一 Worker向Master注册
1.1 Worker启动,调用registerWithMaster,向Master注册
当worker启动的时候,会调用registerWithMaster方法
# 注册状态置为false
# 尝试向所有master注册
# 后台线程定时调度,发送ReregisterWithMaster请求,如果之前已经注册成功,则下一次来注册,则啥也不做
private def registerWithMaster() {
registrationRetryTimermatch {
//如果没有,说明还没有注册,然后会开始去注册
case None =>
// 初始注册状态为false
registered = false
// 尝试向所有master注册
registerMasterFutures= tryRegisterAllMasters()
// 连接尝试次数设为0
connectionAttemptCount= 0
// 后台线程定时调度,发送ReregisterWithMaster请求,如果之前已经注册成功,则下一次来注册,则啥也不做
registrationRetryTimer= Some(forwordMessageScheduler.scheduleAtFixedRate(
new Runnable {
override def run(): Unit = Utils.tryLogNonFatalError {
Option(self).foreach(_.send(ReregisterWithMaster))
}
},
INITIAL_REGISTRATION_RETRY_INTERVAL_SECONDS,
INITIAL_REGISTRATION_RETRY_INTERVAL_SECONDS,
TimeUnit.SECONDS))
// 如果已经有 registrationRetryTimer,就啥都不做
case Some(_) =>
logInfo("Not spawning another attempt to register with the master, sincethere is an" +
" attemptscheduled already.")
}
}
private def tryRegisterAllMasters(): Array[JFuture[_]] = {
masterRpcAddresses.map { masterAddress =>
registerMasterThreadPool.submit(new Runnable {
override def run(): Unit = {
try {
logInfo("Connecting to master " + masterAddress + "...")
// 构造master RpcEndpoint,用于向master发送消息或者请求
val masterEndpoint = rpcEnv.setupEndpointRef(masterAddress, Master.ENDPOINT_NAME)
// 向指定的master注册
registerWithMaster(masterEndpoint)
} catch {
case ie: InterruptedException => // Cancelled
case NonFatal(e) => logWarning(s"Failed to connect to master $masterAddress", e)
}
}
})
}
}
private def registerWithMaster(masterEndpoint: RpcEndpointRef): Unit = {
// 向master发送RegisterWorker请求
masterEndpoint.ask[RegisterWorkerResponse](RegisterWorker(
workerId, host, port, self, cores, memory, workerWebUiUrl))
.onComplete {
// 回调成功,则调用handleRegisterResponse
case Success(msg) =>
Utils.tryLogNonFatalError {
handleRegisterResponse(msg)
}
// 回调失败,则退出
case Failure(e) =>
logError(s"Cannot register with master: ${masterEndpoint.address}", e)
System.exit(1)
}(ThreadUtils.sameThread)
}
1.2 Master接受到Worker的RegisterWorker请求,则开始注册worker
# 检查worker是否已经注册过,如果已经注册过,返回注册失败的RegisterWorkerFailed消息
# 检查master所维护的worker节点中是否有DEAD状态的worker,如果有则移除这些worker
# 检查RpcAddress->Worker的映射是否包含这个RpcAddress,如果包含检查状态是否是为UNKNOWN状态,如果是则移除
# 把这个worker添加到Master所维护的与worker相关列表或者集合中
# 然后向Worker发送RegisteredWorker消息,表示注册已成功
# 重新调用schedule方法,开始进行调度,让worker开始干活
// 如果当前节点状态是standby,返回MasterInStandby
if (state == RecoveryState.STANDBY) {
context.reply(MasterInStandby)
} else if (idToWorker.contains(id)) {
// 判断维护的workerid->WorkerInfo映射是否包含这个worker id
// 如果包含返回wokerid,则返回 worker id重复的RegisterWorkerFailed
context.reply(RegisterWorkerFailed("Duplicate worker ID"))
} else {// 表示当前节点为master,且要注册是worker id之前是不存在的
// 创建worker,并进行注册,注册成功并且返回RegisteredWorker请求,然后开始调度
// 否则返回RegisterWorkerFailed请求,worker注册失败
val worker = new WorkerInfo(id, workerHost, workerPort, cores, memory,
workerRef, workerWebUiUrl)
if (registerWorker(worker)) {
persistenceEngine.addWorker(worker)
context.reply(RegisteredWorker(self, masterWebUiUrl))
schedule()
} else {
val workerAddress = worker.endpoint.address
logWarning("Worker registration failed. Attempted to re-register worker at same " +
"address: " + workerAddress)
context.reply(RegisterWorkerFailed("Attempted to re-register worker at same address: "
+ workerAddress))
}
}
1.3Worker收到Master返回的注册结果,调用handleRegisterResponse处理结果
# 如果接收RegisteredWorker消息,则更新注册状态;后台线程开始定时调度向master发送心跳的线程;向master发送WorkerLatestState请求,获取worker最近状态;
# 如果接收RegisterWorkerFailed消息,则退出
private def handleRegisterResponse(msg: RegisterWorkerResponse): Unit = synchronized {
msg match {
// 如果是RegisteredWorker请求,表示已经注册成功
case RegisteredWorker(masterRef, masterWebUiUrl) =>
logInfo("Successfully registered with master " + masterRef.address.toSparkURL)
registered = true // 更新registered状态
changeMaster(masterRef, masterWebUiUrl)
// 后台线程开始定时调度向master发送心跳的线程
forwordMessageScheduler.scheduleAtFixedRate(new Runnable {
override def run(): Unit = Utils.tryLogNonFatalError {
self.send(SendHeartbeat)
}
}, 0, HEARTBEAT_MILLIS, TimeUnit.MILLISECONDS)
// 如果启用了cleanup功能,后台线程开始定时调度发送WorkDirCleanup指令,清理目录
if (CLEANUP_ENABLED) {
logInfo(
s"Worker cleanup enabled; old application directories will be deleted in: $workDir")
forwordMessageScheduler.scheduleAtFixedRate(new Runnable {
override def run(): Unit = Utils.tryLogNonFatalError {
self.send(WorkDirCleanup)
}
}, CLEANUP_INTERVAL_MILLIS, CLEANUP_INTERVAL_MILLIS, TimeUnit.MILLISECONDS)
}
// 根据worker所持有的executor构造ExecutorDescription对象,描述该executor
val execs = executors.values.map { e =>
new ExecutorDescription(e.appId, e.execId, e.cores, e.state)
}
// 向master发送WorkerLatestState请求,获取worker最近状态
masterRef.send(WorkerLatestState(workerId, execs.toList, drivers.keys.toSeq))
// 如果是RegisterWorkerFailed请求,表示注册失败
case RegisterWorkerFailed(message) =>
// 如果还没有注册成功,则退出
if (!registered) {
logError("Worker registration failed: " + message)
System.exit(1)
}
// 如果是MasterInStandby请求,则啥也不做
case MasterInStandby =>
// Ignore. Master not yet ready.
}
}
二Driver向Master注册
在用spark-submit提交应用程序的时候,会调用SparkSubmit这个类,SparkSubmit会调用prepareSubmitEnvironment准备提交环境,在这个时候会设置集群管理者即Clsuter Manager;然后根据部署模式是standalone集群模式,且不是使用rest方式提交,则会初始化org.apache.spark.deploy.Client这个类,并且给定launch参数
2.1 客户端向Master发起提交driver的请求
Client在启动的时候会调用onstart方法,然后根据给定指令时launch还是kill发送对应的消息。
如果是launch:
则最终会调用ayncSendToMasterAndForwardReply向master发送RequestSubmitDriver消息
如果是kill:
则最终会调用ayncSendToMasterAndForwardReply向master发送RequestKillDriver消息
driverArgs.cmd match {
case "launch" =>
val mainClass = "org.apache.spark.deploy.worker.DriverWrapper"
val classPathConf = "spark.driver.extraClassPath"
val classPathEntries = sys.props.get(classPathConf).toSeq.flatMap { cp =>
cp.split(java.io.File.pathSeparator)
}
val libraryPathConf = "spark.driver.extraLibraryPath"
val libraryPathEntries = sys.props.get(libraryPathConf).toSeq.flatMap { cp =>
cp.split(java.io.File.pathSeparator)
}
val extraJavaOptsConf = "spark.driver.extraJavaOptions"
val extraJavaOpts = sys.props.get(extraJavaOptsConf)
.map(Utils.splitCommandString).getOrElse(Seq.empty)
val sparkJavaOpts = Utils.sparkJavaOpts(conf)
val javaOpts = sparkJavaOpts ++ extraJavaOpts
val command = new Command(mainClass,
Seq("{{WORKER_URL}}", "{{USER_JAR}}", driverArgs.mainClass) ++ driverArgs.driverOptions,
sys.env, classPathEntries, libraryPathEntries, javaOpts)
val driverDescription = new DriverDescription(
driverArgs.jarUrl,
driverArgs.memory,
driverArgs.cores,
driverArgs.supervise,
command)
ayncSendToMasterAndForwardReply[SubmitDriverResponse](
RequestSubmitDriver(driverDescription))
case "kill" =>
val driverId = driverArgs.driverId
ayncSendToMasterAndForwardReply[KillDriverResponse](RequestKillDriver(driverId))
}
}
2.2Master接收客户端的RequestSubmitDriver消息,开始注册driver
# 创建driver
# 持久化引擎添加driver
# 将driver添加到master所维护的driver相关集合或者列表中
# 调用schedule开始调度资源
# 向Client发送SubmitDriverResponse消息
case RequestSubmitDriver(description) =>
// 如果master不是active,返回错误
if (state != RecoveryState.ALIVE) {
val msg = s"${Utils.BACKUP_STANDALONE_MASTER_PREFIX}: $state. " +
"Can only accept driver submissions in ALIVE state."
context.reply(SubmitDriverResponse(self, false, None, msg))
} else {
logInfo("Driver submitted " + description.command.mainClass)
// 创建driver
val driver = createDriver(description)
// 持久化引擎添加drriver
persistenceEngine.addDriver(driver)
// drivers集合和waitingDrivers集合添加driver
waitingDrivers += driver
drivers.add(driver)
schedule()// 开始调度
// 返回成功的请求消息
context.reply(SubmitDriverResponse(self, true, Some(driver.id),
s"Driver successfully submitted as ${driver.id}"))
}
三Application向Master注册
3.1 构建StandaloneAppClient,然后向Master注册应用程序
在Standalone模式下,Driver是通过StandaloneSchedulerBackend来和Master进行资源请求协商的.
# SparkContext在初始化的时候会调用createTaskScheduler方法创建TaskSchedulerImpl和StandaloneSchedulerBackend
# 调用TaskSchedulerImpl的start方法启动TaskScheduler
// Create and start the scheduler
val (sched, ts) = SparkContext.createTaskScheduler(this, master, deployMode)
_schedulerBackend = sched
_taskScheduler = ts
_dagScheduler = new DAGScheduler(this)
_heartbeatReceiver.ask[Boolean](TaskSchedulerIsSet)
_taskScheduler.start()
# 启动TaskScheduler的时候,首先就会启动StandaloneSchedulerBackend
override def start() {
backend.start()
if (!isLocal && conf.getBoolean("spark.speculation", false)) {
logInfo("Starting speculative execution thread")
speculationScheduler.scheduleAtFixedRate(new Runnable {
override def run(): Unit = Utils.tryOrStopSparkContext(sc) {
checkSpeculatableTasks()
}
}, SPECULATION_INTERVAL_MS, SPECULATION_INTERVAL_MS, TimeUnit.MILLISECONDS)
}
}
# 启动StandaloneSchedulerBackend就会创建StandaloneAppClient,并且启动它
override def start() {
// ……省略
val appDesc = new ApplicationDescription(sc.appName, maxCores, sc.executorMemory, command,
appUIAddress, sc.eventLogDir, sc.eventLogCodec, coresPerExecutor, initialExecutorLimit)
client = new StandaloneAppClient(sc.env.rpcEnv, masters, appDesc, this, conf)
client.start()
// ……省略
}
# 启动StandaloneAppClient的时候,会构建通信环境, 会注册一个ClientEndpoint用于通信,然后调用ClientEndpoint的onstart方法
def start() {
// Just launch an rpcEndpoint; it will call back into the listener.
endpoint.set(rpcEnv.setupEndpoint("AppClient", new ClientEndpoint(rpcEnv)))
}
# onstart方法会调用registerWithMaster方法,然后调用tryRegisterAllMasters方法向所有master发送RegisterApplication消息,注册应用程序application
override def onStart(): Unit = {
try {
registerWithMaster(1)
} catch {
case e: Exception =>
logWarning("Failed to connect to master", e)
markDisconnected()
stop()
}
}
private def registerWithMaster(nthRetry: Int) {
registerMasterFutures.set(tryRegisterAllMasters())
registrationRetryTimer.set(registrationRetryThread.schedule(new Runnable {
override def run(): Unit = {
if (registered.get) {
registerMasterFutures.get.foreach(_.cancel(true))
registerMasterThreadPool.shutdownNow()
} else if (nthRetry >= REGISTRATION_RETRIES) {
markDead("All masters are unresponsive! Giving up.")
} else {
registerMasterFutures.get.foreach(_.cancel(true))
registerWithMaster(nthRetry + 1)
}
}
}, REGISTRATION_TIMEOUT_SECONDS, TimeUnit.SECONDS))
}
private def tryRegisterAllMasters(): Array[JFuture[_]] = {
for (masterAddress <- masterRpcAddresses) yield {
registerMasterThreadPool.submit(new Runnable {
override def run(): Unit = try {
if (registered.get) {
return
}
logInfo("Connecting to master " + masterAddress.toSparkURL + "...")
val masterRef = rpcEnv.setupEndpointRef(masterAddress, Master.ENDPOINT_NAME)
masterRef.send(RegisterApplication(appDescription, self))
} catch {
case ie: InterruptedException => // Cancelled
case NonFatal(e) => logWarning(s"Failed to connect to master $masterAddress", e)
}
})
}
}
3.2Master开始注册应用程序
# 创建应用程序
# 如果该应用程序已经注册过,则直接返回
# 把该应用程序注册到master,即添加到master所维护与application相关集合或者列表,放入等待队列
# 持久化引擎添加该应用程序
# 向master发送RegisteredApplication请求,表示注册已完成
# 调用schedule方法,开始调度
case RegisterApplication(description, driver) =>
// 其他的非leader的master是不能进行应用程序的创建和注册
if (state == RecoveryState.STANDBY) {
// ignore, don't send response
} else {
logInfo("Registering app " + description.name)
// 创建应用程序和driver
val app = createApplication(description, driver)
// 注册应用程序
registerApplication(app)
logInfo("Registered app " + description.name + " with ID " + app.id)
// 持久化引擎添加该application
persistenceEngine.addApplication(app)
// 向master发送RegisteredApplication请求,表示注册已完成
driver.send(RegisteredApplication(app.id, self))
schedule()
}
private def registerApplication(app: ApplicationInfo): Unit = {
// 获取app的RpcAddress
val appAddress = app.driver.address
// 如果已经注册过,则直接返回
if (addressToApp.contains(appAddress)) {
logInfo("Attempted to re-register application at same address: " + appAddress)
return
}
applicationMetricsSystem.registerSource(app.appSource)
apps += app // 添加这个app到master所维护的application集合
// 并且把app相关数据存放到对应application映射列表
idToApp(app.id) = app
endpointToApp(app.driver) = app
addressToApp(appAddress) = app
waitingApps += app
if (reverseProxy) {
webUi.addProxyTargets(app.id, app.desc.appUiUrl)
}
}
3.3StandaloneAppClient接收到RegisteredApplication消息
# 为应用程序设置id
# 注册状态设置为true
# 设置master
# StandaloneAppClientListener开始监听应用程序
case RegisteredApplication(appId_, masterRef) =>
appId.set(appId_)
registered.set(true)
master = Some(masterRef)
listener.connected(appId.get)