1、RPC调用
dubbo服务调用只需在spring.xml中如下配置后,就可以调用本地方法一样,调用provider提供的远程服务:
<dubbo:reference id="demoService" interface="com.alibaba.dubbo.demo.DemoService" />
dubbo服务调用链路图
dubbo服务调用前部分链路如下图所示,下面根据这张图以调用com.alibaba.dubbo.demo.DemoService.sayHello("afei")
为例,一步一步分析dubbo服务的调用过程:
InvokerInvocationHandler
demoService.sayHello("afei")
这样的RPC调用,被Dubbo代理后,就会调用InvokerInvocationHandler中的invoke()
方法。源码如下:
public class InvokerInvocationHandler implements InvocationHandler {
private final Invoker<?> invoker;
public InvokerInvocationHandler(Invoker<?> handler){
this.invoker = handler;
}
public Object invoke(Object proxy, Method method, Object[] args) throws Throwable {
// 得到调用的方法名称
String methodName = method.getName();
Class<?>[] parameterTypes = method.getParameterTypes();
if (method.getDeclaringClass() == Object.class) {
return method.invoke(invoker, args);
}
// 调用toString()方法的特殊处理方式
if ("toString".equals(methodName) && parameterTypes.length == 0) {
return invoker.toString();
}
// 调用hashCode()方法的特殊处理方式
if ("hashCode".equals(methodName) && parameterTypes.length == 0) {
return invoker.hashCode();
}
// 调用equals()方法的特殊处理方式
if ("equals".equals(methodName) && parameterTypes.length == 1) {
return invoker.equals(args[0]);
}
// 常规的dubbo调用,都走这里,把调用的方法名称和参数封装成RpcInvocation对象,然后调用MockClusterInvoker中的invoke()方法
return invoker.invoke(new RpcInvocation(method, args)).recreate();
}
}
// RpcInvocation的定义如下,包含了一些RPC调用信息:方法名,参数类型,参数值,Dubbo调用的一些附属信息attachments,以及调用的Invoker(attachments和invoker在后面会赋值)
public class RpcInvocation implements Invocation, Serializable {
private static final long serialVersionUID = -4355285085441097045L;
private String methodName;
private Class<?>[] parameterTypes;
private Object[] arguments;
private Map<String, String> attachments;
private transient Invoker<?> invoker;
... ...
}
invoke()方法中
Object proxy
就是代理的对象;Method method
就是本次调用的方法,即DemoService中的sayHello(String)方法;Object[] args
就是调用的参数,即"afei",组装成Object[]就是new Object[]{"afei"}。
MockClusterInvoker
接下来分析MockClusterInvoker中的invoke()
方法。部分核心源码如下:
public class MockClusterInvoker<T> implements Invoker<T>{
... ...
public Result invoke(Invocation invocation) throws RpcException {
Result result = null;
// 获取mock的值,默认为false;
String value = directory.getUrl().getMethodParameter(invocation.getMethodName(), Constants.MOCK_KEY, Boolean.FALSE.toString()).trim();
if (value.length() == 0 || value.equalsIgnoreCase("false")){
// 如果在<dubbo:reference>中没有申明mock(默认方式),或者申明为false,那么走这里的逻辑
result = this.invoker.invoke(invocation);
} else if (value.startsWith("force")) {
// 强制mock调用方式的WARN日志
if (logger.isWarnEnabled()) {
logger.info("force-mock: " + invocation.getMethodName() + " force-mock enabled , url : " + directory.getUrl());
}
//force:direct mock
result = doMockInvoke(invocation, null);
} else{
//fail-mock
try {
// 普通的mock方式,例如申明mock="com.alibaba.dubbo.demo.consumer.mock.DemoServiceMock",那么在RPC调用抛出RPC异常时才启用mock调用;
result = this.invoker.invoke(invocation);
}catch (RpcException e) {
if (e.isBiz()) {
throw e;
} else {
if (logger.isWarnEnabled()) {
logger.info("fail-mock: " + invocation.getMethodName() + " fail-mock enabled , url : " + directory.getUrl(), e);
}
result = doMockInvoke(invocation, e);
}
}
}
... ...
}
}
mock申明方式:
<dubbo:reference id="demoService" interface="com.alibaba.dubbo.demo.DemoService" version="1.0.0" mock="false"/>
,从这段源码可知,dubbo提供了三种策略:
1、不需要mock,直接调用AbstractClusterInvoker(默认方式)
2、强制mock方式调用;
3、先AbstractClusterInvoker方式调用,如果有RpcException(比如没有任何可用的Provider),再以mock方式调用;想要详细了解mock的使用方式,请参考dubbo一些你不一定知道但是很好用的功能中的"本地伪装";
AbstractClusterInvoker
接下来调用AbstractClusterInvoker中的invoke()方法,部分源码如下所示:
public Result invoke(final Invocation invocation) throws RpcException {
checkWheatherDestoried();
LoadBalance loadbalance;
// 从Diectory中得到所有可用的,经过路由过滤的Invoker集合
List<Invoker<T>> invokers = list(invocation);
if (invokers != null && invokers.size() > 0) {
// 如果有可用的Invoker,那么根据第一个Invoker得到其LoadBalance策略
loadbalance = ExtensionLoader.getExtensionLoader(LoadBalance.class).getExtension(invokers.get(0).getUrl()
.getMethodParameter(invocation.getMethodName(),Constants.LOADBALANCE_KEY, Constants.DEFAULT_LOADBALANCE));
} else {
// // 如果没有可用的Invoker,那么采用默认的LoadBalance策略(随机策略)
loadbalance = ExtensionLoader.getExtensionLoader(LoadBalance.class).getExtension(Constants.DEFAULT_LOADBALANCE);
}
// 如果异步调用,那么在attachment中给id赋值(值是自增的,通过AtomicLong.getAndIncrement()得到)
RpcUtils.attachInvocationIdIfAsync(getUrl(), invocation);
// doInvoke()定义在AbstractClusterInvoker中是一个抽象方法,所以这里采用了模板方法设计模式,调用FailoverClusterInvoker(默认是failover集群容错)中的doInvoke()方法
return doInvoke(invocation, invokers, loadbalance);
}
protected abstract Result doInvoke(Invocation invocation, List<Invoker<T>> invokers,
LoadBalance loadbalance) throws RpcException;
FailoverClusterInvoker
FailoverClusterInvoker的分析,请参考dubbo源码-集群容错,这篇文章对dubbo支持的所有的集群容错处理都一一进行了比较详细的分析;但是不管哪种集群容错处理,接下来都会调用invoker.invoke(invocation)
得到Result;RPC调用invoker.invoke(invocation)
;的调用关系链图如下,根据这张图一步一步分析每个步骤:
InvokerWrapper
InvokerWrapper中会初始化Consumer端的调用过滤链,然后在FailoverClusterInvoker中调用invoker.invoke(invocation)
时一一执行每个Filter:
private static <T> Invoker<T> buildInvokerChain(final Invoker<T> invoker, String key, String group) {
Invoker<T> last = invoker;
// 得到Consumer端的Filter集合
List<Filter> filters = ExtensionLoader.getExtensionLoader(Filter.class).getActivateExtension(invoker.getUrl(), key, group);
if (filters.size() > 0) {
for (int i = filters.size() - 1; i >= 0; i --) {
final Filter filter = filters.get(i);
final Invoker<T> next = last;
last = new Invoker<T>() {
... ...
// 通过Filter的next()方法遍历执行Filter链上所有的Filter
public Result invoke(Invocation invocation) throws RpcException {
return filter.invoke(next, invocation);
}
... ...
};
}
}
return last;
}
Filter
Consumer端Filter有ConsumerContextFilter、FutureFilter、MonitorFilter等;这里不一一讲解,里面的业务都比较简单;执行完Filter链后,调用AbstractInvoker中的invoke()
方法;
AbstractInvoker
AbstractInvoker中申明了抽象方法:protected abstract Result doInvoke(Invocation invocation) throws Throwable;
,所以,这里会以模板方法设计模式调用DubboInvoker中的doInvoker()
方法;接下来的调用关系链如下图所示:
DubboInvoker
DubboInvoker.doInvoke(Invocation)
核心源码如下:
@Override
protected Result doInvoke(final Invocation invocation) throws Throwable {
RpcInvocation inv = (RpcInvocation) invocation;
final String methodName = RpcUtils.getMethodName(invocation);
// RpcInvocation中attachments设置path和version并赋值
inv.setAttachment(Constants.PATH_KEY, getUrl().getPath());
inv.setAttachment(Constants.VERSION_KEY, version);
ExchangeClient currentClient;
// 如果只有一个Client,直接选择;如果多个Client,轮询
if (clients.length == 1) {
currentClient = clients[0];
} else {
currentClient = clients[index.getAndIncrement() % clients.length];
}
try {
// 是否异步调用,默认false
boolean isAsync = RpcUtils.isAsync(getUrl(), invocation);
// 是否单边调用,即不需要等待返回结果,默认false
boolean isOneway = RpcUtils.isOneway(getUrl(), invocation);
// 获取Consumer侧的timeout,默认1s
int timeout = getUrl().getMethodParameter(methodName, Constants.TIMEOUT_KEY,Constants.DEFAULT_TIMEOUT);
if (isOneway) {
... ...
} else if (isAsync) {
... ...
} else {
// 重点关注这里,即默认实现
RpcContext.getContext().setFuture(null);
// 发送请求后,调用DefaultFuture.get()方法获取远程响应的结果
return (Result) currentClient.request(inv, timeout).get();
}
} catch (TimeoutException e) {
throw new RpcException(RpcException.TIMEOUT_EXCEPTION, "Invoke remote method timeout. method: " + invocation.getMethodName() + ", provider: " + getUrl() + ", cause: " + e.getMessage(), e);
} catch (RemotingException e) {
throw new RpcException(RpcException.NETWORK_EXCEPTION, "Failed to invoke remote method: " + invocation.getMethodName() + ", provider: " + getUrl() + ", cause: " + e.getMessage(), e);
}
}
说明:
currentClient.request(inv, timeout)
得到的是ResponseFuture类型结果,调用get()
返回Result对象;
HeaderExchangeChannel
HeaderExchangeChannel.request(Object request, int timeout)
核心源码如下:
public ResponseFuture request(Object request, int timeout) throws RemotingException {
... ...
// 构造一个准备RPC远程调用的Request类型参数
Request req = new Request();
req.setVersion("2.0.0");
req.setTwoWay(true);
// 将调用该方法前的RpcInvocation类型请求参数封装到Request中
req.setData(request);
DefaultFuture future = new DefaultFuture(channel, req, timeout);
try{
channel.send(req);
}catch (RemotingException e) {
future.cancel();
throw e;
}
return future;
}
DefaultFuture future = new DefaultFuture(channel, req, timeout);
源码解读:
这里比较重要,包含了通过netty调用后,如何拿到调用结果。
public DefaultFuture(Channel channel, Request request, int timeout){
this.channel = channel;
this.request = request;
// request.getId即得到这一次请求的id,id生成方式通过AtomicLong.getAndIncrement()得到;源码参考Request.newId();这个方法会不会溢出?getAndIncrement()增长到MAX_VALUE时,再增长会变为MIN_VALUE,负数也可以做为ID,所以不会溢出;
this.id = request.getId();
this.timeout = timeout > 0 ? timeout : channel.getUrl().getPositiveParameter(Constants.TIMEOUT_KEY, Constants.DEFAULT_TIMEOUT);
// put into waiting map.
// dubbo会根据请求ID从这个map中就能拿到对应的响应结果
FUTURES.put(id, this);
CHANNELS.put(id, channel);
}
说明:由于请求ID是从AtomicLong取得,所以理论上是唯一的;即使当达到AtomicLong的最大值后又从MIN_VALUE开始,理论上同一个ID对应的请求不可能存在这么长时间从而导致下一次轮回ID碰撞;
AtomicLong溢出问题可以通过下面一段代码验证:
/**
* @author afei
*/
public class AtomicLongTest {
public static void main(String[] args) {
AtomicLong al = new AtomicLong(Long.MAX_VALUE-2);
for (int i=0; i<5; i++){
System.out.println(al.getAndIncrement());
}
}
}
运行结果如下,达到MAX_VALUE后下一次getAndIncrement()就是MIN_VALUE,所以getAndIncrement()不会溢出:
9223372036854775805
9223372036854775806
9223372036854775807
-9223372036854775808
-9223372036854775807
AbstractClient
Channel准备发送请求消息到远程服务的核心源码:
public void send(Object message, boolean sent) throws RemotingException {
if (send_reconnect && !isConnected()){
connect();
}
Channel channel = getChannel();
//TODO getChannel返回的状态是否包含null需要改进
if (channel == null || ! channel.isConnected()) {
throw new RemotingException(this, "message can not send, because channel is closed . url:" + getUrl());
}
channel.send(message, sent);
}
NettyChannel
NettyChannel.send(Object message, boolean sent)
是真正调用Netty把请求消息通过NIO方式发给远程服务的地方,message即dubbo封装的Request类型请求参数,核心属性是mData,为RpcInvocation类型,源码如下:
public void send(Object message, boolean sent) throws RemotingException {
super.send(message, sent);
boolean success = true;
int timeout = 0;
try {
// 这里就是调用netty的NioClientSocketChannel.write(Object message)方法将请求message发送到Provider
ChannelFuture future = channel.write(message);
if (sent) {
timeout = getUrl().getPositiveParameter(Constants.TIMEOUT_KEY, Constants.DEFAULT_TIMEOUT);
success = future.await(timeout);
}
Throwable cause = future.getCause();
if (cause != null) {
throw cause;
}
}
... ...
}
2. 获取结果
通过Netty以NIO方式发送请求后,接下来分析dubbo如果拿到Provider响应的结果,并把结果和请求对应起来(因为是异步调用,不能把结果和请求对应关系搞混淆);由前面分析HeaderExchangeChannel可知,dubbo调用(Result) currentClient.request(inv, timeout).get()
,通过ResponseFuture.get()
方法得到RpcResult结果,ResponseFuture的实现就是DefaultFuture:
DefaultFuture
DefaultFuture中get()方法的核心源码:
public Object get(int timeout) throws RemotingException {
// 如果Consumer端指定的timeout不大于0,那么设置为默认值1s
if (timeout <= 0) {
timeout = Constants.DEFAULT_TIMEOUT;
}
// isDone()就是判断 response != null
if (! isDone()) {
long start = System.currentTimeMillis();
// 通过ReentrantLock锁保证线程安全,lock定义为:private final Lock lock = new ReentrantLock();
lock.lock();
try {
while (! isDone()) {
done.await(timeout, TimeUnit.MILLISECONDS);
if (isDone() || System.currentTimeMillis() - start > timeout) {
break;
}
}
} catch (InterruptedException e) {
throw new RuntimeException(e);
} finally {
lock.unlock();
}
if (! isDone()) {
throw new TimeoutException(sent > 0, channel, getTimeoutMessage(false));
}
}
return returnFromResponse();
}
private Object returnFromResponse() throws RemotingException {
// 全局申明的private volatile Response response就是结果,后面会分析response是怎么被赋值的;
Response res = response;
if (res == null) {
throw new IllegalStateException("response cannot be null");
}
// 如果是正常的结果,直接返回
if (res.getStatus() == Response.OK) {
return res.getResult();
}
// 如果是超时的结果,那么抛出超时异常
if (res.getStatus() == Response.CLIENT_TIMEOUT || res.getStatus() == Response.SERVER_TIMEOUT) {
throw new TimeoutException(res.getStatus() == Response.SERVER_TIMEOUT, channel, res.getErrorMessage());
}
throw new RemotingException(channel, res.getErrorMessage());
}
HeaderExchangeHandler
发送RPC请求后,在HeaderExchangeHandler.received()
中接收Porvider返回的响应结果(通过dubbo源码-NettyClient分析可知,NettyHandler是消息的handler,NettyHandler中的messageReceived()即消息接收方法,经过解码后,最终调用的就是HeaderExchangeHandler.received()),源码如下:
public void received(Channel channel, Object message) throws RemotingException {
channel.setAttribute(KEY_READ_TIMESTAMP, System.currentTimeMillis());
ExchangeChannel exchangeChannel = HeaderExchangeChannel.getOrAddChannel(channel);
try {
if (message instanceof Request) {
// handle request.
... ...
} else if (message instanceof Response) {
// 这里处理响应结果
handleResponse(channel, (Response) message);
}
... ...
} finally {
HeaderExchangeChannel.removeChannelIfDisconnected(channel);
}
}
static void handleResponse(Channel channel, Response response) throws RemotingException {
if (response != null && !response.isHeartbeat()) {
DefaultFuture.received(channel, response);
}
}
DefaultFure.received()
方法源码:
public static void received(Channel channel, Response response) {
try {
// 在response中封装了请求ID,根据请求ID得到DefaultFuture(根据请求id通过remove方式获取DefaultFuture的好处是,获取的同时也清理了FUTURES中这个ID对应的请求信息,防止FUTURES堆积)
DefaultFuture future = FUTURES.remove(response.getId());
if (future != null) {
// 接收Reponse结果,这就是请求id对应的结果
future.doReceived(response);
} else {
logger.warn("The timeout response finally returned at "
+ (new SimpleDateFormat("yyyy-MM-dd HH:mm:ss.SSS").format(new Date()))
+ ", response " + response
+ (channel == null ? "" : ", channel: " + channel.getLocalAddress()
+ " -> " + channel.getRemoteAddress()));
}
} finally {
CHANNELS.remove(response.getId());
}
}
private void doReceived(Response res) {
lock.lock();
try {
// 将reponse赋值给申明的:private volatile Response response;这就是请求id对应的结果
response = res;
if (done != null) {
done.signal();
}
} finally {
lock.unlock();
}
if (callback != null) {
invokeCallback(callback);
}
}
3、超时请求清理
对于那些耗时超过Consumer端timeout指定的值,且没有任何响应,dubbo如何处理呢?这些请求如果不处理的话,数据一致会积压在FUTURES这个Map中,dubbo采用的方法是在DefaultFuture中开启一个后台线程,死循环检测,源码如下:
private static class RemotingInvocationTimeoutScan implements Runnable {
public void run() {
while (true) {
try {
// 只要有请求,那么FUTURES就不为空,那么遍历这些请求
for (DefaultFuture future : FUTURES.values()) {
if (future == null || future.isDone()) {
continue;
}
// 如果耗时超过了Consumer端指定的timeout,那么返回特定status值的Response(future.isSent() ? Response.SERVER_TIMEOUT : Response.CLIENT_TIMEOUT),Consumer拿到这种Response后,判断它是Response.SERVER_TIMEOUT or Response.CLIENT_TIMEOUT,从而抛出TimeoutException异常;
if (System.currentTimeMillis() - future.getStartTimestamp() > future.getTimeout()) {
// create exception response.
Response timeoutResponse = new Response(future.getId());
// set timeout status.
timeoutResponse.setStatus(future.isSent() ? Response.SERVER_TIMEOUT : Response.CLIENT_TIMEOUT);
timeoutResponse.setErrorMessage(future.getTimeoutMessage(true));
// handle response.
DefaultFuture.received(future.getChannel(), timeoutResponse);
}
}
Thread.sleep(30);
} catch (Throwable e) {
logger.error("Exception when scan the timeout invocation of remoting.", e);
}
}
}
}
static {
// 静态代码块,即初始化创建名为"DubboResponseTimeoutScanTimer"的线程来获取调用超时的请求,并返回特定status的Response
Thread th = new Thread(new RemotingInvocationTimeoutScan(), "DubboResponseTimeoutScanTimer");
th.setDaemon(true);
th.start();
}