Dubbo—Cluster剖析

概述

本文将介绍以下内容：

• 集群容错
• 集群容错模式
• 自定义扩展集群

集群容错

集群容错
在集群调用失败时，Dubbo 提供了多种容错方案，缺省为 failover 重试。

各节点关系：

• 这里的 Invoker 是 Provider 的一个可调用 Service 的抽象，Invoker 封装了 Provider 地址及 Service 接口信息
• Directory 代表多个 Invoker，可以把它看成 List ，但与 List 不同的是，它的值可能是动态变化的，比如注册中心推送变更
• Cluster 将 Directory 中的多个 Invoker 伪装成一个 Invoker，对上层透明，伪装过程包含了容错逻辑，调用失败后，重试另一个
• Router 负责从多个 Invoker 中按路由规则选出子集，比如读写分离，应用隔离等
• LoadBalance 负责从多个 Invoker 中选出具体的一个用于本次调用，选的过程包含了负载均衡算法，调用失败后，需要重选

集群容错模式

Cluster接口

先来看看Cluster接口

@SPI(FailoverCluster.NAME)
public interface Cluster {

    /**
     * Merge the directory invokers to a virtual invoker.
     *
     * @param 
     * @param directory
     * @return cluster invoker
     * @throws RpcException
     */
    @Adaptive
     Invoker join(Directory directory) throws RpcException;

}

dubbo集群实现同样采用Dubbo SPI方式（在前一篇Filter剖析文章中已介绍，FailoverCluster.NAME定义默认集群模式FailoverCluster。join方法是集群具体实现方案的方法接口。

Failover Cluster

失败自动切换，当出现失败，重试其它服务器 [1]。通常用于读操作，但重试会带来更长延迟。可通过 retries=“2” 来设置重试次数(不含第一次)。

重试次数配置如下：

或

或

核心代码如下：

public Result doInvoke(Invocation invocation, final List> invokers, LoadBalance loadbalance) throws RpcException {
        List> copyinvokers = invokers;
        checkInvokers(copyinvokers, invocation);
        String methodName = RpcUtils.getMethodName(invocation);
        int len = getUrl().getMethodParameter(methodName, Constants.RETRIES_KEY, Constants.DEFAULT_RETRIES) + 1;
        if (len <= 0) {
            len = 1;
        }
        // retry loop.
        RpcException le = null; // last exception.
        List> invoked = new ArrayList>(copyinvokers.size()); // invoked invokers.
        Set providers = new HashSet(len);
        for (int i = 0; i < len; i++) {
            //Reselect before retry to avoid a change of candidate `invokers`.
            //NOTE: if `invokers` changed, then `invoked` also lose accuracy.
            if (i > 0) {
                checkWhetherDestroyed();
                copyinvokers = list(invocation);
                // check again
                checkInvokers(copyinvokers, invocation);
            }
            Invoker invoker = select(loadbalance, invocation, copyinvokers, invoked);
            invoked.add(invoker);
            RpcContext.getContext().setInvokers((List) invoked);
            try {
                Result result = invoker.invoke(invocation);
                if (le != null && logger.isWarnEnabled()) {
                    logger.warn("Although retry the method " + methodName
                            + " in the service " + getInterface().getName()
                            + " was successful by the provider " + invoker.getUrl().getAddress()
                            + ", but there have been failed providers " + providers
                            + " (" + providers.size() + "/" + copyinvokers.size()
                            + ") from the registry " + directory.getUrl().getAddress()
                            + " on the consumer " + NetUtils.getLocalHost()
                            + " using the dubbo version " + Version.getVersion() + ". Last error is: "
                            + le.getMessage(), le);
                }
                return result;
            } catch (RpcException e) {
                if (e.isBiz()) { // biz exception.
                    throw e;
                }
                le = e;
            } catch (Throwable e) {
                le = new RpcException(e.getMessage(), e);
            } finally {
                providers.add(invoker.getUrl().getAddress());
            }
        }
        throw new RpcException(le.getCode(), "Failed to invoke the method "
                + methodName + " in the service " + getInterface().getName()
                + ". Tried " + len + " times of the providers " + providers
                + " (" + providers.size() + "/" + copyinvokers.size()
                + ") from the registry " + directory.getUrl().getAddress()
                + " on the consumer " + NetUtils.getLocalHost() + " using the dubbo version "
                + Version.getVersion() + ". Last error is: "
                + le.getMessage(), le.getCause() != null ? le.getCause() : le);
    }

Failfast Cluster

快速失败，只发起一次调用，失败立即报错。通常用于非幂等性的写操作，比如新增记录，核心代码实现如下：

public Result doInvoke(Invocation invocation, List> invokers, LoadBalance loadbalance) throws RpcException {
        checkInvokers(invokers, invocation);
        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;
            }
            throw new RpcException(e instanceof RpcException ? ((RpcException) e).getCode() : 0,
                    "Failfast invoke providers " + invoker.getUrl() + " " + loadbalance.getClass().getSimpleName()
                            + " select from all providers " + invokers + " for service " + getInterface().getName()
                            + " method " + invocation.getMethodName() + " on consumer " + NetUtils.getLocalHost()
                            + " use dubbo version " + Version.getVersion()
                            + ", but no luck to perform the invocation. Last error is: " + e.getMessage(),
                    e.getCause() != null ? e.getCause() : e);
        }
    }

Failsafe Cluster

失败安全，出现异常时，直接忽略。通常用于写入审计日志等操作。

public Result doInvoke(Invocation invocation, List> invokers, LoadBalance loadbalance) throws RpcException {
        try {
            checkInvokers(invokers, invocation);
            Invoker invoker = select(loadbalance, invocation, invokers, null);
            return invoker.invoke(invocation);
        } catch (Throwable e) {
            logger.error("Failsafe ignore exception: " + e.getMessage(), e);
            return new RpcResult(); // ignore
        }
    }

Failback Cluster
失败自动恢复，后台记录失败请求，定时重发。通常用于消息通知操作。
FailbackClusterInvoker中维护了一个ConcurrentMap类型的failed，key为Invocation，value为AbstractClusterInvoker。执行doInvoke方法时，若调用异常，则在捕获时将其put到failed中，然后启动定时任务遍历failed中所有失败调用并重试，若重试调用成功则将其从failed中剔除。

public class FailbackClusterInvoker extends AbstractClusterInvoker {

    private static final Logger logger = LoggerFactory.getLogger(FailbackClusterInvoker.class);

    private static final long RETRY_FAILED_PERIOD = 5 * 1000;

    /**
     * Use {@link NamedInternalThreadFactory} to produce {@link org.apache.dubbo.common.threadlocal.InternalThread}
     * which with the use of {@link org.apache.dubbo.common.threadlocal.InternalThreadLocal} in {@link RpcContext}.
     */
    private final ScheduledExecutorService scheduledExecutorService = Executors.newScheduledThreadPool(2,
            new NamedInternalThreadFactory("failback-cluster-timer", true));

    private final ConcurrentMap> failed = new ConcurrentHashMap<>();
    private volatile ScheduledFuture retryFuture;

    public FailbackClusterInvoker(Directory directory) {
        super(directory);
    }

    private void addFailed(Invocation invocation, AbstractClusterInvoker router) {
        if (retryFuture == null) {
            synchronized (this) {
                if (retryFuture == null) {
                    retryFuture = scheduledExecutorService.scheduleWithFixedDelay(new Runnable() {

                        @Override
                        public void run() {
                            // collect retry statistics
                            try {
                                retryFailed();
                            } catch (Throwable t) { // Defensive fault tolerance
                                logger.error("Unexpected error occur at collect statistic", t);
                            }
                        }
                    }, RETRY_FAILED_PERIOD, RETRY_FAILED_PERIOD, TimeUnit.MILLISECONDS);
                }
            }
        }
        failed.put(invocation, router);
    }

    void retryFailed() {
        if (failed.size() == 0) {
            return;
        }
        for (Map.Entry> entry : new HashMap<>(failed).entrySet()) {
            Invocation invocation = entry.getKey();
            Invoker invoker = entry.getValue();
            try {
                invoker.invoke(invocation);
                failed.remove(invocation);
            } catch (Throwable e) {
                logger.error("Failed retry to invoke method " + invocation.getMethodName() + ", waiting again.", e);
            }
        }
    }

    @Override
    protected Result doInvoke(Invocation invocation, List> invokers, LoadBalance loadbalance) throws RpcException {
        try {
            checkInvokers(invokers, invocation);
            Invoker invoker = select(loadbalance, invocation, invokers, null);
            return invoker.invoke(invocation);
        } catch (Throwable e) {
            logger.error("Failback to invoke method " + invocation.getMethodName() + ", wait for retry in background. Ignored exception: "
                    + e.getMessage() + ", ", e);
            addFailed(invocation, this);
            return new RpcResult(); // ignore
        }
    }

}

Forking Cluster
并行调用多个服务器，只要一个成功即返回。通常用于实时性要求较高的读操作，但需要浪费更多服务资源。可通过 forks=“2” 来设置最大并行数。

public Result doInvoke(final Invocation invocation, List> invokers, LoadBalance loadbalance) throws RpcException {
        try {
            checkInvokers(invokers, invocation);
            final List> selected;
            final int forks = getUrl().getParameter(Constants.FORKS_KEY, Constants.DEFAULT_FORKS);
            final int timeout = getUrl().getParameter(Constants.TIMEOUT_KEY, Constants.DEFAULT_TIMEOUT);
            if (forks <= 0 || forks >= invokers.size()) {
                selected = invokers;
            } else {
                selected = new ArrayList<>();
                for (int i = 0; i < forks; i++) {
                    // TODO. Add some comment here, refer chinese version for more details.
                    Invoker invoker = select(loadbalance, invocation, invokers, selected);
                    if (!selected.contains(invoker)) {
                        //Avoid add the same invoker several times.
                        selected.add(invoker);
                    }
                }
            }
            RpcContext.getContext().setInvokers((List) selected);
            final AtomicInteger count = new AtomicInteger();
            final BlockingQueue