dubbo源码阅读-Filter默认实现(十一)之ActiveLimitFilter&ExecuteLimitFilter

文章主目录

• 文档
• ActiveLimitFilter
• ExecuteLimitFilter

回到顶部

文档

http://dubbo.apache.org/zh-cn/docs/user/demos/concurrency-control.html

回到顶部

ActiveLimitFilter

/**
 * 控制调用服务的并发量 限流
 * 同时支持多少请求 consumer有效 含有actives配置
 */
@Activate(
        group = {"consumer"},
        value = {"actives"}
)
public class ActiveLimitFilter implements Filter {
    public ActiveLimitFilter() {
    }

    public Result invoke(Invoker<?> invoker, Invocation invocation) throws RpcException {
        URL url = invoker.getUrl();
        String methodName = invocation.getMethodName();
        //获得 <dubbo:reference actives="1"> actives的数量
        int max = invoker.getUrl().getMethodParameter(methodName, "actives", 0);
        //获取当前service当前方法的请求数量
        RpcStatus count = RpcStatus.getStatus(invoker.getUrl(), invocation.getMethodName());
        long timeout;
        //配置并发控制大于0才写
        if (max > 0) {
            //获得当前方法的等待时间
            timeout = (long)invoker.getUrl().getMethodParameter(invocation.getMethodName(), "timeout", 0);
            long start = System.currentTimeMillis();
            long remain = timeout;
            //判断是否大于并发数 如果大于则等待
            int active = count.getActive();
            if (active >= max) {
                synchronized(count) {
                    /**
                     *1.while循环是有必要的
                     *  当收到其他线程notify 获得执行权
                     *  但是这个时候其他线程提前进入(active >= max)  判断为false获得执行权 count+1 
                     *  这个时候 还需要while判断是否还有空闲请求 否则继续wait
                     *
                     */
                    while((active = count.getActive()) >= max) {
                        try {
                            //超时时间为 配置的超时时间
                            count.wait(remain);
                        } catch (InterruptedException var32) {
                            ;
                        }

                        long elapsed = System.currentTimeMillis() - start;
                        remain = timeout - elapsed;
                         //当其他线程通知等待线程执行 判断是否超时 如果超时了则不执行了
                        if (remain <= 0L) {
                            throw new RpcException("Waiting concurrent invoke timeout in client-side for service:  " + invoker.getInterface().getName() + ", method: " + invocation.getMethodName() + ", elapsed: " + elapsed + ", timeout: " + timeout + ". concurrent invokes: " + active + ". max concurrent invoke limit: " + max);
                        }
                    }
                }
            }
        }

        boolean var28 = false;

        Result var10;
        try {
            var28 = true;
            timeout = System.currentTimeMillis();
            //获得执行权的 count+1
            RpcStatus.beginCount(url, methodName);

            try {
                //执行
                Result result = invoker.invoke(invocation);
                //执行完毕关闭
                RpcStatus.endCount(url, methodName, System.currentTimeMillis() - timeout, true);
                var10 = result;
                var28 = false;
            } catch (RuntimeException var31) {
                RpcStatus.endCount(url, methodName, System.currentTimeMillis() - timeout, false);
                throw var31;
            }
        } finally {
            if (var28) {
                if (max > 0) {
                    //通知等待的线程执行
                    synchronized(count) {
                        count.notify();
                    }
                }

            }
        }

        if (max > 0) {
            synchronized(count) {
                //通知等待的线程执行
                count.notify();
            }
        }

        return var10;
    }
}

回到顶部

ExecuteLimitFilter

**
 * ThreadLimitInvokerFilter
 * 只针对provider有效 含有executes配置
 */
@Activate(group = Constants.PROVIDER, value = Constants.EXECUTES_KEY)
public class ExecuteLimitFilter implements Filter {

    @Override
    public Result invoke(Invoker<?> invoker, Invocation invocation) throws RpcException {
        URL url = invoker.getUrl();
        String methodName = invocation.getMethodName();
        Semaphore executesLimit = null;
        boolean acquireResult = false;
        //获取服务提供者的最大访问并发数配置
        int max = url.getMethodParameter(methodName, Constants.EXECUTES_KEY, 0);
        if (max > 0) {
            //<1> 获得 RpcStatus 对象，基于服务 URL + 方法维度
            RpcStatus count = RpcStatus.getStatus(url, invocation.getMethodName());
//            if (count.getActive() >= max) {
            /**
             * http://manzhizhen.iteye.com/blog/2386408
             * use semaphore for concurrency control (to limit thread number)
             */
            //获得信号量
            executesLimit = count.getSemaphore(max);
            //    //如果达到限流条件直接报错
            if(executesLimit != null && !(acquireResult = executesLimit.tryAcquire())) {
                throw new RpcException("Failed to invoke method " + invocation.getMethodName() + " in provider " + url + ", cause: The service using threads greater than <dubbo:service executes=\"" + max + "\" /> limited.");
            }
        }
        long begin = System.currentTimeMillis();
        boolean isSuccess = true;
        //<2>调用计数+1
        RpcStatus.beginCount(url, methodName);
        try {
            Result result = invoker.invoke(invocation);
            return result;
        } catch (Throwable t) {
            isSuccess = false;
            if (t instanceof RuntimeException) {
                throw (RuntimeException) t;
            } else {
                throw new RpcException("unexpected exception when ExecuteLimitFilter", t);
            }
        } finally {
            //<3>调用结束的计数-1
            RpcStatus.endCount(url, methodName, System.currentTimeMillis() - begin, isSuccess);
            //<4>释放信号量
            if(acquireResult) {
                executesLimit.release();
            }
        }
    }

}