case RegisterApplication(description, driver) => { + case RegisterApplication(description, driver) => // TODO Prevent repeated registrations from some driver if (state == RecoveryState.STANDBY) { // ignore, don't send response @@ -255,12 +253,11 @@ private[deploy] class Master( driver.send(RegisteredApplication(app.id, self)) schedule() } - } - case ExecutorStateChanged(appId, execId, state, message, exitStatus) => { + case ExecutorStateChanged(appId, execId, state, message, exitStatus) => val execOption = idToApp.get(appId).flatMap(app => app.executors.get(execId)) execOption match { - case Some(exec) => { + case Some(exec) => val appInfo = idToApp(appId) val oldState = exec.state exec.state = state @@ -298,22 +295,19 @@ private[deploy] class Master( } } } - } case None => logWarning(s"Got status update for unknown executor $appId/$execId") } - } - case DriverStateChanged(driverId, state, exception) => { + case DriverStateChanged(driverId, state, exception) => state match { case DriverState.ERROR | DriverState.FINISHED | DriverState.KILLED | DriverState.FAILED => removeDriver(driverId, state, exception) case _ => throw new Exception(s"Received unexpected state update for driver $driverId: $state") } - } - case Heartbeat(workerId, worker) => { + case Heartbeat(workerId, worker) => idToWorker.get(workerId) match { case Some(workerInfo) => workerInfo.lastHeartbeat = System.currentTimeMillis() @@ -327,9 +321,8 @@ private[deploy] class Master( " This worker was never registered, so ignoring the heartbeat.") } } - } - case MasterChangeAcknowledged(appId) => { + case MasterChangeAcknowledged(appId) => idToApp.get(appId) match { case Some(app) => logInfo("Application has been re-registered: " + appId) @@ -339,9 +332,8 @@ private[deploy] class Master( } if (canCompleteRecovery) { completeRecovery() } - } - case WorkerSchedulerStateResponse(workerId, executors, driverIds) => { + case WorkerSchedulerStateResponse(workerId, executors, driverIds) => idToWorker.get(workerId) match { case Some(worker) => logInfo("Worker has been re-registered: " + workerId) @@ -367,7 +359,6 @@ private[deploy] class Master( } if (canCompleteRecovery) { completeRecovery() } - } case WorkerLatestState(workerId, executors, driverIds) => idToWorker.get(workerId) match { @@ -397,9 +388,8 @@ private[deploy] class Master( logInfo(s"Received unregister request from application $applicationId") idToApp.get(applicationId).foreach(finishApplication) - case CheckForWorkerTimeOut => { + case CheckForWorkerTimeOut => timeOutDeadWorkers() - } case AttachCompletedRebuildUI(appId) => // An asyncRebuildSparkUI has completed, so need to attach to master webUi @@ -408,7 +398,7 @@ private[deploy] class Master( override def receiveAndReply(context: RpcCallContext): PartialFunction[Any, Unit] = { case RegisterWorker( - id, workerHost, workerPort, workerRef, cores, memory, workerWebUiUrl) => { + id, workerHost, workerPort, workerRef, cores, memory, workerWebUiUrl) => logInfo("Registering worker %s:%d with %d cores, %s RAM".format( workerHost, workerPort, cores, Utils.megabytesToString(memory))) if (state == RecoveryState.STANDBY) { @@ -430,9 +420,8 @@ private[deploy] class Master( + workerAddress)) } } - } - case RequestSubmitDriver(description) => { + case RequestSubmitDriver(description) => if (state != RecoveryState.ALIVE) { val msg = s"${Utils.BACKUP_STANDALONE_MASTER_PREFIX}: $state. " + "Can only accept driver submissions in ALIVE state." @@ -451,9 +440,8 @@ private[deploy] class Master( context.reply(SubmitDriverResponse(self, true, Some(driver.id), s"Driver successfully submitted as ${driver.id}")) } - } - case RequestKillDriver(driverId) => { + case RequestKillDriver(driverId) => if (state != RecoveryState.ALIVE) { val msg = s"${Utils.BACKUP_STANDALONE_MASTER_PREFIX}: $state. " + s"Can only kill drivers in ALIVE state." @@ -484,9 +472,8 @@ private[deploy] class Master( context.reply(KillDriverResponse(self, driverId, success = false, msg)) } } - } - case RequestDriverStatus(driverId) => { + case RequestDriverStatus(driverId) => if (state != RecoveryState.ALIVE) { val msg = s"${Utils.BACKUP_STANDALONE_MASTER_PREFIX}: $state. " + "Can only request driver status in ALIVE state." @@ -501,18 +488,15 @@ private[deploy] class Master( context.reply(DriverStatusResponse(found = false, None, None, None, None)) } } - } - case RequestMasterState => { + case RequestMasterState => context.reply(MasterStateResponse( address.host, address.port, restServerBoundPort, workers.toArray, apps.toArray, completedApps.toArray, drivers.toArray, completedDrivers.toArray, state)) - } - case BoundPortsRequest => { + case BoundPortsRequest => context.reply(BoundPortsResponse(address.port, webUi.boundPort, restServerBoundPort)) - } case RequestExecutors(appId, requestedTotal) => context.reply(handleRequestExecutors(appId, requestedTotal)) @@ -859,10 +843,10 @@ private[deploy] class Master( addressToApp -= app.driver.address if (completedApps.size >= RETAINED_APPLICATIONS) { val toRemove = math.max(RETAINED_APPLICATIONS / 10, 1) - completedApps.take(toRemove).foreach( a => { + completedApps.take(toRemove).foreach { a => Option(appIdToUI.remove(a.id)).foreach { ui => webUi.detachSparkUI(ui) } applicationMetricsSystem.removeSource(a.appSource) - }) + } completedApps.trimStart(toRemove) } completedApps += app // Remember it in our history |