本期內(nèi)容:
我們提供的服務(wù)有:成都做網(wǎng)站、成都網(wǎng)站建設(shè)、微信公眾號(hào)開發(fā)、網(wǎng)站優(yōu)化、網(wǎng)站認(rèn)證、柴桑ssl等。為1000+企事業(yè)單位解決了網(wǎng)站和推廣的問(wèn)題。提供周到的售前咨詢和貼心的售后服務(wù),是有科學(xué)管理、有技術(shù)的柴桑網(wǎng)站制作公司
1、Receiver啟動(dòng)方式的設(shè)想
2、Receiver啟動(dòng)源碼徹底分析
一:Receiver啟動(dòng)方式的設(shè)想
1. Spark Streaming通過(guò)Receiver持續(xù)不斷的從外部數(shù)據(jù)源接收數(shù)據(jù),并把數(shù)據(jù)匯報(bào)給Driver端,由此每個(gè)Batch Durations就可以根據(jù)匯報(bào)的數(shù)據(jù)生成不同的Job。
2. Receiver是在Spark Streaming應(yīng)用程序啟動(dòng)時(shí)啟動(dòng)的,那么我們找Receiver在哪里啟動(dòng)就應(yīng)該去找Spark Streaming的啟動(dòng)。
3. Receivers和InputDStreams是一一對(duì)應(yīng)的,默認(rèn)情況下一般只有一個(gè)Receiver.
如何啟動(dòng)Receiver?
1. 從Spark Core的角度來(lái)看,Receiver的啟動(dòng)Spark Core并不知道,就相當(dāng)于Linux的內(nèi)核之上所有的都是應(yīng)用程序,因此Receiver是通過(guò)Job的方式啟動(dòng)的
2. 一般情況下,只有一個(gè)Receiver,但是可以創(chuàng)建不同的數(shù)據(jù)來(lái)源的InputDStream.
final private[streaming] class DStreamGraph extends Serializable with Logging { private val inputStreams = new ArrayBuffer[InputDStream[_]]() //數(shù)組 private val outputStreams = new ArrayBuffer[DStream[_]]()
3. 啟動(dòng)Receiver的時(shí)候,啟動(dòng)一個(gè)Job,這個(gè)Job里面有RDD的transformations操作和action的操作,這個(gè)Job只有一個(gè)partition.這個(gè)partition的特殊是里面只有一個(gè)成員, 這個(gè)成員就是啟動(dòng)的Receiver. 4. 這樣做的問(wèn)題: a) 如果有多個(gè)InputDStream,那就要啟動(dòng)多個(gè)Receiver,每個(gè)Receiver也就相當(dāng)于分片partition,那我們啟動(dòng)Receiver的時(shí)候理想的情況下是在不同的機(jī)器上啟動(dòng)Receiver, 但是Spark Core的角度來(lái)看就是應(yīng)用程序,感覺不到Receiver的特殊性,所以就會(huì)按照正常的Job啟動(dòng)的方式來(lái)處理,極有可能在一個(gè)Executor上啟動(dòng)多個(gè)Receiver. 這樣的話就可能導(dǎo)致負(fù)載不均衡。 b) 有可能啟動(dòng)Receiver失敗,只要集群存在Receiver就不應(yīng)該失敗。 c) 運(yùn)行過(guò)程中,就默認(rèn)的而言如果是一個(gè)partition的話,那啟動(dòng)的時(shí)候就是一個(gè)Task,但是此Task也很可能失敗,因此以Task啟動(dòng)的Receiver也會(huì)掛掉。
由此,可以得出,對(duì)于Receiver失敗的話,后果是非常嚴(yán)重的,那么Spark Streaming如何防止這些事的呢,下面就尋找Receiver的創(chuàng)建
這里先給出答案,后面源碼會(huì)詳細(xì)分析:
a) Spark使用一個(gè)Job啟動(dòng)一個(gè)Receiver.最大程度的保證了負(fù)載均衡。
b) Spark Streaming指定每個(gè)Receiver運(yùn)行在哪些Executor上。
c) 如果Receiver啟動(dòng)失敗,此時(shí)并不是Job失敗,在內(nèi)部會(huì)重新啟動(dòng)Receiver.
接下來(lái)我們通過(guò)代碼一步一步解析Receiver是如何啟動(dòng)的
1、首先我們?cè)诰帉懢唧w的應(yīng)用程序的時(shí)候,都會(huì)調(diào)用StreamingContext的start方法,其實(shí)這就是job啟動(dòng)的源頭,我們先來(lái)看下start方法的源碼:
def start(): Unit = synchronized { state match { case INITIALIZED => startSite.set(DStream.getCreationSite()) StreamingContext.ACTIVATION_LOCK.synchronized { StreamingContext.assertNoOtherContextIsActive() try { validate() // Start the streaming scheduler in a new thread, so that thread local properties // like call sites and job groups can be reset without affecting those of the // current thread. ThreadUtils.runInNewThread("streaming-start") { sparkContext.setCallSite(startSite.get) sparkContext.clearJobGroup() sparkContext.setLocalProperty(SparkContext.SPARK_JOB_INTERRUPT_ON_CANCEL, "false") scheduler.start() //啟動(dòng)JobScheduler的start方法,啟動(dòng)子線程,一方面為了本地初始化工作,另外一方面是不要阻塞主線程。 } state = StreamingContextState.ACTIVE } catch { case NonFatal(e) => logError("Error starting the context, marking it as stopped", e) scheduler.stop(false) state = StreamingContextState.STOPPED throw e } StreamingContext.setActiveContext(this) } shutdownHookRef = ShutdownHookManager.addShutdownHook( StreamingContext.SHUTDOWN_HOOK_PRIORITY)(stopOnShutdown) // Registering Streaming Metrics at the start of the StreamingContext assert(env.metricsSystem != null) env.metricsSystem.registerSource(streamingSource) uiTab.foreach(_.attach()) logInfo("StreamingContext started") case ACTIVE => logWarning("StreamingContext has already been started") case STOPPED => throw new IllegalStateException("StreamingContext has already been stopped") } }
2、上面調(diào)用start方法的時(shí)候,會(huì)調(diào)用JobScheduler的start()方法,在該方法里面,receiverTracker啟動(dòng)了,源碼如下:
def start(): Unit = synchronized { if (eventLoop != null) return // scheduler has already been started logDebug("Starting JobScheduler") eventLoop = new EventLoop[JobSchedulerEvent]("JobScheduler") { override protected def onReceive(event: JobSchedulerEvent): Unit = processEvent(event) override protected def onError(e: Throwable): Unit = reportError("Error in job scheduler", e) } eventLoop.start() // attach rate controllers of input streams to receive batch completion updates for { inputDStream <- ssc.graph.getInputStreams rateController <- inputDStream.rateController } ssc.addStreamingListener(rateController) listenerBus.start(ssc.sparkContext) receiverTracker = new ReceiverTracker(ssc) inputInfoTracker = new InputInfoTracker(ssc) receiverTracker.start() //啟動(dòng)Receiver jobGenerator.start() logInfo("Started JobScheduler") }
3、我們接著看下receiverTracker的start()方法,在start方法里啟動(dòng)了RPC消息通信體,為啥呢?因?yàn)閞eceiverTracker會(huì)監(jiān)控整個(gè)集群中的Receiver,Receiver轉(zhuǎn)過(guò)來(lái)要向ReceiverTrackerEndpoint匯報(bào)自己的狀態(tài),接收的數(shù)據(jù),包括生命周期等信息
/** Start the endpoint and receiver execution thread. */ def start(): Unit = synchronized { if (isTrackerStarted) { throw new SparkException("ReceiverTracker already started") } if (!receiverInputStreams.isEmpty) { //Receiver的啟動(dòng)是依據(jù)數(shù)據(jù)流的 endpoint = ssc.env.rpcEnv.setupEndpoint( "ReceiverTracker", new ReceiverTrackerEndpoint(ssc.env.rpcEnv)) //匯報(bào)狀態(tài)信息 if (!skipReceiverLaunch) launchReceivers() //發(fā)起Receiver logInfo("ReceiverTracker started") trackerState = Started } }
4、基于ReceiverInputDStream(是在Driver端)來(lái)獲得具體的Receivers實(shí)例,然后再把他們分不到Worker節(jié)點(diǎn)上。一個(gè)ReceiverInputDStream只產(chǎn)生一個(gè)Receiver
/** * Get the receivers from the ReceiverInputDStreams, distributes them to the * worker nodes as a parallel collection, and runs them. */ private def launchReceivers(): Unit = { val receivers = receiverInputStreams.map(nis => { //一個(gè)輸入數(shù)據(jù)來(lái)源只產(chǎn)生一個(gè)Receiver val rcvr = nis.getReceiver() rcvr.setReceiverId(nis.id) rcvr }) runDummySparkJob() //啟動(dòng)虛擬Job來(lái)分配Receiver到不同的executor上 logInfo("Starting " + receivers.length + " receivers") endpoint.send(StartAllReceivers(receivers)) }
5、其中runDummySparkJob()為了確保所有節(jié)點(diǎn)活著,而且避免所有的receivers集中在一個(gè)節(jié)點(diǎn)上。
private def runDummySparkJob(): Unit = { if (!ssc.sparkContext.isLocal) { ssc.sparkContext.makeRDD(1 to 50, 50).map(x => (x, 1)).reduceByKey(_ + _, 20).collect() } assert(getExecutors.nonEmpty) }
ReceiverInputDStream中的getReceiver()方法獲得receiver對(duì)象然后將它發(fā)送到worker節(jié)點(diǎn)上實(shí)例化receiver,然后去接收數(shù)據(jù)。
此方法必須要在子類中實(shí)現(xiàn)。
/** * Gets the receiver object that will be sent to the worker nodes * to receive data. This method needs to defined by any specific implementation * of a ReceiverInputDStream. */ def getReceiver(): Receiver[T]
ReceiverInputDStream是抽象類,所以getReceiver方法必須要在繼承的子類中實(shí)現(xiàn)
private[streaming] class SocketInputDStream[T: ClassTag]( ssc_ : StreamingContext, host: String, port: Int, bytesToObjects: InputStream => Iterator[T], storageLevel: StorageLevel ) extends ReceiverInputDStream[T](ssc_) { def getReceiver(): Receiver[T] = { new SocketReceiver(host, port, bytesToObjects, storageLevel) //調(diào)用SocketReceiver } } private[streaming] class SocketReceiver[T: ClassTag]( host: String, port: Int, bytesToObjects: InputStream => Iterator[T], storageLevel: StorageLevel ) extends Receiver[T](storageLevel) with Logging { def onStart() { // Start the thread that receives data over a connection new Thread("Socket Receiver") { setDaemon(true) override def run() { receive() } //啟動(dòng)線程,調(diào)用Receiver方法 }.start() }
在receive()方法中啟動(dòng)socket接收數(shù)據(jù)
/** Create a socket connection and receive data until receiver is stopped */ def receive() { var socket: Socket = null try { logInfo("Connecting to " + host + ":" + port) socket = new Socket(host, port) //根據(jù)我們應(yīng)用程序傳入的host和post創(chuàng)建socket對(duì)象 logInfo("Connected to " + host + ":" + port) val iterator = bytesToObjects(socket.getInputStream()) //接收數(shù)據(jù) while(!isStopped && iterator.hasNext) { store(iterator.next) //接收后的數(shù)據(jù)進(jìn)行存儲(chǔ) } if (!isStopped()) { restart("Socket data stream had no more data") } else { logInfo("Stopped receiving") } } catch { case e: java.net.ConnectException => restart("Error connecting to " + host + ":" + port, e) case NonFatal(e) => logWarning("Error receiving data", e) restart("Error receiving data", e) } finally { if (socket != null) { socket.close() logInfo("Closed socket to " + host + ":" + port) } } } }
6、ReceiverTrackerEndpoint源碼如下:
override def receive: PartialFunction[Any, Unit] = { // Local messages case StartAllReceivers(receivers) => val scheduledLocations = schedulingPolicy.scheduleReceivers(receivers, getExecutors) // receivers就是要啟動(dòng)的receiver,getExecutors獲得集群中的Executors的列表 for (receiver <- receivers) { val executors = scheduledLocations(receiver.streamId) updateReceiverScheduledExecutors(receiver.streamId, executors) receiverPreferredLocations(receiver.streamId) = receiver.preferredLocation startReceiver(receiver, executors) //循環(huán)receivers,每次將一個(gè)receiver傳入過(guò)去。 } case RestartReceiver(receiver) => // Old scheduled executors minus the ones that are not active any more val oldScheduledExecutors = getStoredScheduledExecutors(receiver.streamId) val scheduledLocations = if (oldScheduledExecutors.nonEmpty) { // Try global scheduling again oldScheduledExecutors } else { val oldReceiverInfo = receiverTrackingInfos(receiver.streamId) // Clear "scheduledLocations" to indicate we are going to do local scheduling val newReceiverInfo = oldReceiverInfo.copy( state = ReceiverState.INACTIVE, scheduledLocations = None) receiverTrackingInfos(receiver.streamId) = newReceiverInfo schedulingPolicy.rescheduleReceiver( receiver.streamId, receiver.preferredLocation, receiverTrackingInfos, getExecutors) } // Assume there is one receiver restarting at one time, so we don't need to update // receiverTrackingInfos startReceiver(receiver, scheduledLocations) case c: CleanupOldBlocks => receiverTrackingInfos.values.flatMap(_.endpoint).foreach(_.send(c)) case UpdateReceiverRateLimit(streamUID, newRate) => for (info <- receiverTrackingInfos.get(streamUID); eP <- info.endpoint) { eP.send(UpdateRateLimit(newRate)) } // Remote messages case ReportError(streamId, message, error) => reportError(streamId, message, error) }
從注釋中可以看到,Spark Streaming指定receiver在那些Executors運(yùn)行,而不是基于Spark Core中的Task來(lái)指定。 通過(guò)StartAllReceivers將消息發(fā)送給ReceiverTrackerEndpoint
在for循環(huán)中為每個(gè)receiver分配相應(yīng)的executor。并調(diào)用startReceiver方法:
Receiver是以job的方式啟動(dòng)的?。?! 這里你可能會(huì)有疑惑,沒有RDD和來(lái)的Job呢?首先,在startReceiver方法中,會(huì)將Receiver封裝成RDD
receiverRDD: RDD[Receiver[_]] = (scheduledLocations.isEmpty) { ssc..makeRDD((receiver)) } { preferredLocations = scheduledLocations.map(_.toString).distinct ssc..makeRDD((receiver -> preferredLocations)) }
封裝成RDD后,將RDD提交到集群中運(yùn)行
future = ssc.sparkContext.submitJob[Receiver[_]]( receiverRDDstartReceiverFunc()(__) => ())
task被發(fā)送到executor中,從RDD中取出“Receiver”然后對(duì)它執(zhí)行startReceiverFunc:
// Function to start the receiver on the worker node val startReceiverFunc: Iterator[Receiver[_]] => Unit = (iterator: Iterator[Receiver[_]]) => { if (!iterator.hasNext) { throw new SparkException( "Could not start receiver as object not found.") } if (TaskContext.get().attemptNumber() == 0) { val receiver = iterator.next() assert(iterator.hasNext == false) val supervisor = new ReceiverSupervisorImpl( //Receiver注冊(cè) receiver, SparkEnv.get, serializableHadoopConf.value, checkpointDirOption) supervisor.start() //啟動(dòng)Receiver supervisor.awaitTermination() } else { // It's restarted by TaskScheduler, but we want to reschedule it again. So exit it. } }
在函數(shù)中創(chuàng)建了一個(gè)ReceiverSupervisorImpl對(duì)象。它用來(lái)管理具體的Receiver。
首先它會(huì)將Receiver注冊(cè)到ReceiverTracker中
override protected def onReceiverStart(): Boolean = {
val msg = RegisterReceiver(
streamId, receiver.getClass.getSimpleName, host, executorId, endpoint)
trackerEndpoint.askWithRetry[Boolean](msg)
}
如果注冊(cè)成功,通過(guò)supervisor.start()來(lái)啟動(dòng)Receiver
/** Start the supervisor */ def start() { onStart() startReceiver() //啟動(dòng)Receiver }
// We will keep restarting the receiver job until ReceiverTracker is stopped future.onComplete { case Success(_) => if (!shouldStartReceiver) { onReceiverJobFinish(receiverId) } else { logInfo(s"Restarting Receiver $receiverId") self.send(RestartReceiver(receiver)) } case Failure(e) => if (!shouldStartReceiver) { onReceiverJobFinish(receiverId) } else { logError("Receiver has been stopped. Try to restart it.", e) logInfo(s"Restarting Receiver $receiverId") self.send(RestartReceiver(receiver)) } }(submitJobThreadPool) logInfo(s"Receiver ${receiver.streamId} started")
回到receiverTracker的startReceiver方法中,只要Receiver對(duì)應(yīng)的Job結(jié)束了(無(wú)論是正常還是異常結(jié)束),而ReceiverTracker還沒有停止。
它將會(huì)向ReceiverTrackerEndpoint發(fā)送一個(gè)ReStartReceiver的方法。
// We will keep restarting the receiver job until ReceiverTracker is stopped future.onComplete { case Success(_) => if (!shouldStartReceiver) { onReceiverJobFinish(receiverId) } else { logInfo(s"Restarting Receiver $receiverId") self.send(RestartReceiver(receiver)) } case Failure(e) => if (!shouldStartReceiver) { onReceiverJobFinish(receiverId) } else { logError("Receiver has been stopped. Try to restart it.", e) logInfo(s"Restarting Receiver $receiverId") self.send(RestartReceiver(receiver)) } }(submitJobThreadPool) logInfo(s"Receiver ${receiver.streamId} started")
重新為Receiver選擇一個(gè)executor,并再次運(yùn)行Receiver。直到ReceiverTracker啟動(dòng)為止。
在ReceiverTracker的receive方法中startReceiver方法第一個(gè)參數(shù)就是receiver,從實(shí)現(xiàn)的可以看出for循環(huán)不斷取出receiver,然后調(diào)用startReceiver。由此就可以得出一個(gè)Job只啟動(dòng)一個(gè)Receiver. 如果Receiver啟動(dòng)失敗,此時(shí)并不會(huì)認(rèn)為是作業(yè)失敗,會(huì)重新發(fā)消息給ReceiverTrackerEndpoint重新啟動(dòng)Receiver,這樣也就確保了Receivers一定會(huì)被啟動(dòng),這樣就不會(huì)像Task啟動(dòng)Receiver的話如果失敗受重試次數(shù)的影響。
簡(jiǎn)單的流程圖: