Question

我目前正在从事使用akka 2.4的项目。该项目是使用akka DistributedPubSub介体编写的。在初始测试阶段，一切似乎都工作正常，但是随着我们现在的规模扩展，我们经常看到Distributed PubSub中介程序发布和订阅服务中的内存消耗非常大。发布服务会很快填满内存，并停止处理更多消息。

发布服务的application.conf是：

akka {

  loggers = ["de.heikoseeberger.akkalog4j.Log4jLogger"]
  logging-filter = "de.heikoseeberger.akkalog4j.Log4jLoggingFilter"
  loglevel = debug
  log-dead-letters = off


  actor {
    provider = "akka.cluster.ClusterActorRefProvider"
     serializers {
    java = "akka.serialization.JavaSerializer"
    }
    serialization-bindings {
    "java.lang.String" = java
    }
  }
  remote {
    log-remote-lifecycle-events = off
    netty.tcp {
      hostname = "<IP>"
      port =2551
    }
  }

  cluster {
    seed-nodes = [
      "akka.tcp://ClusterSystem@<InternalIP1>:2551",
      "akka.tcp://ClusterSystem@<InternalIP2>:2551"]

    # auto downing is NOT safe for production deployments.
    # you may want to use it during development, read more about it in the docs.
    #
     auto-down-unreachable-after = 10s
  pub-sub {
       removed-time-to-live = 120s
        }
 }
}
akka.cluster.metrics.enabled=true
akka.extensions=["akka.cluster.pubsub.DistributedPubSub","akka.cluster.client.ClusterClientReceptionist"]

发布者代码如下：

 @Override
  public void onReceive(Object message) throws Exception {
    if (message instanceof InitializeLogger) {
      getSender().tell(Logging.loggerInitialized(), getSelf());
    } else if (message instanceof MemberUp) {
      MemberUp mUp = (MemberUp) message;
      LOGGER.trace("Member is Up: {}", mUp.member());

    } else if (message instanceof UnreachableMember) {
      UnreachableMember mUnreachable = (UnreachableMember) message;
      LOGGER.trace("Member detected as unreachable: {}", mUnreachable.member());

    } else if (message instanceof MemberRemoved) {
      MemberRemoved mRemoved = (MemberRemoved) message;
      LOGGER.trace("Member is Removed: {}", mRemoved.member());

    } else if (message instanceof MemberEvent) {
      // ignore

    } else if (message instanceof Message) {

      LOGGER.info("Obtained a message in message type{}", message);
      // counter.increment();
      mediator.tell(new DistributedPubSubMediator.Publish(TOPIC_NAME, message), getSelf());

    } else
      unhandled(message);

  }

问题1：因为我们使用的是默认邮箱（即无限制队列），所以邮箱消息在被订阅者角色消费后是否可能没有被清除？。如果是这样，我该如何清除它们。我以为“删除时间”将清除这些消息，但我认为我的理解是错误的。

在运行在单独节点上的订户端，我有一个参与者订阅了主题并使用消息。该演员充当委派者，并对另一个本地演员执行“讲述”。订户应用程序conf如下：

akka {

  loglevel = "INFO"
  log-dead-letters = off
    log-dead-letters-during-shutdown = off

  actor {
    provider = "akka.cluster.ClusterActorRefProvider"
  }
  remote {
    log-remote-lifecycle-events = off
    netty.tcp {
      hostname = "InternalIP1"
      port = 2551
    }
  }

  cluster {
    seed-nodes = [
      "akka.tcp://ClusterSystem@<InternalIP1>:2551"
      "akka.tcp://ClusterSystem@<InternalIP2>:2551"
    ]

    auto-down-unreachable-after = 120s

  }
      blocking-io-dispatcher = "akka.stream.default-blocking-io-dispatcher"
      akka.extensions=["akka.cluster.client.ClusterClientReceptionist"]

    default-blocking-io-dispatcher {
      type = "Dispatcher"
      executor = "thread-pool-executor"
      throughput = 1

      thread-pool-executor {
        core-pool-size-min = 2
        core-pool-size-factor = 2.0
        core-pool-size-max = 16
      }
    }
    akka.remote.transport-failure-detector.acceptable-heartbeat-pause = 10s
}

问题2：在执行本地“告诉”操作时，我发现在处理消息方面存在巨大的时间滞后（接近7分钟）。我认为这是因为加载了订户actor的邮箱（我们也为订户使用了默认邮箱，这是一个无界队列。）。我们看到订户也慢慢地开始消耗内存。邮件耗尽后，如何确保清除邮箱。

现在，我们已经开始重新启动服务，这是清除邮箱的一种非常粗糙且不正确的方法。为了避免此内存泄漏，我在配置中缺少什么。

如果有人可以就此问题提供帮助，我们将不胜感激。

akka邮箱占用内存

0 个答案: