dubbo服务调用过程浅析

完成了服务的初始化之后,生成了代理,生成的动态代理类对接口的方法进行了包装,每次调用都会调用到InvocationHandler的invoke()方法,此方法中会进行远程服务调用一些列复杂过程,诸如网络通信,编码,解码,序列化等,然后将结果返回。在InvokerInvocationHandler.invoker()方法中,最后调用invoker.invoke(new RpcInvocation(method, args)),首先invoker为构造InvokerInvocationHandler传入的,具体类型为MockClusterInvoker,然后调用MockClusterInvoker.invoke-->FailfastClusterInvoker.doInvoke()

public Result invoke(Invocation invocation) throws RpcException {
        Result result = null;
        String value = directory.getUrl().getMethodParameter(invocation.getMethodName(), Constants.MOCK_KEY, Boolean.FALSE.toString()).trim(); 
        if (value.length() == 0 || value.equalsIgnoreCase("false")){//没有mock
            //no mock
            result = this.invoker.invoke(invocation);
   ..................................................................
}

然后调用到FailfastClusterInvoker.invoke(),此对象持有接口服务的RegistryDirectory,里面含有远程提供者的具体信息,是在服务消费者初始化时,通过订阅zk相应节点得到的。此方法的调用过程比较复杂,会调用负载均衡算法,根据一定的策略,选取一个提供者,生成DubboInvoker对象。调用过程:AbstractClusterInvoker.invoke()--> FailfastClusterInvoker.doInvoke()

public Result invoke(final Invocation invocation) throws RpcException {
        checkWhetherDestroyed();
        LoadBalance loadbalance;//加载loadbalance
        List> invokers = list(invocation);//根据method加载invokers
        if (invokers != null && invokers.size() > 0) {//默认采用随机策略的loadbalance
            loadbalance = ExtensionLoader.getExtensionLoader(LoadBalance.class).getExtension(invokers.get(0).getUrl()
                    .getMethodParameter(invocation.getMethodName(),Constants.LOADBALANCE_KEY, Constants.DEFAULT_LOADBALANCE));
        } else {
            loadbalance = ExtensionLoader.getExtensionLoader(LoadBalance.class).getExtension(Constants.DEFAULT_LOADBALANCE);
        }
        RpcUtils.attachInvocationIdIfAsync(getUrl(), invocation);
        /** invocation:方法,参数类型,参数
         *  invokers:此接口提供端的Invoker;
         *  loadbalance:均衡负载
         */
        return doInvoke(invocation, invokers, loadbalance);
    }

下面看如何加载invokers的,进入到 list(invocation)方法,调用过程:AbstractDirectory.list()-->RegistryDirectory.doList()

 public List> doList(Invocation invocation) {
         ........................................................
        List> invokers = null;
        // 本地method-->lnvokers对应关系
        Map>> localMethodInvokerMap = this.methodInvokerMap; 
        if (localMethodInvokerMap != null && localMethodInvokerMap.size() > 0) {
             //获取方法名称和参数
            String methodName = RpcUtils.getMethodName(invocation);
            Object[] args = RpcUtils.getArguments(invocation);
            if(args != null && args.length > 0 && args[0] != null
                    && (args[0] instanceof String || args[0].getClass().isEnum())) {
                invokers = localMethodInvokerMap.get(methodName + "." + args[0]); // 可根据第一个参数枚举路由
            }
            if(invokers == null) {//根据方法获取
                invokers = localMethodInvokerMap.get(methodName);
            }
            if(invokers == null) {//获取*匹配的
                invokers = localMethodInvokerMap.get(Constants.ANY_VALUE);
            }
            if(invokers == null) {//获取所有
                Iterator>> iterator = localMethodInvokerMap.values().iterator();
                if (iterator.hasNext()) {
                    invokers = iterator.next();
                }
            }
        }
        return invokers == null ? new ArrayList>(0) : invokers;
    }

下面回到AbstractDirectory.list(),在这个方法中进行路由规则的匹配,匹配规则,路由规则也是spi机制的,有脚本、文件、表达式,一般通过服务治理中心配置,与接口服务关联,具体参见dubbo官方文档。路由匹配完成之后,获取到provider端的所有可用的Invokers,回到AbstractClusterInvoker.invoke()方法,最终返回provider端可用的Invokers,然后调用FailfastClusterInvoker.doInvoke():

 public Result doInvoke(Invocation invocation, List> invokers, LoadBalance loadbalance) throws RpcException {
        checkInvokers(invokers, invocation);//检查invokers的可用性
        Invoker invoker = select(loadbalance, invocation, invokers, null);
        try {
            return invoker.invoke(invocation);
        } catch (Throwable e) {
            if (e instanceof RpcException && ((RpcException)e).isBiz()) { // biz exception.
                throw (RpcException) e;
            }
           .........................
        }
    }

select(loadbalance, invocation, invokers, null)会根据loadbalance进行负载均衡算法匹配,此处为随机选择:
com.alibaba.dubbo.rpc.cluster.loadbalance.RandomLoadBalance

 protected  Invoker doSelect(List> invokers, URL url, Invocation invocation) {
        int length = invokers.size(); // 总个数
        int totalWeight = 0; // 总权重
        boolean sameWeight = true; // 权重是否都一样
        for (int i = 0; i < length; i++) {
            int weight = getWeight(invokers.get(i), invocation);
            totalWeight += weight; // 累计总权重
            if (sameWeight && i > 0
                    && weight != getWeight(invokers.get(i - 1), invocation)) {
                sameWeight = false; // 计算所有权重是否一样
            }
        }
        if (totalWeight > 0 && ! sameWeight) {
            // 如果权重不相同且权重大于0则按总权重数随机
            int offset = random.nextInt(totalWeight);
            // 并确定随机值落在哪个片断上
            for (int i = 0; i < length; i++) {
                offset -= getWeight(invokers.get(i), invocation);
                if (offset < 0) {
                    return invokers.get(i);
                }
            }
        }
        // 如果权重相同或权重为0则均等随机
        return invokers.get(random.nextInt(length));
    }

回到FailfastClusterInvoker.doInvoke(),根据负载均衡算法返回一个Invoker,为dubboInvoker,下面进行DubboInvoke.invoke()的调用

  protected Result doInvoke(final Invocation invocation) throws Throwable {
        //RpcInvocation封装了调用方法的对象和参数类型,具体参数
        RpcInvocation inv = (RpcInvocation) invocation;
         //获取具体的方法名称
        final String methodName = RpcUtils.getMethodName(invocation);
        inv.setAttachment(Constants.PATH_KEY, getUrl().getPath());
        inv.setAttachment(Constants.VERSION_KEY, version);
        //调用客户端,在refer中已经初始化好了,为HeaderExchangeClient
        ExchangeClient currentClient;
        if (clients.length == 1) {
            currentClient = clients[0];
        } else {
            currentClient = clients[index.getAndIncrement() % clients.length];
        }
        try {
            boolean isAsync = RpcUtils.isAsync(getUrl(), invocation);
            boolean isOneway = RpcUtils.isOneway(getUrl(), invocation);
            int timeout = getUrl().getMethodParameter(methodName, Constants.TIMEOUT_KEY,Constants.DEFAULT_TIMEOUT);
            if (isOneway) {//没有返回参数
                boolean isSent = getUrl().getMethodParameter(methodName, Constants.SENT_KEY, false);
                currentClient.send(inv, isSent);
                RpcContext.getContext().setFuture(null);
                return new RpcResult();
            } else if (isAsync) {//异步
                ResponseFuture future = currentClient.request(inv, timeout) ;
                RpcContext.getContext().setFuture(new FutureAdapter(future));
                return new RpcResult();
            } else {//同步返回
                RpcContext.getContext().setFuture(null);
                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);
        }
    }
 
 

下面分析常用的同步返回,进入HeaderExchangeClient.request()方法:

 public ResponseFuture request(Object request, int timeout) throws RemotingException {
        return channel.request(request, timeout);
    }

上面方法的channel为HeaderExchangeChannel,在HeaderExchangeClient初始化的时候直接初始化的,里面封装了NettyClient:
com.alibaba.dubbo.remoting.exchange.support.header.HeaderExchangeChannel

 public ResponseFuture request(Object request, int timeout) throws RemotingException {
        if (closed) {
            throw new RemotingException(this.getLocalAddress(), null, "Failed to send request " + request + ", cause: The channel " + this + " is closed!");
        }
        // create request.
        Request req = new Request();
        req.setVersion("2.0.0");
        req.setTwoWay(true);
        req.setData(request);//request为RpcInvocation对象
        DefaultFuture future = new DefaultFuture(channel, req, timeout);
        try{
            channel.send(req);
        }catch (RemotingException e) {
            future.cancel();
            throw e;
        }
        return future;
    }

上面方法中channel为NettyClient,进入NettyClient.send()-->NettyChannel.send(),channel是netty的一个组件,负责客户端与服务端之间的链路传递,调用Netty框架的IO事件之后会触发Netty框架的IO事件处理链。
com.alibaba.dubbo.remoting.transport.netty.NettyChannel:

public void send(Object message, boolean sent) throws RemotingException {
        super.send(message, sent);
        boolean success = true;
        int timeout = 0;
        try {
            //调用netty channel写入请求 
            ChannelFuture future = channel.write(message);
            if (sent) {//是否等待请求消息发出,默认不等待
                timeout = getUrl().getPositiveParameter(Constants.TIMEOUT_KEY, Constants.DEFAULT_TIMEOUT);
                success = future.await(timeout);
            }
            Throwable cause = future.getCause();
            if (cause != null) {// 等待消息发出,消息发送失败将抛出异常
                throw cause;
            }
        } catch (Throwable e) {
            throw new RemotingException(this, "Failed to send message " + message + " to " + getRemoteAddress() + ", cause: " + e.getMessage(), e);
        }
        
        if(! success) {
            throw new RemotingException(this, "Failed to send message " + message + " to " + getRemoteAddress()
                    + "in timeout(" + timeout + "ms) limit");
        }
    }

下面来分析消费端和提供端网络请求的部分,利用netty的encoder和decoder,来写入协议内容和请求数据,这里涉及到序列化机制,dubbo默认使用hessian序列化,过程为:消费端将请求编码-->发送-->提供端将请求解码-->响应请求-->提供端响应编码-->发送-->消费端响应解码-->结果返回

1) 消费端请求编码

消费端初始化nettyclient时,添加了事件处理链:NettyCodecAdapter.decoder-->NettyCodecAdapter.encoder-->NettyHandler,NettyCodecAdapter.decoder为上行事件处理器,NettyCodecAdapter.encoder为下行事件处理器,NettyHandler为上下行事件处理器,chanel.write()是一个下行事件,NettyCodecAdapter.decoder和NettyHandler将会被调用,调用顺序为NettyHandler-->NettyCodecAdapter.decoder
调用NettyHandler中的writeRequested方法:(从注释可以看出)

   /**
     * Invoked when a message object (e.g: {@link ChannelBuffer}) was received
     * from a remote peer.
     */
  @Override
    public void writeRequested(ChannelHandlerContext ctx, MessageEvent e) throws Exception {
        super.writeRequested(ctx, e);
        NettyChannel channel = NettyChannel.getOrAddChannel(ctx.getChannel(), url, handler);
        try {
            handler.sent(channel, e.getMessage());
        } finally {
            NettyChannel.removeChannelIfDisconnected(ctx.getChannel());
        }
    }

handler.sent()调用链路为:NettyClient.send()-->AbstractPeer.send()

  public void sent(Channel ch, Object msg) throws RemotingException {
        if (closed) {
            return;
        }
        handler.sent(ch, msg);
    }

此handler为DubboProtocol一路传过来的,包装了几层

public ExchangeClient connect(URL url, ExchangeHandler handler) throws RemotingException {
        return new HeaderExchangeClient(Transporters.connect(url, new DecodeHandler(new HeaderExchangeHandler(handler))));
    }

调用链路为:DecodeHandler.send()-->HeaderExchangeHandler.send()-->ExchangeHandlerAdapter.send(),最后调到ChannelHandlerAdapter.send()没有任何操作,NettyHandler就调用完毕了。下面看核心encoder的处理,调用方法为:InternalEncoder.encode()

protected Object encode(ChannelHandlerContext ctx, Channel ch, Object msg) throws Exception {
            com.alibaba.dubbo.remoting.buffer.ChannelBuffer buffer =
                com.alibaba.dubbo.remoting.buffer.ChannelBuffers.dynamicBuffer(1024);
            NettyChannel channel = NettyChannel.getOrAddChannel(ch, url, handler);
            try {
                codec.encode(channel, buffer, msg);
            } finally {
                NettyChannel.removeChannelIfDisconnected(ch);
            }
            return ChannelBuffers.wrappedBuffer(buffer.toByteBuffer());
        }

codec是NettyCodecAdapter初始化时传过来的,类型为DubboCountCodec,DubboCountCodec.encode()会调用到DubboCodec.encode()

 public void encode(Channel channel, ChannelBuffer buffer, Object msg) throws IOException {
        if (msg instanceof Request) {
            encodeRequest(channel, buffer, (Request) msg);
        } else if (msg instanceof Response) {
            encodeResponse(channel, buffer, (Response) msg);
        } else {
            super.encode(channel, buffer, msg);
        }
    }

接着调用到encodeRequest(),下面的方法是按照dubbo协议进行编码,dubbo协议头,长度为16个字节,共128位,的内容如下:


协议头.PNG

0-15:魔数
16-20:序列化ID
21:是否事件数据
22:需要响应标识
23:标识是请求还是响应
24-31:状态(req为空)
32-95:请求id
96-127:实体数据长度

     protected void encodeRequest(Channel channel, ChannelBuffer buffer, Request req) throws IOException {
        Serialization serialization = getSerialization(channel);
        // header.
        byte[] header = new byte[HEADER_LENGTH];//16个字节的消息头
        // set magic number.
        Bytes.short2bytes(MAGIC, header);//前两个字节魔数

        // 第三个字节标识序列化id、请求信息等,对应16-23位
        header[2] = (byte) (FLAG_REQUEST | serialization.getContentTypeId());

        if (req.isTwoWay()) header[2] |= FLAG_TWOWAY;//
        if (req.isEvent()) header[2] |= FLAG_EVENT;

        //32-95 请求id
        Bytes.long2bytes(req.getId(), header, 4);

        // encode request data.
        int savedWriteIndex = buffer.writerIndex();
        buffer.writerIndex(savedWriteIndex + HEADER_LENGTH);
        ChannelBufferOutputStream bos = new ChannelBufferOutputStream(buffer);
        ObjectOutput out = serialization.serialize(channel.getUrl(), bos);
        if (req.isEvent()) {//序列化数据
            encodeEventData(channel, out, req.getData());
        } else {
            encodeRequestData(channel, out, req.getData());
        }
        out.flushBuffer();
        bos.flush();
        bos.close();
        int len = bos.writtenBytes();
        checkPayload(channel, len);//检查请求数据实体长度
        Bytes.int2bytes(len, header, 12);//数据实体长度写入请求头

        // write
        buffer.writerIndex(savedWriteIndex);
        buffer.writeBytes(header); // write header.
        buffer.writerIndex(savedWriteIndex + HEADER_LENGTH + len);
    }

下面进入 encodeRequestData(channel, out, req.getData()),看看序列化,默认使用hessian2序列化,消息体数据包含dubbo版本号、接口名称、接口版本、方法名称、参数类型列表、参数、附加信息,把它们按顺序依次序列化,数据写入到类型为ChannelBuffer的buffer参数中,接着将ChannelBuffer封装成netty的org.jboss.netty.buffer.ChannelBuffer,netty会将数据写到链路发送到服务端。

 @Override
    protected void encodeRequestData(Channel channel, ObjectOutput out, Object data) throws IOException {
        RpcInvocation inv = (RpcInvocation) data;

        out.writeUTF(inv.getAttachment(Constants.DUBBO_VERSION_KEY, DUBBO_VERSION));
        out.writeUTF(inv.getAttachment(Constants.PATH_KEY));
        out.writeUTF(inv.getAttachment(Constants.VERSION_KEY));

        out.writeUTF(inv.getMethodName());
        out.writeUTF(ReflectUtils.getDesc(inv.getParameterTypes()));
        Object[] args = inv.getArguments();
        if (args != null)
        for (int i = 0; i < args.length; i++){
            out.writeObject(encodeInvocationArgument(channel, inv, i));
        }
        out.writeObject(inv.getAttachments());
    }
2) 提供端响应请求

此步骤包括:提供端将请求解码,响应请求,将结果编码,返回到消费端
2.1 提供端将请求解码
客户端将数据发送到服务端,netty服务端会触发decoder和handler两个处理器,进入com.alibaba.dubbo.remoting.transport.netty.NettyCodecAdapter.InternalDecoder.messageReceived()方法,
首先将netty的ChannelBuffer转成dubbo的ChannelBuffer,这里需要处理半包问题。

 @Override
 public void messageReceived(ChannelHandlerContext ctx, MessageEvent event) throws Exception {
   .................................................
   NettyChannel channel = NettyChannel.getOrAddChannel(ctx.getChannel(), url, handler);
            Object msg;
            int saveReaderIndex;

            try {
                // decode object.
                do {//循环处理半包问题
                    saveReaderIndex = message.readerIndex();//保存当前解析的位置等待链路的下次IO事件
                    try {
                        msg = codec.decode(channel, message);如果数据不完整返回NEED_MORE_INPUT
                    } catch (IOException e) {
                        buffer = com.alibaba.dubbo.remoting.buffer.ChannelBuffers.EMPTY_BUFFER;
                        throw e;
                    }
                    if (msg == Codec2.DecodeResult.NEED_MORE_INPUT) {//从上次解析位置读取
                        message.readerIndex(saveReaderIndex);
                        break;
                    } else {
                        if (saveReaderIndex == message.readerIndex()) {
                            buffer = com.alibaba.dubbo.remoting.buffer.ChannelBuffers.EMPTY_BUFFER;
                            throw new IOException("Decode without read data.");
                        }
                        if (msg != null) {
                            Channels.fireMessageReceived(ctx, msg, event.getRemoteAddress());
                        }
                    }
                } while (message.readable());
     .................................................
  }

DubboCodec.decode()->ExchangeCodec.decode(),进入到ExchangeCodec.decode():

 protected Object decode(Channel channel, ChannelBuffer buffer, int readable, byte[] header) throws IOException {
   ....................................................//检查协议头中的长度、魔数等
    ChannelBufferInputStream is = new ChannelBufferInputStream(buffer, len);
    return decodeBody(channel, is, header);
}

下面进入decodeBody(),首先初始化一个Request对象,创建DecodeableRpcInvocation对象,此对象为rpcInvocation的子类,调用此对象的decode()序列化出请求的method,paramtype等信息 set到DecodeableRpcInvocation中,最后将此对象set到request的data对象中。

protected Object decodeBody(Channel channel, InputStream is, byte[] header) throws IOException {
        byte flag = header[2], proto = (byte) (flag & SERIALIZATION_MASK);
        Serialization s = CodecSupport.getSerialization(channel.getUrl(), proto);
        // get request id.
        long id = Bytes.bytes2long(header, 4);
        if ((flag & FLAG_REQUEST) == 0) {
            // decode response.
        }else {
              Object data;
             ......................................
               DecodeableRpcInvocation inv;
               if (channel.getUrl().getParameter(
                        Constants.DECODE_IN_IO_THREAD_KEY,
                        Constants.DEFAULT_DECODE_IN_IO_THREAD)) {
                        inv = new DecodeableRpcInvocation(channel, req, is, proto);
                        inv.decode();
               } else {
                        inv = new DecodeableRpcInvocation(channel, req,
                                                          new UnsafeByteArrayInputStream(readMessageData(is)), proto);
               }
                    data = inv;
            }
          req.setData(data);
         ..................................
          return req;
        }

回调InternalDecoder.messageReceived()方法,调用Channels.fireMessageReceived()方法,激活下一个处理器的messageReceived事件,并且把解码后的对象封装在MessageEvent中。

2.2 提供端处理请求
decoder之后,继续出发netty的下一个处理链handler,进入NettyHandler的messageReceived()方法,

   @Override
    public void messageReceived(ChannelHandlerContext ctx, MessageEvent e) throws Exception {
        NettyChannel channel = NettyChannel.getOrAddChannel(ctx.getChannel(), url, handler);
        try {
            handler.received(channel, e.getMessage());
        } finally {
            NettyChannel.removeChannelIfDisconnected(ctx.getChannel());
        }
    }

此handler为createServer传递过来的,封装了多层,涉及到服务调用线程的初始化,具体可看《dubbo 线程模型浅析》,此处只分析任务丢到线程池的执行过程:DecodeHandler->HeaderExchangeHandler->ExchangeHandler

 public class HeaderExchanger implements Exchanger {
    
    public static final String NAME = "header";

    public ExchangeClient connect(URL url, ExchangeHandler handler) throws RemotingException {
        return new HeaderExchangeClient(Transporters.connect(url, new DecodeHandler(new HeaderExchangeHandler(handler))));
    }

    public ExchangeServer bind(URL url, ExchangeHandler handler) throws RemotingException {
        return new HeaderExchangeServer(Transporters.bind(url, new DecodeHandler(new HeaderExchangeHandler(handler))));
    }

}

最终调到HeaderExchangeHandler.receive()方法,此处的message为刚才decoder传递过来的request类型

 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.
                Request request = (Request) message;
                if (request.isEvent()) {
                    handlerEvent(channel, request);
                } else {
                    if (request.isTwoWay()) {
                        Response response = handleRequest(exchangeChannel, request);
                        channel.send(response);
                    } else {
                        handler.received(exchangeChannel, request.getData());
                    }
                }
            }
         .......................................................
    }

进入handleRequest(),取出request中的data对象,为invocation,然后进入到DubboProtocol内部类ExchangeHandlerAdapter的reply()方法

 Response handleRequest(ExchangeChannel channel, Request req) throws RemotingException {
        Response res = new Response(req.getId(), req.getVersion());
        ..........................................................
        // find handler by message class.
        Object msg = req.getData();
        try {
            // handle data.
            Object result = handler.reply(channel, msg);
            res.setStatus(Response.OK);
            res.setResult(result);
        } catch (Throwable e) {
            res.setStatus(Response.SERVICE_ERROR);
            res.setErrorMessage(StringUtils.toString(e));
        }
        return res;
    }

在DubboProtocol.ExchangeHandlerAdapter.reply()中,传入的参数分别是ExchangeChannel、Invocation,根据这两个参数拼装serviceKey,然后从exporterMap中找到Invoker,exportMap在服务发布的时候,DubboProtocol.export()方法put进去的。

private ExchangeHandler requestHandler = new ExchangeHandlerAdapter() {
        
        public Object reply(ExchangeChannel channel, Object message) throws RemotingException {
            if (message instanceof Invocation) {
                Invocation inv = (Invocation) message;
                Invoker invoker = getInvoker(channel, inv);
                //如果是callback 需要处理高版本调用低版本的问题
                if (Boolean.TRUE.toString().equals(inv.getAttachments().get(IS_CALLBACK_SERVICE_INVOKE))){
                    String methodsStr = invoker.getUrl().getParameters().get("methods");
                    boolean hasMethod = false;
                    if (methodsStr == null || methodsStr.indexOf(",") == -1){
                        hasMethod = inv.getMethodName().equals(methodsStr);
                    } else {
                        String[] methods = methodsStr.split(",");
                        for (String method : methods){
                            if (inv.getMethodName().equals(method)){
                                hasMethod = true;
                                break;
                            }
                        }
                    }
                    if (!hasMethod){
                        logger.warn(new IllegalStateException("The methodName "+inv.getMethodName()+" not found in callback service interface ,invoke will be ignored. please update the api interface. url is:" + invoker.getUrl()) +" ,invocation is :"+inv );
                        return null;
                    }
                }
                RpcContext.getContext().setRemoteAddress(channel.getRemoteAddress());
                return invoker.invoke(inv);//执行调用
            }
      ...................................................

此invoker为包装类型,包含filter执行链、提供端接口实现类的包装类,执行invoke()方法,先调用filter执行链,然后调用AbstractProxyInvoker实现类的doInvoke()方法,进行调用的包装类的invokeMethod()方法,包装类持有具体实现类,至此调用完毕,回到HeaderExchangeHandler.receive()方法,调用channel.send()写到客户端

 Response response = handleRequest(exchangeChannel, request);
 channel.send(response);

写到客户端仍然会触发netty事件链,接着调用NettyCodecAdapter.encoder和NettyHandler,InternalEncoder.encode()-->ExchangeCodec.encode():

public void encode(Channel channel, ChannelBuffer buffer, Object msg) throws IOException {
        if (msg instanceof Request) {
            encodeRequest(channel, buffer, (Request) msg);
        } else if (msg instanceof Response) {
            encodeResponse(channel, buffer, (Response) msg);
        } else {
            super.encode(channel, buffer, msg);
        }
    }

在encodeResponse()中,与消费端一样,将协议头,结果作为消息体写入ChannelBuffer中,在encodeResponse()中调用了encodeResponseData(),进行序列化

 protected void encodeResponseData(Channel channel, ObjectOutput out, Object data) throws IOException {
        Result result = (Result) data;

        Throwable th = result.getException();
        if (th == null) {//处理结果没有异常
            Object ret = result.getValue();
            if (ret == null) {//结果为null
                out.writeByte(RESPONSE_NULL_VALUE);
            } else {//结果不为空
                out.writeByte(RESPONSE_VALUE);
                out.writeObject(ret);
            }
        } else {//处理结果异常
            out.writeByte(RESPONSE_WITH_EXCEPTION);
            out.writeObject(th);
        }
    }

回到InternalEncoder.encode()中,调用ChannelBuffers.wrappedBuffer(buffer.toByteBuffer()),将ChannelBuffer写回链路,服务端所有处理结束。

2.3消费端响应结果
服务端写数据到客户端调用netty事件处理器:NettyCodecAdapter.decoder和NettyHandler,调用链路:NettyCodecAdapter.InternalDecoder. messageReceived()-->ExchangeCodec.decode()-->DubboCodec.decodeBody()

protected Object decodeBody(Channel channel, InputStream is, byte[] header) throws IOException {
 byte flag = header[2], proto = (byte) (flag & SERIALIZATION_MASK);
        Serialization s = CodecSupport.getSerialization(channel.getUrl(), proto);
        // get request id.
        long id = Bytes.bytes2long(header, 4);
        if ((flag & FLAG_REQUEST) == 0) {
            // decode response.
            Response res = new Response(id);
            if ((flag & FLAG_EVENT) != 0) {
                res.setEvent(Response.HEARTBEAT_EVENT);
            }
            // get status.
            byte status = header[3];
            res.setStatus(status);
            if (status == Response.OK) {
                try {
                    Object data;
                    ............................................
                        DecodeableRpcResult result;
                        if (channel.getUrl().getParameter(
                            Constants.DECODE_IN_IO_THREAD_KEY,
                            Constants.DEFAULT_DECODE_IN_IO_THREAD)) {
                            result = new DecodeableRpcResult(channel, res, is,
                                                             (Invocation)getRequestData(id), proto);
                            result.decode();//DecodeableRpcResult.decode()
                        } 
                      .......................................
                        data = result;
                    }
                    res.setResult(data);
              .....................................................
            return res;
        } else{
      }

}

上面的方法会调用DecodeableRpcResult.decode(),调完之后,将DecodeableRpcResult set到response对象的data属性中,DecodeableRpcResult为RpcResult的子类,进入到DecodeableRpcResult.decode():

public Object decode(Channel channel, InputStream input) throws IOException {
        ObjectInput in = CodecSupport.getSerialization(channel.getUrl(), serializationType)
            .deserialize(channel.getUrl(), input);

        byte flag = in.readByte();
        switch (flag) {
            case DubboCodec.RESPONSE_NULL_VALUE://正常返回,空值
                break;
            case DubboCodec.RESPONSE_VALUE://正常返回,有值
                try {
                    Type[] returnType = RpcUtils.getReturnTypes(invocation);
                    setValue(returnType == null || returnType.length == 0 ? in.readObject() :
                                 (returnType.length == 1 ? in.readObject((Class) returnType[0])
                                     : in.readObject((Class) returnType[0], returnType[1])));
                } catch (ClassNotFoundException e) {
                    throw new IOException(StringUtils.toString("Read response data failed.", e));
                }
                break;
            case DubboCodec.RESPONSE_WITH_EXCEPTION://异常返回
             ..................................
            default:
                throw new IOException("Unknown result flag, expect '0' '1' '2', get " + flag);
        }
        return this;
    }

在执行结果有值,正常返回的情况下,会序列化出返回值,调用setValue()设置到result中。
然后触发NettyHandler事件,调用NettyHandler中的messageReceived():

 @Override
    public void messageReceived(ChannelHandlerContext ctx, MessageEvent e) throws Exception {
        NettyChannel channel = NettyChannel.getOrAddChannel(ctx.getChannel(), url, handler);
        try {
            handler.received(channel, e.getMessage());
        } finally {
            NettyChannel.removeChannelIfDisconnected(ctx.getChannel());
        }
    }

这个handler的封装关系在前面已经说明:DecodeHandler->HeaderExchangeHandler->ExchangeHandler,一路调用直到HeaderExchangeHandler.received():

 public void received(Channel channel, Object message) throws RemotingException {
  ........................
  else if (message instanceof Response) {
      handleResponse(channel, (Response) message);
   }
  .......................
}

handleResponse()的调用链路为:DefaultFuture.received()-->DefaultFuture.doReceive():

private void doReceived(Response res) {
        lock.lock();
        try {
            response = res;//将resposne设置到DefaultFuture中
            if (done != null) {
                done.signal();//唤起消费端调用等待线程
            }
        } finally {
            lock.unlock();
        }
        if (callback != null) {
            invokeCallback(callback);
        }
    }

调用 done.signal()的作用是唤起消费端调用等待线程,在消费端向提供端发送调用请求后,会调用DefaultFuture.get阻塞等待响应结果,回到DubboInvoker.doInvoke()方法:

 protected Result doInvoke(final Invocation invocation) throws Throwable {
            ..............................
            boolean isAsync = RpcUtils.isAsync(getUrl(), invocation);
            boolean isOneway = RpcUtils.isOneway(getUrl(), invocation);
            int timeout = getUrl().getMethodParameter(methodName, Constants.TIMEOUT_KEY,Constants.DEFAULT_TIMEOUT);
            if (isOneway) {
              .............................
            } else if (isAsync) {
            .............................
            } else {
                RpcContext.getContext().setFuture(null);
                return (Result) currentClient.request(inv, timeout).get();
            }
   }

currentClient.request()最终调到HeaderExchangeChannel.request(Object request, int timeout):

public ResponseFuture request(Object request, int timeout) throws RemotingException {
       ...........................................
        // create request.
        Request req = new Request();//初始化的时候会设置一个id
        req.setVersion("2.0.0");
        req.setTwoWay(true);
        req.setData(request);//RpcInvocation对象
        //DefaultFuture初始化的时候会将id 当前对象保存起来,以备返回的时候使用rpcInvocation
        DefaultFuture future = new DefaultFuture(channel, req, timeout);
        try{
            channel.send(req);//调用netty发送
        }catch (RemotingException e) {
            future.cancel();
            throw e;
        }
        return future;
    }

可以看到返回的是DefaultFuture对象,进入此对象的get()方法:

  public Object get(int timeout) throws RemotingException {
       ..............................................
         while (! isDone()) {//等待的条件为response为null
              done.await(timeout, TimeUnit.MILLISECONDS);
              if (isDone() || System.currentTimeMillis() - start > timeout) {
                   break;
                }
             }
         } 
    ..............................................
        return returnFromResponse();
    }

  private Object returnFromResponse() throws RemotingException {
        Response res = response;
        if (res == null) {
            throw new IllegalStateException("response cannot be null");
        }
        if (res.getStatus() == Response.OK) {
            //see Dubbocodec.decodebody(),返回 DecodeableRpcResult
            return res.getResult();
        }
        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());
    }

回到DubboInvoker.doInvoke(),将结果返回,回到最初的InvokerInvocationHandler.invoke()

 public Object invoke(Object proxy, Method method, Object[] args) throws Throwable {
        String methodName = method.getName();
        Class[] parameterTypes = method.getParameterTypes();
        if (method.getDeclaringClass() == Object.class) {
            return method.invoke(invoker, args);
        }
        if ("toString".equals(methodName) && parameterTypes.length == 0) {
            return invoker.toString();
        }
        if ("hashCode".equals(methodName) && parameterTypes.length == 0) {
            return invoker.hashCode();
        }
        if ("equals".equals(methodName) && parameterTypes.length == 1) {
            return invoker.equals(args[0]);
        }
        return invoker.invoke(new RpcInvocation(method, args)).recreate();
    }

经过一些列复杂的过程,执行 invoker.invoke(new RpcInvocation(method, args))完成,返回ecodeableRpcResult,然后调用它的recreate()获取result值,此值就是提供端的执行结果,具体可以回顾上面的流程,将具体的结果返回,整个rpc调用流程完毕。

参考:
http://blog.csdn.net/pentiumchen/article/details/53227844
http://www.imooc.com/article/details/id/22597
http://blog.kazaff.me/2015/02/02/dubbo%E7%9A%84%E6%9C%8D%E5%8A%A1%E6%B2%BB%E7%90%86%E7%BB%86%E8%8A%82/
http://blog.csdn.net/manzhizhen/article/details/73436619

你可能感兴趣的:(dubbo服务调用过程浅析)