RocketMQ源码学习-通信与协议

系列文章总目录

github clone 最新源码,结构如下:
源码结构图
本篇文章要讲的通信与协议部分的源代码在remoting模块下。remoting模块是复杂网络通信的模块,为其他需要网络通信的模块所依赖。在这个模块中,RocketMQ定义了基础的通信协议,结合Netty,使得端与端之间的数据交互变得统一而高效。

基本类图

先来看这个模块的类关系图:
通信模块类图
针对每个类分别做解释:

  • RemotingService 为最上层接口
  • RemotingClient 继承自RemotingService,提供了client端接口定义
  • RemotingServer 继承自RemotingService,提供了server端接口定义
  • NettyRemotingAbstract 使用Netty作为通信框架的抽象类,包含很多公共的处理逻辑和数据结构
  • NettyRemotingClient 继承了NettyRemotingAbstract,并实现了RemotingClient接口,作为通信的client端
  • NettyRemotingServer 继承了NettyRemotingAbstract,并实现了RemotingServer接口,作为通信的server端
  • NettyEvent、NettyEncoder、NettyDecoder、RemotingCommand等为通信框架使用的类

通信协议

RocketMQ的通信协议如下:
整个通信消息分为四个部分:
通信协议

  1. 整体消息长度,占用四个资格
  2. 序列化类型和消息头长度,一共占用四个字节,第一个字节表示序列化类型,后三个字节表示消息头长度
  3. 消息头数据
  4. 消息主体数据

源码详细分析之通信流程

RemotingService
public interface RemotingService {
   
    void start();
    void shutdown();
    void registerRPCHook(RPCHook rpcHook);
}

RemotingService定义了三个基本的方法,registerRPCHook()注册一些钩子,用来做一些通信前后的处理。

RemotingClient 和 RemotingServer
public interface RemotingClient extends RemotingService {
   
	
    void updateNameServerAddressList(final List<String> addrs);
	
    List<String> getNameServerAddressList();
	// 同步调用
    RemotingCommand invokeSync(final String addr, final RemotingCommand request,
        final long timeoutMillis) throws InterruptedException, RemotingConnectException,
        RemotingSendRequestException, RemotingTimeoutException;
	// 异步调用
    void invokeAsync(final String addr, final RemotingCommand request, final long timeoutMillis,
        final InvokeCallback invokeCallback) throws InterruptedException, RemotingConnectException,
        RemotingTooMuchRequestException, RemotingTimeoutException, RemotingSendRequestException;
	// 单向调用
    void invokeOneway(final String addr, final RemotingCommand request, final long timeoutMillis)
        throws InterruptedException, RemotingConnectException, RemotingTooMuchRequestException,
        RemotingTimeoutException, RemotingSendRequestException;
	
    void registerProcessor(final int requestCode, final NettyRequestProcessor processor,
        final ExecutorService executor);
	
    void setCallbackExecutor(final ExecutorService callbackExecutor);

    ExecutorService getCallbackExecutor();

    boolean isChannelWritable(final String addr);
}

public interface RemotingServer extends RemotingService {
   

    void registerProcessor(final int requestCode, final NettyRequestProcessor processor,
        final ExecutorService executor);

    void registerDefaultProcessor(final NettyRequestProcessor processor, final ExecutorService executor);

    int localListenPort();

    Pair<NettyRequestProcessor, ExecutorService> getProcessorPair(final int requestCode);
	// 同步调用
    RemotingCommand invokeSync(final Channel channel, final RemotingCommand request,
        final long timeoutMillis) throws InterruptedException, RemotingSendRequestException,
        RemotingTimeoutException;
	// 异步调用
    void invokeAsync(final Channel channel, final RemotingCommand request, final long timeoutMillis,
        final InvokeCallback invokeCallback) throws InterruptedException,
        RemotingTooMuchRequestException, RemotingTimeoutException, RemotingSendRequestException;
	// 单向调用
    void invokeOneway(final Channel channel, final RemotingCommand request, final long timeoutMillis)
        throws InterruptedException, RemotingTooMuchRequestException, RemotingTimeoutException,
        RemotingSendRequestException;

}

比较重要的三个方法:同步调用、异步调用和单向调用。

NettyRemotingAbstract 部分核心代码

先看NettyRemotingAbstract定义的属性和构造方法

public abstract class NettyRemotingAbstract {
   
    protected final Semaphore semaphoreOneway; // 单向调用信号量,控制并发
    protected final Semaphore semaphoreAsync; // 异步调用信号量,控制并发
    protected final ConcurrentMap<Integer /* opaque */, ResponseFuture> responseTable =
        new ConcurrentHashMap<Integer, ResponseFuture>(256); // 处理中的请求,opaque是唯一的请求id
    protected final HashMap<Integer/* request code */, Pair<NettyRequestProcessor, ExecutorService>> processorTable = // requestCode对应的请求处理Processor和线程池
        new HashMap<Integer, Pair<NettyRequestProcessor, ExecutorService>>(64);
    protected Pair<NettyRequestProcessor, ExecutorService> defaultRequestProcessor; // 默认处理Processor

    public NettyRemotingAbstract(final int permitsOneway, final int permitsAsync) {
   
        this.semaphoreOneway = new Semaphore(permitsOneway, true);
        this.semaphoreAsync = new Semaphore(permitsAsync, true);
    }

上面是定义的属性,两个信号量用来控制单向调用和异步调用的并发量。responseTable维护处理中的请求,请求处理完成后会进行剔除,详见下面的scanResponseTable()方法。processorTable维护requestCode和处理线程的映射关系,处理请求时从processorTable根据requestCode查询出Pair,使用Pair进行处理。构造方法初始化两个信号量。Semaphore继承自AQS,这里不详细描述,可以参见AQS源码学习

下面是处理请求的逻辑, processMessageReceived()为入口,根据请求类型(request or reponse)做不同处理

   public void processMessageReceived(ChannelHandlerContext ctx, RemotingCommand msg) throws Exception {
   
       final RemotingCommand cmd = msg;
       if (cmd != null) {
   
           switch (cmd.getType()) {
   
               case REQUEST_COMMAND:
                   processRequestCommand(ctx, cmd);
                   break;
               case RESPONSE_COMMAND:
                   processResponseCommand(ctx, cmd);
                   break;
               default:
                   break;
           }
       }
   }

先来看处理Request的逻辑,使用processorTable中查询的processor和线程池进行逻辑处理,并进行rpcHook的调用。

    public void processRequestCommand(final ChannelHandlerContext ctx, final RemotingCommand cmd) {
   
        final Pair<NettyRequestProcessor, ExecutorService> matched = this.processorTable.get(cmd.getCode());// 根据请求类型查询处理的processor
        final Pair<NettyRequestProcessor, ExecutorService> pair = null == matched ? this.defaultRequestProcessor : matched;
        final int opaque = cmd.getOpaque();
        if (pair != null) {
   
            Runnable run = new Runnable() {
   
                @Override
                public void run() {
   
                    try {
   
                        RPCHook rpcHook = NettyRemotingAbstract.this.getRPCHook();
                        if (rpcHook != null) {
    // 请求前置处理
                            rpcHook.doBeforeRequest(RemotingHelper.parseChannelRemoteAddr(ctx.channel()), cmd);
                        }
                        final RemotingCommand response = pair.getObject1().processRequest(ctx, cmd);
                        if (rpcHook != null) {
    // 请求后置处理
                            rpcHook.doAfterResponse(RemotingHelper.parseChannelRemoteAddr(ctx.channel()), cmd, response);
                        }
                        if (!cmd.isOnewayRPC()) {
    // 非单向请求,回写处理结果
                            if (response != null) {
   
                                response.setOpaque(opaque);
                                response.markResponseType();
                                try {
   
                                    ctx.writeAndFlush(response);
                                } catch (Throwable e) {
   
                                    // 略
                                }
                            } else {
   
                            }
                        }
                    } catch (Throwable e) {
   
							// 略
                    }
                }
            };
            if (pair.getObject1().rejectRequest()) 
  • 1
    点赞
  • 2
    收藏
    觉得还不错? 一键收藏
  • 0
    评论

“相关推荐”对你有帮助么?

  • 非常没帮助
  • 没帮助
  • 一般
  • 有帮助
  • 非常有帮助
提交
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值