当前位置: 首页>>代码示例>>Java>>正文


Java ZooKeeperListener类代码示例

本文整理汇总了Java中org.apache.hadoop.hbase.zookeeper.ZooKeeperListener的典型用法代码示例。如果您正苦于以下问题:Java ZooKeeperListener类的具体用法?Java ZooKeeperListener怎么用?Java ZooKeeperListener使用的例子?那么恭喜您, 这里精选的类代码示例或许可以为您提供帮助。


ZooKeeperListener类属于org.apache.hadoop.hbase.zookeeper包,在下文中一共展示了ZooKeeperListener类的3个代码示例,这些例子默认根据受欢迎程度排序。您可以为喜欢或者感觉有用的代码点赞,您的评价将有助于系统推荐出更棒的Java代码示例。

示例1: registerRegionServerListener

import org.apache.hadoop.hbase.zookeeper.ZooKeeperListener; //导入依赖的package包/类
public void registerRegionServerListener(ZooKeeperListener listener) {
  this.zookeeper.registerListener(listener);
}
 
开发者ID:fengchen8086,项目名称:LCIndex-HBase-0.94.16,代码行数:4,代码来源:ReplicationZookeeper.java

示例2: tryRecoveringExpiredZKSession

import org.apache.hadoop.hbase.zookeeper.ZooKeeperListener; //导入依赖的package包/类
/**
 * We do the following in a different thread.  If it is not completed
 * in time, we will time it out and assume it is not easy to recover.
 *
 * 1. Create a new ZK session. (since our current one is expired)
 * 2. Try to become a primary master again
 * 3. Initialize all ZK based system trackers.
 * 4. Assign root and meta. (they are already assigned, but we need to update our
 * internal memory state to reflect it)
 * 5. Process any RIT if any during the process of our recovery.
 *
 * @return True if we could successfully recover from ZK session expiry.
 * @throws InterruptedException
 * @throws IOException
 * @throws KeeperException
 * @throws ExecutionException
 */
private boolean tryRecoveringExpiredZKSession() throws InterruptedException,
    IOException, KeeperException, ExecutionException {

  this.zooKeeper.unregisterAllListeners();
  // add back listeners which were registered before master initialization
  // because they won't be added back in below Master re-initialization code
  if (this.registeredZKListenersBeforeRecovery != null) {
    for (ZooKeeperListener curListener : this.registeredZKListenersBeforeRecovery) {
      this.zooKeeper.registerListener(curListener);
    }
  }

  this.zooKeeper.reconnectAfterExpiration();

  Callable<Boolean> callable = new Callable<Boolean> () {
    public Boolean call() throws InterruptedException,
        IOException, KeeperException {
      MonitoredTask status =
        TaskMonitor.get().createStatus("Recovering expired ZK session");
      try {
        if (!becomeActiveMaster(status)) {
          return Boolean.FALSE;
        }
        serverManager.disableSSHForRoot();
        serverShutdownHandlerEnabled = false;
        initialized = false;
        finishInitialization(status, true);
        return Boolean.TRUE;
      } finally {
        status.cleanup();
      }
    }
  };

  long timeout =
    conf.getLong("hbase.master.zksession.recover.timeout", 300000);
  java.util.concurrent.ExecutorService executor =
    Executors.newSingleThreadExecutor();
  Future<Boolean> result = executor.submit(callable);
  executor.shutdown();
  if (executor.awaitTermination(timeout, TimeUnit.MILLISECONDS)
      && result.isDone()) {
    Boolean recovered = result.get();
    if (recovered != null) {
      return recovered.booleanValue();
    }
  }
  executor.shutdownNow();
  return false;
}
 
开发者ID:fengchen8086,项目名称:LCIndex-HBase-0.94.16,代码行数:68,代码来源:HMaster.java

示例3: tryRecoveringExpiredZKSession

import org.apache.hadoop.hbase.zookeeper.ZooKeeperListener; //导入依赖的package包/类
/**
 * We do the following in a different thread.  If it is not completed
 * in time, we will time it out and assume it is not easy to recover.
 *
 * 1. Create a new ZK session. (since our current one is expired)
 * 2. Try to become a primary master again
 * 3. Initialize all ZK based system trackers.
 * 4. Assign meta. (they are already assigned, but we need to update our
 * internal memory state to reflect it)
 * 5. Process any RIT if any during the process of our recovery.
 *
 * @return True if we could successfully recover from ZK session expiry.
 * @throws InterruptedException
 * @throws IOException
 * @throws KeeperException
 * @throws ExecutionException
 */
private boolean tryRecoveringExpiredZKSession() throws InterruptedException,
    IOException, KeeperException, ExecutionException {

  this.zooKeeper.unregisterAllListeners();
  // add back listeners which were registered before master initialization
  // because they won't be added back in below Master re-initialization code
  if (this.registeredZKListenersBeforeRecovery != null) {
    for (ZooKeeperListener curListener : this.registeredZKListenersBeforeRecovery) {
      this.zooKeeper.registerListener(curListener);
    }
  }

  this.zooKeeper.reconnectAfterExpiration();

  Callable<Boolean> callable = new Callable<Boolean> () {
    @Override
    public Boolean call() throws InterruptedException,
        IOException, KeeperException {
      MonitoredTask status =
        TaskMonitor.get().createStatus("Recovering expired ZK session");
      try {
        if (!becomeActiveMaster(status)) {
          return Boolean.FALSE;
        }
        serverShutdownHandlerEnabled = false;
        initialized = false;
        finishInitialization(status, true);
        return !stopped;
      } finally {
        status.cleanup();
      }
    }
  };

  long timeout =
    conf.getLong("hbase.master.zksession.recover.timeout", 300000);
  java.util.concurrent.ExecutorService executor =
    Executors.newSingleThreadExecutor();
  Future<Boolean> result = executor.submit(callable);
  executor.shutdown();
  if (executor.awaitTermination(timeout, TimeUnit.MILLISECONDS)
      && result.isDone()) {
    Boolean recovered = result.get();
    if (recovered != null) {
      return recovered.booleanValue();
    }
  }
  executor.shutdownNow();
  return false;
}
 
开发者ID:tenggyut,项目名称:HIndex,代码行数:68,代码来源:HMaster.java


注:本文中的org.apache.hadoop.hbase.zookeeper.ZooKeeperListener类示例由纯净天空整理自Github/MSDocs等开源代码及文档管理平台,相关代码片段筛选自各路编程大神贡献的开源项目,源码版权归原作者所有,传播和使用请参考对应项目的License;未经允许,请勿转载。