版權聲明:本套技術專欄是作者(秦凱新)平時工作的總結和升華,通過從真實商業環境抽取案例進行總結和分享,并給出商業應用的調優建議和集群環境容量規劃等內容,請持續關注本套博客。版權聲明:禁止轉載,歡迎學習。
Spark商業環境實戰及調優進階系列
1. Spark 內置框架rpc通訊機制
TransportContext 內部握有創建TransPortClient和TransPortServer的方法實現,但卻屬于最底層的RPC通訊設施。為什么呢?
因為成員變量RPCHandler是抽象的,并沒有具體的消息處理,而且TransportContext功能也在于創建TransPortClient客戶端和TransPortServer服務端。具體解釋如下:
Contains the context to create a {@link TransportServer}, {@link TransportClientFactory}, and to
setup Netty Channel pipelines with a
{@link org.apache.spark.network.server.TransportChannelHandler}.
所以TransportContext只能為最底層的通訊基礎。上層為NettyRPCEnv高層封裝,并持有TransportContext引用,在TransportContext中傳入NettyRpcHandler實體,來實現netty通訊回調Handler處理。TransportContext代碼片段如下:
/* The TransportServer and TransportClientFactory both create a TransportChannelHandler for each
* channel. As each TransportChannelHandler contains a TransportClient, this enables server
* processes to send messages back to the client on an existing channel.
*/
public class TransportContext {
private final Logger logger = LoggerFactory.getLogger(TransportContext.class);
private final TransportConf conf;
private final RpcHandler rpcHandler;
private final boolean closeIdleConnections;
private final MessageEncoder encoder;
private final MessageDecoder decoder;
public TransportContext(TransportConf conf, RpcHandler rpcHandler) {
this(conf, rpcHandler, false);
}
1.1 客戶端和服務端統一的消息接收處理器 TransportChannelHandlerer
TransportClient 和TransportServer 在配置Netty的pipeLine的handler處理器時,均采用TransportChannelHandler, 來做統一的消息receive處理。為什么呢?在于統一消息處理入口,TransportChannelHandlerer根據消息類型執行不同的處理,代碼片段如下:
public void channelRead(ChannelHandlerContext ctx, Object request) throws Exception {
if (request instanceof RequestMessage) {
requestHandler.handle((RequestMessage) request);
} else if (request instanceof ResponseMessage) {
responseHandler.handle((ResponseMessage) request);
} else {
ctx.fireChannelRead(request);
}
}
TransportContext初始化Pipeline的代碼片段:
public TransportChannelHandler initializePipeline(
SocketChannel channel,
RpcHandler channelRpcHandler) {
try {
TransportChannelHandler channelHandler = createChannelHandler(channel,
channelRpcHandler);
channel.pipeline()
.addLast("encoder", ENCODER)
.addLast(TransportFrameDecoder.HANDLER_NAME, NettyUtils.createFrameDecoder())
.addLast("decoder", DECODER)
.addLast("idleStateHandler", new IdleStateHandler(0, 0,
conf.connectionTimeoutMs() / 1000))
.addLast("handler", channelHandler);
return channelHandler;
} catch (RuntimeException e) {
logger.error("Error while initializing Netty pipeline", e);
throw e;
}
客戶端和服務端統一的消息接收處理器 TransportChannelHandlerer 是這個函數:createChannelHandler(channel, channelRpcHandler)實現的,也即統一了這個netty的消息接受處理,代碼片段如下:
/**
* Creates the server- and client-side handler which is used to handle both RequestMessages and
* ResponseMessages. The channel is expected to have been successfully created, though certain
* properties (such as the remoteAddress()) may not be available yet.
*/
private TransportChannelHandler createChannelHandler(Channel channel, RpcHandler rpcHandler) {
TransportResponseHandler responseHandler = new
TransportResponseHandler(channel);
TransportClient client = new TransportClient(channel, responseHandler);
TransportRequestHandler requestHandler = new TransportRequestHandler(channel, client,
rpcHandler, conf.maxChunksBeingTransferred());
return new TransportChannelHandler(client, responseHandler, requestHandler,
conf.connectionTimeoutMs(), closeIdleConnections);
}
不過transportClient對應的是TransportResponseHander,TransportServer對應的的是TransportRequestHander。
在進行消息處理時,首先會經過TransportChannelHandler根據消息類型進行處理器選擇,分別進行netty的消息生命周期管理:
- exceptionCaught
- channelActive
- channelInactive
- channelRead
- userEventTriggered
1.2 transportClient對應的是ResponseMessage
客戶端一旦發送消息(均為Request消息),就會在
private final Map<Long, RpcResponseCallback> outstandingRpcs;
private final Map<StreamChunkId, ChunkReceivedCallback> outstandingFetches
中緩存,用于回調處理。
1.3 transportServer對應的是RequestMessage
服務端接收消息類型(均為Request消息)
- ChunkFetchRequest
- RpcRequest
- OneWayMessage
- StremRequest
服務端響應類型(均為Response消息):
- ChunkFetchSucess
- ChunkFetchFailure
- RpcResponse
- RpcFailure
2. Spark RpcEnv基礎設施
2.1 上層建筑NettyRPCEnv
上層建筑NettyRPCEnv,持有TransportContext引用,在TransportContext中傳入NettyRpcHandler實體,來實現netty通訊回調Handler處理
- Dispatcher
- TransportContext
- TransPortClientFactroy
- TransportServer
- TransportConf
2.2 RpcEndPoint 與 RPCEndPointRef 端點
- RpcEndPoint 為服務端
- RPCEndPointRef 為客戶端
2.2 Dispacher 與 Inbox 與 Outbox
- 一個端點對應一個Dispacher,一個Inbox , 多個OutBox
- RpcEndpoint:RPC端點 ,Spark針對于每個節點(Client/Master/Worker)都稱之一個Rpc端點 ,且都實現RpcEndpoint接口,內部根據不同端點的需求,設計不同的消息和不同的業務處理,如果需要發送(詢問)則調用Dispatcher
- RpcEnv:RPC上下文環境,每個Rpc端點運行時依賴的上下文環境稱之為RpcEnv
- Dispatcher:消息分發器,針對于RPC端點需要發送消息或者從遠程RPC接收到的消息,分發至對應的指令收件箱/發件箱。如果指令接收方是自己存入收件箱,如果指令接收方為非自身端點,則放入發件箱
- Inbox:指令消息收件箱,一個本地端點對應一個收件箱,Dispatcher在每次向Inbox存入消息時,都將對應EndpointData加入內部待Receiver Queue中,另外Dispatcher創建時會啟動一個單獨線程進行輪詢Receiver Queue,進行收件箱消息消費
- OutBox:指令消息發件箱,一個遠程端點對應一個發件箱,當消息放入Outbox后,緊接著將消息通過TransportClient發送出去。消息放入發件箱以及發送過程是在同一個線程中進行,這樣做的主要原因是遠程消息分為RpcOutboxMessage, OneWayOutboxMessage兩種消息,而針對于需要應答的消息直接發送且需要得到結果進行處理
- TransportClient:Netty通信客戶端,根據OutBox消息的receiver信息,請求對應遠程TransportServer
- TransportServer:Netty通信服務端,一個RPC端點一個TransportServer,接受遠程消息后調用Dispatcher分發消息至對應收發件箱
Spark在Endpoint的設計上核心設計即為Inbox與Outbox,其中Inbox核心要點為:
- 內部的處理流程拆分為多個消息指令(InboxMessage)存放入Inbox
- 當Dispatcher啟動最后,會啟動一個名為【dispatcher-event-loop】的線程掃描Inbox待處理InboxMessage,并調用Endpoint根據InboxMessage類型做相應處理
- 當Dispatcher啟動最后,默認會向Inbox存入OnStart類型的InboxMessage,Endpoint在根據OnStart指令做相關的額外啟動工作,端點啟動后所有的工作都是對OnStart指令處理衍生出來的,因此可以說OnStart指令是相互通信的源頭。
-
注意: 一個端點對應一個Dispacher,一個Inbox , 多個OutBox,可以看到 inbox在Dispacher 中且在EndPointData內部:
private final RpcHandler rpcHandler; /** * A message dispatcher, responsible for routing RPC messages to the appropriate endpoint(s). */ private[netty] class Dispatcher(nettyEnv: NettyRpcEnv) extends Logging { private class EndpointData( val name: String, val endpoint: RpcEndpoint, val ref: NettyRpcEndpointRef) { val inbox = new Inbox(ref, endpoint) } private val endpoints = new ConcurrentHashMap[String, EndpointData] private val endpointRefs = new ConcurrentHashMap[RpcEndpoint, RpcEndpointRef] // Track the receivers whose inboxes may contain messages. private val receivers = new LinkedBlockingQueue[EndpointData]
-
注意: 一個端點對應一個Dispacher,一個Inbox , 多個OutBox,可以看到 OutBox在NettyRpcEnv內部:
private[netty] class NettyRpcEnv( val conf: SparkConf, javaSerializerInstance: JavaSerializerInstance, host: String, securityManager: SecurityManager) extends RpcEnv(conf) with Logging { private val dispatcher: Dispatcher = new Dispatcher(this) private val streamManager = new NettyStreamManager(this) private val transportContext = new TransportContext(transportConf, new NettyRpcHandler(dispatcher, this, streamManager)) /** * A map for [[RpcAddress]] and [[Outbox]]. When we are connecting to a remote [[RpcAddress]], * we just put messages to its [[Outbox]] to implement a non-blocking `send` method. */ private val outboxes = new ConcurrentHashMap[RpcAddress, Outbox]()
2.3 Dispacher 與 Inbox 與 Outbox
Dispatcher的代碼片段中,包含了核心的消息發送代碼邏輯,意思是:向服務端發送一條消息,也即同時放進Dispatcher中的receiverrs中,也放進inbox的messages中。這個高層封裝,如Master和Worker端點發送消息都是通過NettyRpcEnv中的 Dispatcher來實現的。在Dispatcher中有一個線程,叫做MessageLoop,實現消息的及時處理。
/**
* Posts a message to a specific endpoint.
*
* @param endpointName name of the endpoint.
* @param message the message to post
* @param callbackIfStopped callback function if the endpoint is stopped.
*/
private def postMessage(
endpointName: String,
message: InboxMessage,
callbackIfStopped: (Exception) => Unit): Unit = {
val error = synchronized {
val data = endpoints.get(endpointName)
if (stopped) {
Some(new RpcEnvStoppedException())
} else if (data == null) {
Some(new SparkException(s"Could not find $endpointName."))
} else {
data.inbox.post(message)
receivers.offer(data)
None
}
}
注意:默認第一條消息為onstart,為什么呢?看這里:
看到下面的 new EndpointData(name, endpoint, endpointRef) 了嗎?
def registerRpcEndpoint(name: String, endpoint: RpcEndpoint): NettyRpcEndpointRef = {
val addr = RpcEndpointAddress(nettyEnv.address, name)
val endpointRef = new NettyRpcEndpointRef(nettyEnv.conf, addr, nettyEnv)
synchronized {
if (stopped) {
throw new IllegalStateException("RpcEnv has been stopped")
}
if (endpoints.putIfAbsent(name, new EndpointData(name, endpoint, endpointRef)) != null) {
throw new IllegalArgumentException(s"There is already an RpcEndpoint called $name")
}
val data = endpoints.get(name)
endpointRefs.put(data.endpoint, data.ref)
receivers.offer(data) // for the OnStart message
}
endpointRef
}
注意EndpointData里面包含了inbox,因此Inbox初始化的時候,放進了onstart
private class EndpointData(
val name: String,
val endpoint: RpcEndpoint,
val ref: NettyRpcEndpointRef) {
val inbox = new Inbox(ref, endpoint)
}
onstart在Inbox初始化時出現了,注意每一個端點只有一個inbox,比如:master 節點。
2.4 發送消息流程為分為兩種,一種端點(Master)自己把消息發送到本地Inbox,一種端點(Master)接收到消息后,通過TransPortRequestHander接收后處理,扔進Inbox
2.4.1 端點(Master)自己把消息發送到本地Inbox
- endpoint(Master) -> NettyRpcEnv-> Dispatcher -> postMessage -> MessageLoop(Dispatcher) -> inbox -> process -> endpoint.receiveAndReply
解釋如下:端點通過自己的RPCEnv環境,向自己的Inbox中發送消息,然后交由Dispatch來進行消息的處理,調用了端點自己的receiveAndReply方法
-
這里著重講一下MessageLoop是什么時候啟動的,參照Dispatcher的代碼段如下,一旦初始化就會啟動,因為是成員變量:
private val threadpool: ThreadPoolExecutor = { val numThreads = nettyEnv.conf.getInt("spark.rpc.netty.dispatcher.numThreads", math.max(2, Runtime.getRuntime.availableProcessors())) val pool = ThreadUtils.newDaemonFixedThreadPool(numThreads, "dispatcher-event-loop") for (i <- 0 until numThreads) { pool.execute(new MessageLoop) } pool }
接著講nettyRpcEnv是何時初始化的,Dispatcher是何時初始化的?
master初始化RpcEnv環境時,調用NettyRpcEnvFactory().create(config)進行初始化nettyRpcEnv,然后其成員變量Dispatcher開始初始化,然后Dispatcher內部成員變量threadpool開始啟動messageLoop,然后開始處理消息,可謂是一環套一環啊。如下是Master端點初始化RPCEnv。
在NettyRpcEnv中,NettyRpcEnvFactory的create方法如下:
其中nettyRpcEnv.startServer,代碼段如下,然后調用底層 transportContext.createServer來創建Server,并初始化netty 的 pipeline:
server = transportContext.createServer(host, port, bootstraps)
dispatcher.registerRpcEndpoint(
RpcEndpointVerifier.NAME, new RpcEndpointVerifier(this, dispatcher))
最終端點開始不斷向自己的Inboxz中發送消息即可,代碼段如下:
private def postMessage(
endpointName: String,
message: InboxMessage,
callbackIfStopped: (Exception) => Unit): Unit = {
error = synchronized {
val data = endpoints.get(endpointName)
if (stopped) {
Some(new RpcEnvStoppedException())
} else if (data == null) {
Some(new SparkException(s"Could not find $endpointName."))
} else {
data.inbox.post(message)
receivers.offer(data)
None
}
}
2.4.2 端點(Master)接收到消息后,通過TransPortRequestHander接收后處理,扔進Inbox
- endpointRef(Worker) ->TransportChannelHandler -> channelRead0 -> TransPortRequestHander -> handle -> processRpcRequest ->NettyRpcHandler(在NettyRpcEnv中) -> receive -> internalReceive -> dispatcher.postToAll(RemoteProcessConnected(remoteEnvAddress)) (響應)-> dispatcher.postRemoteMessage(messageToDispatch, callback) (發送遠端來的消息放進inbox)-> postMessage -> inbox -> process
如下圖展示了整個消息接收到inbox的流程:
下圖展示了 TransportChannelHandler接收消息:
@Override
public void channelRead0(ChannelHandlerContext ctx, Message request) throws Exception {
if (request instanceof RequestMessage) {
requestHandler.handle((RequestMessage) request);
} else {
responseHandler.handle((ResponseMessage) request);
}
}
然后TransPortRequestHander來進行消息匹配處理:
最終交給inbox的process方法,實際上由端點 endpoint.receiveAndReply(context)方法處理:
/**
* Process stored messages.
*/
def process(dispatcher: Dispatcher): Unit = {
var message: InboxMessage = null
inbox.synchronized {
if (!enableConcurrent && numActiveThreads != 0) {
return
}
message = messages.poll()
if (message != null) {
numActiveThreads += 1
} else {
return
}
}
while (true) {
safelyCall(endpoint) {
message match {
case RpcMessage(_sender, content, context) =>
try {
endpoint.receiveAndReply(context).applyOrElse[Any, Unit](content, { msg =>
throw new SparkException(s"Unsupported message $message from ${_sender}")
})
} catch {
case NonFatal(e) =>
context.sendFailure(e)
// Throw the exception -- this exception will be caught by the safelyCall function.
// The endpoint's onError function will be called.
throw e
}
case OneWayMessage(_sender, content) =>
endpoint.receive.applyOrElse[Any, Unit](content, { msg =>
throw new SparkException(s"Unsupported message $message from ${_sender}")
})
case OnStart =>
endpoint.onStart()
if (!endpoint.isInstanceOf[ThreadSafeRpcEndpoint]) {
inbox.synchronized {
if (!stopped) {
enableConcurrent = true
}
}
}
case OnStop =>
val activeThreads = inbox.synchronized { inbox.numActiveThreads }
assert(activeThreads == 1,
s"There should be only a single active thread but found $activeThreads threads.")
dispatcher.removeRpcEndpointRef(endpoint)
endpoint.onStop()
assert(isEmpty, "OnStop should be the last message")
case RemoteProcessConnected(remoteAddress) =>
endpoint.onConnected(remoteAddress)
case RemoteProcessDisconnected(remoteAddress) =>
endpoint.onDisconnected(remoteAddress)
case RemoteProcessConnectionError(cause, remoteAddress) =>
endpoint.onNetworkError(cause, remoteAddress)
}
}
inbox.synchronized {
// "enableConcurrent" will be set to false after `onStop` is called, so we should check it
// every time.
if (!enableConcurrent && numActiveThreads != 1) {
// If we are not the only one worker, exit
numActiveThreads -= 1
return
}
message = messages.poll()
if (message == null) {
numActiveThreads -= 1
return
}
}
}
}
3 結語
本文花了將近兩天時間進行剖析Spark的 Rpc 工作原理,真是不容易,關鍵是你看懂了嗎?歡迎評論
秦凱新 于深圳 2018-10-28