motan源码分析十一：部分特性

本章将描述motan部分的特性并对源码进行分析。

1.requestid的维护，使用了当前时间左移20位，再和一个自增变量组合

public class RequestIdGenerator {
    protected static final AtomicLong offset = new AtomicLong(0);
    protected static final int BITS = 20;
    protected static final long MAX_COUNT_PER_MILLIS = 1 << BITS;


    /**
     * 获取 requestId
     * 
     * @return
     */
    public static long getRequestId() {
        long currentTime = System.currentTimeMillis();
        long count = offset.incrementAndGet();
        while(count >= MAX_COUNT_PER_MILLIS){
            synchronized (RequestIdGenerator.class){
                if(offset.get() >= MAX_COUNT_PER_MILLIS){
                    offset.set(0);
                }
            }
            count = offset.incrementAndGet();
        }
        return (currentTime << BITS) + count;
    }

    public static long getRequestIdFromClient() {
        // TODO 上下文 requestid
        return 0;

    }

}

2.限流，motan支持简单的限流，是利用filter来实现的

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29

@SpiMeta (name =  "active" ) @Activation (sequence =  1 ) public  class  ActiveLimitFilter  implements  Filter {        @Override      public  Response filter(Caller<?> caller, Request request) {          int  maxAcvitivyCount = caller.getUrl().getIntParameter(URLParamType.actives.getName(), URLParamType.actives.getIntValue());          if  (maxAcvitivyCount >  0 ) {              int  activeCount = RpcStats.getServiceStat(caller.getUrl()).getActiveCount();              if  (activeCount >= maxAcvitivyCount) {                  throw  new  MotanServiceException(String.format( "Request(%s) active count exceed the limit (%s), referer:%s" , request,                          maxAcvitivyCount, caller.getUrl()), MotanErrorMsgConstant.SERVICE_REJECT);              }          }            long  startTime = System.currentTimeMillis();          RpcStats.beforeCall(caller.getUrl(), request);          try  {              Response rs = caller.call(request);              RpcStats.afterCall(caller.getUrl(), request,  true , System.currentTimeMillis() - startTime);              return  rs;          }  catch  (RuntimeException re) {              RpcStats.afterCall(caller.getUrl(), request,  false , System.currentTimeMillis() - startTime);              throw  re;          }        }   }

3.对于连续失败的client进行不可用操作

    void incrErrorCount() {
        long count = errorCount.incrementAndGet();

        // 如果节点是可用状态，同时当前连续失败的次数超过限制maxClientConnection次，那么把该节点标示为不可用
        if (count >= maxClientConnection && state.isAliveState()) {
            synchronized (this) {
                count = errorCount.longValue();

                if (count >= maxClientConnection && state.isAliveState()) {
                    LoggerUtil.error("NettyClient unavailable Error: url=" + url.getIdentity() + " "
                            + url.getServerPortStr());
                    state = ChannelState.UNALIVE;
                }
            }
        }
    }

    void resetErrorCount() {
        errorCount.set(0);

        if (state.isAliveState()) {
            return;
        }

        synchronized (this) {
            if (state.isAliveState()) {
                return;
            }

            // 如果节点是unalive才进行设置，而如果是 close 或者 uninit，那么直接忽略
            if (state.isUnAliveState()) {
                long count = errorCount.longValue();

                // 过程中有其他并发更新errorCount的，因此这里需要进行一次判断
                if (count < maxClientConnection) {
                    state = ChannelState.ALIVE;
                    LoggerUtil.info("NettyClient recover available: url=" + url.getIdentity() + " "
                            + url.getServerPortStr());
                }
            }
        }
    }

4.支持多注册中心，因此cluster的refer集合是所有注册中心包含服务器的集合，如果同一个服务器在不同的注册中心注册，则cluster中当作两个服务器

5.服务端的采用boss线程池+工作线程池+业务线程池的处理方式

1 2 3 4 5 6 7 8 9 10 11 12 13

private  final  static  ChannelFactory channelFactory =  new  NioServerSocketChannelFactory( //boss线程池和工作线程池，主要负责接收消息          Executors.newCachedThreadPool( new  DefaultThreadFactory( "nettyServerBoss" ,  true )),          Executors.newCachedThreadPool( new  DefaultThreadFactory( "nettyServerWorker" ,  true )));        private  StandardThreadExecutor standardThreadExecutor =  null ; //业务线程池，负责具体的业务处理        standardThreadExecutor = (standardThreadExecutor !=  null  && !standardThreadExecutor.isShutdown()) ? standardThreadExecutor              :  new  StandardThreadExecutor(minWorkerThread, maxWorkerThread, workerQueueSize,                      new  DefaultThreadFactory( "NettyServer-"  + url.getServerPortStr(),  true ));      standardThreadExecutor.prestartAllCoreThreads();       final  NettyChannelHandler handler =  new  NettyChannelHandler(NettyServer. this , messageHandler,              standardThreadExecutor); //handler使用业务线程池今天处理具体的业务