介紹
作為常用的http協(xié)議服務(wù)器，Tomcat應(yīng)用非常廣泛。tomcat也是遵循Servelt協(xié)議的，Servelt協(xié)議可以讓服務(wù)器與真實(shí)服務(wù)邏輯代碼進(jìn)行解耦。各自只需要關(guān)注Servlet協(xié)議即可。
對(duì)于tomcat是如何作為一個(gè)高性能的服務(wù)器的呢？你是不是也會(huì)有這樣的疑問(wèn)？
tomcat是如何接收網(wǎng)絡(luò)請(qǐng)求？
如何做到高性能的http協(xié)議服務(wù)器？
tomcat從8.0往后開(kāi)始使用了NIO非阻塞io模型，提高了吞吐量，本文的源碼是tomcat 9.0.48版本
接收Socket請(qǐng)求
org.Apache.tomcat.util.net.Acceptor實(shí)現(xiàn)了Runnable接口，在一個(gè)單獨(dú)的線程中以死循環(huán)的方式一直進(jìn)行socket的監(jiān)聽(tīng)
線程的初始化及啟動(dòng)是在方法org.apache.tomcat.util.net.AbstractEndpoint#startAcceptorThread
有個(gè)很重要的屬性org.apache.tomcat.util.net.AbstractEndpoint；同時(shí)實(shí)現(xiàn)了run方法，方法中主要有以下功能：
- 請(qǐng)求最大連接數(shù)限制： 最大為 8*1024；請(qǐng)你注意到達(dá)最大連接數(shù)后操作系統(tǒng)底層還是會(huì)接收客戶端連接，但用戶層已經(jīng)不再接收
- 獲取socketChannel
public void run() {
int errorDelay = 0;
try {
// Loop until we receive a shutdown command
while (!stopCalled) {
...
if (stopCalled) {
break;
}
state = AcceptorState.RUNNING;
try {
//if we have reached max connections, wait
// 如果連接超過(guò)了 8*1024，則線程阻塞等待; 是使用org.apache.tomcat.util.threads.LimitLatch類實(shí)現(xiàn)了分享鎖(內(nèi)部實(shí)現(xiàn)了AbstractQueuedSynchronizer)
// 請(qǐng)你注意到達(dá)最大連接數(shù)后操作系統(tǒng)底層還是會(huì)接收客戶端連接，但用戶層已經(jīng)不再接收。
endpoint.countUpOrAwaitConnection();
// Endpoint might have been paused while waiting for latch
// If that is the case, don't accept new connections
if (endpoint.isPaused()) {
continue;
}
U socket = null;
try {
// Accept the next incoming connection from the server
// socket
// 抽象方法，不同的endPoint有不同的實(shí)現(xiàn)方法。NioEndPoint為例，實(shí)現(xiàn)方法為serverSock.accept()，這個(gè)方法主要看serverSock實(shí)例化時(shí)如果為阻塞，accept方法為阻塞；反之為立即返回，如果沒(méi)有socket鏈接，則為null
socket = endpoint.serverSocketAccept();
} catch (Exception ioe) {
// We didn't get a socket
endpoint.countDownConnection();
if (endpoint.isRunning()) {
// Introduce delay if necessary
errorDelay = handleExceptionWithDelay(errorDelay);
// re-throw
throw ioe;
} else {
break;
}
}
// Successful accept, reset the error delay
errorDelay = 0;
// Configure the socket
if (!stopCalled && !endpoint.isPaused()) {
// setSocketOptions() will hand the socket off to
// an Appropriate processor if successful
// endPoint類的抽象方法，不同的endPoint有不同的實(shí)現(xiàn)。處理獲取到的socketChannel鏈接，如果該socket鏈接能正常處理，那么該方法會(huì)返回true，否則為false
if (!endpoint.setSocketOptions(socket)) {
endpoint.closeSocket(socket);
}
} else {
endpoint.destroySocket(socket);
}
} catch (Throwable t) {
...
}
}
} finally {
stopLatch.countDown();
}
state = AcceptorState.ENDED;
}
再來(lái)看下org.apache.tomcat.util.net.NioEndpoint#setSocketOptions方法的具體實(shí)現(xiàn)(NioEndpoint為例)
這個(gè)方法中主要做的事：
- 創(chuàng)建NioChannel
- 設(shè)置socket為非阻塞
- 將socket添加到Poller的隊(duì)列中
protected boolean setSocketOptions(SocketChannel socket) {
NIOSocketWrapper socketWrapper = null;
try {
// Allocate channel and wrapper
// 優(yōu)先使用已有的緩存nioChannel
NioChannel channel = null;
if (nioChannels != null) {
channel = nioChannels.pop();
}
if (channel == null) {
SocketBufferHandler bufhandler = new SocketBufferHandler(
socketProperties.getAppReadBufSize(),
socketProperties.getAppWriteBufSize(),
socketProperties.getDirectBuffer());
if (isSSLEnabled()) {
channel = new SecureNioChannel(bufhandler, this);
} else {
channel = new NioChannel(bufhandler);
}
}
// 將nioEndpoint與NioChannel進(jìn)行包裝
NioSocketWrapper newWrapper = new NioSocketWrapper(channel, this);
channel.reset(socket, newWrapper);
connections.put(socket, newWrapper);
socketWrapper = newWrapper;
// Set socket properties
// Disable blocking, polling will be used
// 設(shè)置當(dāng)前鏈接的socket為非阻塞
socket.configureBlocking(false);
if (getUnixDomainSocketPath() == null) {
socketProperties.setProperties(socket.socket());
}
socketWrapper.setReadTimeout(getConnectionTimeout());
socketWrapper.setWriteTimeout(getConnectionTimeout());
socketWrapper.setKeepAliveLeft(NioEndpoint.this.getMaxKeepAliveRequests());
// 將包裝后的nioChannel與nioEndpoint進(jìn)行注冊(cè)，注冊(cè)到Poller，將對(duì)應(yīng)的socket包裝類添加到Poller的隊(duì)列中，同時(shí)喚醒selector
poller.register(socketWrapper);
return true;
} catch (Throwable t) {
ExceptionUtils.handleThrowable(t);
try {
log.error(sm.getString("endpoint.socketOptionsError"), t);
} catch (Throwable tt) {
ExceptionUtils.handleThrowable(tt);
}
if (socketWrapper == null) {
destroySocket(socket);
}
}
// Tell to close the socket if needed
return false;
}
Socket請(qǐng)求輪詢
上一小節(jié)是接收到了socket請(qǐng)求，進(jìn)行包裝之后，將socket添加到了Poller的隊(duì)列上，并可能喚醒了Selector，本小節(jié)就來(lái)看看，Poller是如何進(jìn)行socket的輪詢的。
首先org.apache.tomcat.util.net.NioEndpoint.Poller也是實(shí)現(xiàn)了Runnable接口，是一個(gè)可以單獨(dú)啟動(dòng)的線程
初始化及啟動(dòng)是在org.apache.tomcat.util.net.NioEndpoint#startInternal
重要的屬性：
- JAVA.nio.channels.Selector：在Poller對(duì)象初始化的時(shí)候，就會(huì)啟動(dòng)輪詢器
- SynchronizedQueue<PollerEvent>：同步的事件隊(duì)列
再來(lái)看下具體處理邏輯，run方法的源碼
public void run() {
// Loop until destroy() is called
while (true) {
boolean hasEvents = false;
try {
if (!close) {
// 去SynchronizedQueue事件隊(duì)列中拉去，看是否已經(jīng)有了事件，如果有，則返回true
// 如果從隊(duì)列中拉取到了event(即上一步將NioSocketWrapper封裝為PollerEvent添加到次隊(duì)列中)，將socketChannel注冊(cè)到Selector上，標(biāo)記為SelectionKey.OP_READ，添加處理函數(shù)attachment(為Accetpor添加到Poller時(shí)的
// NioSocketWrapper)
hasEvents = events();
if (wakeupCounter.getAndSet(-1) > 0) {
// If we are here, means we have other stuff to do
// Do a non blocking select
keyCount = selector.selectNow();
} else {
keyCount = selector.select(selectorTimeout);
}
wakeupCounter.set(0);
}
if (close) {
events();
timeout(0, false);
try {
selector.close();
} catch (IOException ioe) {
log.error(sm.getString("endpoint.nio.selectorCloseFail"), ioe);
}
break;
}
// Either we timed out or we woke up, process events first
if (keyCount == 0) {
hasEvents = (hasEvents | events());
}
} catch (Throwable x) {
ExceptionUtils.handleThrowable(x);
log.error(sm.getString("endpoint.nio.selectorLoopError"), x);
continue;
}
Iterator<SelectionKey> iterator =
keyCount > 0 ? selector.selectedKeys().iterator() : null;
// Walk through the collection of ready keys and dispatch
// any active event.
// selector輪詢獲取已經(jīng)注冊(cè)的事件，如果有事件準(zhǔn)備好，此時(shí)通過(guò)selectKeys方法就能拿到對(duì)應(yīng)的事件
while (iterator != null && iterator.hasNext()) {
SelectionKey sk = iterator.next();
// 獲取到事件后，從迭代器刪除事件，防止事件重復(fù)輪詢
iterator.remove();
// 獲取事件的處理器，這個(gè)attachment是在event()方法中注冊(cè)的，后續(xù)這個(gè)事件的處理，就交給這個(gè)wrapper去處理
NioSocketWrapper socketWrapper = (NioSocketWrapper) sk.attachment();
// Attachment may be null if another thread has called
// cancelledKey()
if (socketWrapper != null) {
processKey(sk, socketWrapper);
}
}
// Process timeouts
timeout(keyCount,hasEvents);
}
getStopLatch().countDown();
}
在這里，有一個(gè)很重要的方法，org.apache.tomcat.util.net.NioEndpoint.Poller#events()，他是從Poller的事件隊(duì)列中獲取Acceptor接收到的可用socket，并將其注冊(cè)到Selector上
/**
* Processes events in the event queue of the Poller.
*
* @return <code>true</code> if some events were processed,
* <code>false</code> if queue was empty
*/
public boolean events() {
boolean result = false;
PollerEvent pe = null;
// 如果Acceptor將socket添加到隊(duì)列中，那么events.poll()方法就能拿到對(duì)應(yīng)的事件，否則拿不到就返回false
for (int i = 0, size = events.size(); i < size && (pe = events.poll()) != null; i++ ) {
result = true;
NioSocketWrapper socketWrapper = pe.getSocketWrapper();
SocketChannel sc = socketWrapper.getSocket().getIOChannel();
int interestOps = pe.getInterestOps();
if (sc == null) {
log.warn(sm.getString("endpoint.nio.nullSocketChannel"));
socketWrapper.close();
} else if (interestOps == OP_REGISTER) {
// 如果是Acceptor剛添加到隊(duì)列中的事件，那么此時(shí)的ops就是OP_REGISTER
try {，
// 將次socket注冊(cè)到selector上，標(biāo)記為OP_READ事件，添加事件觸發(fā)時(shí)處理函數(shù)socketWrapper
sc.register(getSelector(), SelectionKey.OP_READ, socketWrapper);
} catch (Exception x) {
log.error(sm.getString("endpoint.nio.registerFail"), x);
}
} else {
// ？？這里的邏輯，不清楚什么情況下會(huì)進(jìn)入到這個(gè)分支里面
final SelectionKey key = sc.keyFor(getSelector());
if (key == null) {
// The key was cancelled (e.g. due to socket closure)
// and removed from the selector while it was being
// processed. Count down the connections at this point
// since it won't have been counted down when the socket
// closed.
socketWrapper.close();
} else {
final NioSocketWrapper attachment = (NioSocketWrapper) key.attachment();
if (attachment != null) {
// We are registering the key to start with, reset the fairness counter.
try {
int ops = key.interestOps() | interestOps;
attachment.interestOps(ops);
key.interestOps(ops);
} catch (CancelledKeyException ckx) {
cancelledKey(key, socketWrapper);
}
} else {
cancelledKey(key, socketWrapper);
}
}
}
if (running && !paused && eventCache != null) {
pe.reset();
eventCache.push(pe);
}
}
return result;
}
還有一個(gè)重要方法就是org.apache.tomcat.util.net.NioEndpoint.Poller#processKey，上一個(gè)方法是獲取event，并注冊(cè)到selector，那這個(gè)方法就是通過(guò)Selector獲取到的數(shù)據(jù)準(zhǔn)備好的event，并開(kāi)始封裝成對(duì)應(yīng)的業(yè)務(wù)處理線程SocketProcessorBase，扔到線程池里開(kāi)始處理
protected void processKey(SelectionKey sk, NioSocketWrapper socketWrapper) {
try {
if (close) {
cancelledKey(sk, socketWrapper);
} else if (sk.isValid()) {
if (sk.isReadable() || sk.isWritable()) {
if (socketWrapper.getSendfileData() != null) {
processSendfile(sk, socketWrapper, false);
} else {
unreg(sk, socketWrapper, sk.readyOps());
boolean closeSocket = false;
// Read goes before write
if (sk.isReadable()) {
//這里如果是異步的操作，就會(huì)走這里
if (socketWrapper.readOperation != null) {
if (!socketWrapper.readOperation.process()) {
closeSocket = true;
}
} else if (socketWrapper.readBlocking) {
// readBlocking默認(rèn)為false
synchronized (socketWrapper.readLock) {
socketWrapper.readBlocking = false;
socketWrapper.readLock.notify();
}
} else if (!processSocket(socketWrapper, SocketEvent.OPEN_READ, true)) {
// 處理正常的事件，這里的processSocket就要正式開(kāi)始處理請(qǐng)求了。
// 將對(duì)應(yīng)的事件封裝成對(duì)應(yīng)的線程，然后交給線程池去處理正式的請(qǐng)求業(yè)務(wù)
closeSocket = true;
}
}
if (!closeSocket && sk.isWritable()) {
if (socketWrapper.writeOperation != null) {
if (!socketWrapper.writeOperation.process()) {
closeSocket = true;
}
} else if (socketWrapper.writeBlocking) {
synchronized (socketWrapper.writeLock) {
socketWrapper.writeBlocking = false;
socketWrapper.writeLock.notify();
}
} else if (!processSocket(socketWrapper, SocketEvent.OPEN_WRITE, true)) {
closeSocket = true;
}
}
if (closeSocket) {
cancelledKey(sk, socketWrapper);
}
}
}
} else {
// Invalid key
cancelledKey(sk, socketWrapper);
}
} catch (CancelledKeyException ckx) {
cancelledKey(sk, socketWrapper);
} catch (Throwable t) {
ExceptionUtils.handleThrowable(t);
log.error(sm.getString("endpoint.nio.keyProcessingError"), t);
}
}
請(qǐng)求具體處理
上一步，Selector獲取到了就緒的請(qǐng)求socket，然后根據(jù)socket注冊(cè)的觸發(fā)處理函數(shù)等，將這些數(shù)據(jù)進(jìn)行封裝，扔到了線程池里，開(kāi)始具體的業(yè)務(wù)邏輯處理。本節(jié)就是從工作線程封裝開(kāi)始，org.apache.tomcat.util.net.SocketProcessorBase為工作線程類的抽象類，實(shí)現(xiàn)了Runnable接口，不同的Endpoint實(shí)現(xiàn)具體的處理邏輯，本節(jié)以NioEndpoint為例
以下為org.apache.tomcat.util.net.AbstractEndpoint#processSocket方法源碼
/**
* Process the given SocketWrapper with the given status. Used to trigger
* processing as if the Poller (for those endpoints that have one)
* selected the socket.
*
* @param socketWrapper The socket wrapper to process
* @param event The socket event to be processed
* @param dispatch Should the processing be performed on a new
* container thread
*
* @return if processing was triggered successfully
*/
public boolean processSocket(SocketWrapperBase<S> socketWrapper,
SocketEvent event, boolean dispatch) {
try {
if (socketWrapper == null) {
return false;
}
// 優(yōu)先使用已經(jīng)存在的線程
SocketProcessorBase<S> sc = null;
if (processorCache != null) {
sc = processorCache.pop();
}
if (sc == null) {
sc = createSocketProcessor(socketWrapper, event);
} else {
sc.reset(socketWrapper, event);
}
// 獲取線程池。線程池的初始化，是在Acceptor、Poller這兩個(gè)單獨(dú)線程啟動(dòng)之前創(chuàng)建
// tomcat使用了自定義的org.apache.tomcat.util.threads.TaskQueue，這塊tomcat也進(jìn)行了小的適配開(kāi)發(fā)
// 核心線程為10個(gè)，最大200線程
Executor executor = getExecutor();
if (dispatch && executor != null) {
executor.execute(sc);
} else {
sc.run();
}
} catch (RejectedExecutionException ree) {
getLog().warn(sm.getString("endpoint.executor.fail", socketWrapper) , ree);
return false;
} catch (Throwable t) {
ExceptionUtils.handleThrowable(t);
// This means we got an OOM or similar creating a thread, or that
// the pool and its queue are full
getLog().error(sm.getString("endpoint.process.fail"), t);
return false;
}
return true;
}
上面的方法是得到了處理業(yè)務(wù)邏輯的線程SocketProcessorBase，NioEndpoint內(nèi)部類org.apache.tomcat.util.net.NioEndpoint.SocketProcessor繼承了這個(gè)抽象類，也就是具體的業(yè)務(wù)處理邏輯在org.apache.tomcat.util.net.NioEndpoint.SocketProcessor#doRun方法中，最終調(diào)用到我們的Servlet
protected void doRun() {
/*
* Do not cache and re-use the value of socketWrapper.getSocket() in
* this method. If the socket closes the value will be updated to
* CLOSED_NIO_CHANNEL and the previous value potentially re-used for
* a new connection. That can result in a stale cached value which
* in turn can result in unintentionally closing currently active
* connections.
*/
Poller poller = NioEndpoint.this.poller;
if (poller == null) {
socketWrapper.close();
return;
}
try {
int handshake = -1;
try {
// 握手相關(guān)判斷邏輯
...
} catch (IOException x) {
...
}
// 三次握手成功了
if (handshake == 0) {
SocketState state = SocketState.OPEN;
// Process the request from this socket
// event為SocketEvent.OPEN_READ，這個(gè)變量是org.apache.tomcat.util.net.NioEndpoint.Poller#processKey方法賦值
if (event == null) {
state = getHandler().process(socketWrapper, SocketEvent.OPEN_READ);
} else {
// 這里就開(kāi)始正式處理請(qǐng)求了
state = getHandler().process(socketWrapper, event);
}
if (state == SocketState.CLOSED) {
poller.cancelledKey(getSelectionKey(), socketWrapper);
}
} else if (handshake == -1 ) {
getHandler().process(socketWrapper, SocketEvent.CONNECT_FAIL);
poller.cancelledKey(getSelectionKey(), socketWrapper);
} else if (handshake == SelectionKey.OP_READ){
socketWrapper.registerReadInterest();
} else if (handshake == SelectionKey.OP_WRITE){
socketWrapper.registerWriteInterest();
}
} catch (CancelledKeyException cx) {
poller.cancelledKey(getSelectionKey(), socketWrapper);
} catch (VirtualmachineError vme) {
ExceptionUtils.handleThrowable(vme);
} catch (Throwable t) {
log.error(sm.getString("endpoint.processing.fail"), t);
poller.cancelledKey(getSelectionKey(), socketWrapper);
} finally {
socketWrapper = null;
event = null;
//return to cache
if (running && !paused && processorCache != null) {
processorCache.push(this);
}
}
}
總結(jié)
- Tomcat是如何接收網(wǎng)絡(luò)請(qǐng)求？
- 使用java nio的同步非阻塞去進(jìn)行網(wǎng)絡(luò)監(jiān)聽(tīng)。
- org.apache.tomcat.util.net.AbstractEndpoint#bindWithCleanup中初始化網(wǎng)絡(luò)監(jiān)聽(tīng)、SSL
- { .... serverSock = ServerSocketChannel.open(); socketProperties.setProperties(serverSock.socket()); InetSocketAddress addr = new InetSocketAddress(getAddress(), getPortWithOffset()); // 當(dāng)應(yīng)用層面的連接數(shù)到達(dá)最大值時(shí)，操作系統(tǒng)可以繼續(xù)接收連接，那么操作系統(tǒng)能繼續(xù)接收的最大連接數(shù)就是這個(gè)隊(duì)列長(zhǎng)度，可以通過(guò)acceptCount 參數(shù)配置，默認(rèn)是 100 serverSock.bind(addr, getAcceptCount()); } serverSock.configureBlocking(true); //mimic APR behavior
- org.apache.tomcat.util.net.NioEndpoint#startInternal中初始化業(yè)務(wù)處理的線程池、連接限制器、Poller線程、Acceptor線程
- 如何做到高性能的http協(xié)議服務(wù)器？
- Tomcat把接收連接、檢測(cè) I/O 事件以及處理請(qǐng)求進(jìn)行了拆分，用不同規(guī)模的線程去做對(duì)應(yīng)的事情，這也是tomcat能高并發(fā)處理請(qǐng)求的原因。不讓線程阻塞，盡量讓CPU忙起來(lái)
- 是怎么設(shè)計(jì)的呢？
- 通過(guò)接口、抽象類等，將不同的處理邏輯拆分，各司其職
- org.apache.tomcat.util.net.AbstractEndpoint：I/O事件的檢測(cè)、處理邏輯都在這個(gè)類的實(shí)現(xiàn)類里面。使用模板方法，不同的協(xié)議有不同的實(shí)現(xiàn)方法。NioEndpoint/Nio2Endpoint/AprEndpointorg.apache.tomcat.util.net.NioEndpoint.Poller：引用了java.nio.channels.Selector，內(nèi)部有個(gè)事件隊(duì)列，監(jiān)聽(tīng)I(yíng)/O事件具體就是在這里做的org.apache.tomcat.util.net.NioEndpoint.NioSocketWrapperorg.apache.tomcat.util.net.NioEndpoint.SocketProcessor: 具體處理請(qǐng)求的線程類
參考：
NioEndpoint組件：Tomcat如何實(shí)現(xiàn)非阻塞I/O？
Java NIO淺析
來(lái)源：
https://www.cnblogs.com/chenzw93/p/16072325.html
