org.apache.hadoop.hbase.zookeeper.ZKUtil.deleteNode()方法的使用及代码示例

x33g5p2x  于2022-02-05 转载在 其他  
字(11.0k)|赞(0)|评价(0)|浏览(173)

本文整理了Java中org.apache.hadoop.hbase.zookeeper.ZKUtil.deleteNode()方法的一些代码示例，展示了ZKUtil.deleteNode()的具体用法。这些代码示例主要来源于Github/Stackoverflow/Maven等平台，是从一些精选项目中提取出来的代码，具有较强的参考意义，能在一定程度帮忙到你。ZKUtil.deleteNode()方法的具体详情如下：
包路径：org.apache.hadoop.hbase.zookeeper.ZKUtil
类名称：ZKUtil
方法名：deleteNode

ZKUtil.deleteNode介绍

[英]Delete the specified node. Sets no watches. Throws all exceptions.
[中]删除指定的节点。没有手表。抛出所有异常。

代码示例

代码示例来源：origin: apache/hbase

  1. /**
  2.  * Delete the specified node.  Sets no watches.  Throws all exceptions.
  3.  */
  4. public static void deleteNode(ZKWatcher zkw, String node)
  5.  throws KeeperException {
  6.  deleteNode(zkw, node, -1);
  7. }

代码示例来源：origin: apache/hbase

  1. /**
  2.  * Create a bunch of znodes in a hierarchy, try deleting one that has childs (it will fail), then
  3.  * delete it recursively, then delete the last znode
  4.  */
  5. @Test
  6. public void testZNodeDeletes() throws Exception {
  7.  ZKUtil.createWithParents(ZKW, "/l1/l2/l3/l4");
  8.  try {
  9.   ZKUtil.deleteNode(ZKW, "/l1/l2");
  10.   fail("We should not be able to delete if znode has childs");
  11.  } catch (KeeperException ex) {
  12.   assertNotNull(ZKUtil.getDataNoWatch(ZKW, "/l1/l2/l3/l4", null));
  13.  }
  14.  ZKUtil.deleteNodeRecursively(ZKW, "/l1/l2");
  15.  // make sure it really is deleted
  16.  assertNull(ZKUtil.getDataNoWatch(ZKW, "/l1/l2/l3/l4", null));
  17.  // do the same delete again and make sure it doesn't crash
  18.  ZKUtil.deleteNodeRecursively(ZKW, "/l1/l2");
  19.  ZKUtil.deleteNode(ZKW, "/l1");
  20.  assertNull(ZKUtil.getDataNoWatch(ZKW, "/l1/l2", null));
  21. }

代码示例来源：origin: apache/hbase

  1. private void deleteMyEphemeralNode() throws KeeperException {
  2.  ZKUtil.deleteNode(this.zooKeeper, getMyEphemeralNodePath());
  3. }

代码示例来源：origin: apache/hbase

  1. public static void deleteMetaLocation(ZKWatcher zookeeper, int replicaId)
  2.  throws KeeperException {
  3.  if (replicaId == RegionInfo.DEFAULT_REPLICA_ID) {
  4.   LOG.info("Deleting hbase:meta region location in ZooKeeper");
  5.  } else {
  6.   LOG.info("Deleting hbase:meta for " + replicaId + " region location in ZooKeeper");
  7.  }
  8.  try {
  9.   // Just delete the node.  Don't need any watches.
  10.   ZKUtil.deleteNode(zookeeper, zookeeper.getZNodePaths().getZNodeForReplica(replicaId));
  11.  } catch(KeeperException.NoNodeException nne) {
  12.   // Has already been deleted
  13.  }
  14. }
  15. /**

代码示例来源：origin: apache/hbase

  1. /**
  2.  * Sets the cluster as down by deleting the znode.
  3.  * @throws KeeperException unexpected zk exception
  4.  */
  5. public void setClusterDown()
  6.  throws KeeperException {
  7.  try {
  8.   ZKUtil.deleteNode(watcher, watcher.getZNodePaths().clusterStateZNode);
  9.  } catch(KeeperException.NoNodeException nne) {
  10.   LOG.warn("Attempted to set cluster as down but already down, cluster " +
  11.     "state node (" + watcher.getZNodePaths().clusterStateZNode + ") not found");
  12.  }
  13. }

代码示例来源：origin: apache/hbase

  1. private void cleanupHbckZnode() {
  2.  try {
  3.   if (zkw != null && hbckZodeCreated) {
  4.    ZKUtil.deleteNode(zkw, hbckEphemeralNodePath);
  5.    hbckZodeCreated = false;
  6.   }
  7.  } catch (KeeperException e) {
  8.   // Ignore
  9.   if (!e.code().equals(KeeperException.Code.NONODE)) {
  10.    LOG.warn("Delete HBCK znode " + hbckEphemeralNodePath + " failed ", e);
  11.   }
  12.  }
  13. }

代码示例来源：origin: apache/hbase

  1. @Override
  2. public void removeWAL(ServerName serverName, String queueId, String fileName)
  3.   throws ReplicationException {
  4.  String fileNode = getFileNode(serverName, queueId, fileName);
  5.  try {
  6.   ZKUtil.deleteNode(zookeeper, fileNode);
  7.  } catch (NoNodeException e) {
  8.   LOG.warn("{} already deleted when removing log", fileNode);
  9.  } catch (KeeperException e) {
  10.   throw new ReplicationException("Failed to remove wal from queue (serverName=" + serverName +
  11.    ", queueId=" + queueId + ", fileName=" + fileName + ")", e);
  12.  }
  13. }

代码示例来源：origin: apache/hbase

  1. if (znode.endsWith(suffix)) {
  2.  sb.append(ZNodePaths.ZNODE_PATH_SEPARATOR).append(znode);
  3.  ZKUtil.deleteNode(zookeeper, sb.toString());
  4.  sb.setLength(levelTwoLength);

代码示例来源：origin: apache/hbase

  1. /***
  2.   * Delete the acl notify node of namespace
  3.   */
  4.  public void deleteNamespaceACLNode(final String namespace) {
  5.   String zkNode = ZNodePaths.joinZNode(watcher.getZNodePaths().baseZNode, ACL_NODE);
  6.   zkNode = ZNodePaths.joinZNode(zkNode, AccessControlLists.NAMESPACE_PREFIX + namespace);
  8.   try {
  9.    ZKUtil.deleteNode(watcher, zkNode);
  10.   } catch (KeeperException.NoNodeException e) {
  11.    LOG.warn("No acl notify node of namespace '" + namespace + "'");
  12.   } catch (KeeperException e) {
  13.    LOG.error("Failed deleting acl node of namespace '" + namespace + "'", e);
  14.    watcher.abort("Failed deleting node " + zkNode, e);
  15.   }
  16.  }
  17. }

代码示例来源：origin: apache/hbase

  1. /***
  2.  * Delete the acl notify node of table
  3.  * @param tableName
  4.  */
  5. public void deleteTableACLNode(final TableName tableName) {
  6.  String zkNode = ZNodePaths.joinZNode(watcher.getZNodePaths().baseZNode, ACL_NODE);
  7.  zkNode = ZNodePaths.joinZNode(zkNode, tableName.getNameAsString());
  8.  try {
  9.   ZKUtil.deleteNode(watcher, zkNode);
  10.  } catch (KeeperException.NoNodeException e) {
  11.   LOG.warn("No acl notify node of table '" + tableName + "'");
  12.  } catch (KeeperException e) {
  13.   LOG.error("Failed deleting acl node of table '" + tableName + "'", e);
  14.   watcher.abort("Failed deleting node " + zkNode, e);
  15.  }
  16. }

代码示例来源：origin: apache/hbase

  1. public void removeKeyFromZK(AuthenticationKey key) {
  2.  String keyZNode = getKeyNode(key.getKeyId());
  3.  try {
  4.   ZKUtil.deleteNode(watcher, keyZNode);
  5.  } catch (KeeperException.NoNodeException nne) {
  6.   LOG.error("Non-existent znode "+keyZNode+" for key "+key.getKeyId(), nne);
  7.  } catch (KeeperException ke) {
  8.   LOG.error(HBaseMarkers.FATAL, "Failed removing znode "+keyZNode+" for key "+
  9.     key.getKeyId(), ke);
  10.   watcher.abort("Unhandled zookeeper error removing znode "+keyZNode+
  11.     " for key "+key.getKeyId(), ke);
  12.  }
  13. }

代码示例来源：origin: apache/hbase

  1. /**
  2.   * delete the master znode if its content is same as the parameter
  3.   * @param zkw must not be null
  4.   * @param content must not be null
  5.   */
  6.  public static boolean deleteIfEquals(ZKWatcher zkw, final String content) {
  7.   if (content == null){
  8.    throw new IllegalArgumentException("Content must not be null");
  9.   }
  11.   try {
  12.    Stat stat = new Stat();
  13.    byte[] data = ZKUtil.getDataNoWatch(zkw, zkw.getZNodePaths().masterAddressZNode, stat);
  14.    ServerName sn = ProtobufUtil.parseServerNameFrom(data);
  15.    if (sn != null && content.equals(sn.toString())) {
  16.     return (ZKUtil.deleteNode(zkw, zkw.getZNodePaths().masterAddressZNode, stat.getVersion()));
  17.    }
  18.   } catch (KeeperException e) {
  19.    LOG.warn("Can't get or delete the master znode", e);
  20.   } catch (DeserializationException e) {
  21.    LOG.warn("Can't get or delete the master znode", e);
  22.   }
  24.   return false;
  25.  }
  26. }

代码示例来源：origin: apache/hbase

  1. public void stop() {
  2.   try {
  3.    synchronized (clusterHasActiveMaster) {
  4.     // Master is already stopped, wake up the manager
  5.     // thread so that it can shutdown soon.
  6.     clusterHasActiveMaster.notifyAll();
  7.    }
  8.    // If our address is in ZK, delete it on our way out
  9.    ServerName activeMaster = null;
  10.    try {
  11.     activeMaster = MasterAddressTracker.getMasterAddress(this.watcher);
  12.    } catch (IOException e) {
  13.     LOG.warn("Failed get of master address: " + e.toString());
  14.    }
  15.    if (activeMaster != null &&  activeMaster.equals(this.sn)) {
  16.     ZKUtil.deleteNode(watcher, watcher.getZNodePaths().masterAddressZNode);
  17.     // We may have failed to delete the znode at the previous step, but
  18.     //  we delete the file anyway: a second attempt to delete the znode is likely to fail again.
  19.     ZNodeClearer.deleteMyEphemeralNodeOnDisk();
  20.    }
  21.   } catch (KeeperException e) {
  22.    LOG.debug(this.watcher.prefix("Failed delete of our master address node; " +
  23.      e.getMessage()));
  24.   }
  25.  }
  26. }

代码示例来源：origin: apache/hbase

  1. private void unassignMetaReplica(HbckInfo hi) throws IOException, InterruptedException,
  2. KeeperException {
  3.  undeployRegions(hi);
  4.  ZKUtil.deleteNode(zkw, zkw.getZNodePaths().getZNodeForReplica(hi.metaEntry.getReplicaId()));
  5. }

代码示例来源：origin: apache/hbase

  1. private void unassignExcessMetaReplica(int numMetaReplicasConfigured) {
  2.   final ZKWatcher zooKeeper = master.getZooKeeper();
  3.   // unassign the unneeded replicas (for e.g., if the previous master was configured
  4.   // with a replication of 3 and now it is 2, we need to unassign the 1 unneeded replica)
  5.   try {
  6.    List<String> metaReplicaZnodes = zooKeeper.getMetaReplicaNodes();
  7.    for (String metaReplicaZnode : metaReplicaZnodes) {
  8.     int replicaId = zooKeeper.getZNodePaths().getMetaReplicaIdFromZnode(metaReplicaZnode);
  9.     if (replicaId >= numMetaReplicasConfigured) {
  10.      RegionState r = MetaTableLocator.getMetaRegionState(zooKeeper, replicaId);
  11.      LOG.info("Closing excess replica of meta region " + r.getRegion());
  12.      // send a close and wait for a max of 30 seconds
  13.      ServerManager.closeRegionSilentlyAndWait(master.getClusterConnection(),
  14.        r.getServerName(), r.getRegion(), 30000);
  15.      ZKUtil.deleteNode(zooKeeper, zooKeeper.getZNodePaths().getZNodeForReplica(replicaId));
  16.     }
  17.    }
  18.   } catch (Exception ex) {
  19.    // ignore the exception since we don't want the master to be wedged due to potential
  20.    // issues in the cleanup of the extra regions. We can do that cleanup via hbck or manually
  21.    LOG.warn("Ignoring exception " + ex);
  22.   }
  23.  }
  24. }

代码示例来源：origin: apache/hbase

  1. @Test
  2. public void testRegionServerRemovedEvent() throws Exception {
  3.  ZKUtil.createAndWatch(zkw,
  4.   ZNodePaths.joinZNode(zkw.getZNodePaths().rsZNode, "hostname2.example.org:1234"),
  5.   HConstants.EMPTY_BYTE_ARRAY);
  6.  rt.registerListener(new DummyReplicationListener());
  7.  // delete one
  8.  ZKUtil.deleteNode(zkw,
  9.   ZNodePaths.joinZNode(zkw.getZNodePaths().rsZNode, "hostname2.example.org:1234"));
  10.  // wait for event
  11.  while (rsRemovedCount.get() < 1) {
  12.   Thread.sleep(5);
  13.  }
  14.  assertEquals("hostname2.example.org:1234", rsRemovedData);
  15. }

代码示例来源：origin: apache/hbase

  1. @Test
  2. public void testGetListOfRegionServers() throws Exception {
  3.  // 0 region servers
  4.  assertEquals(0, rt.getListOfRegionServers().size());
  5.  // 1 region server
  6.  ZKUtil.createWithParents(zkw,
  7.   ZNodePaths.joinZNode(zkw.getZNodePaths().rsZNode, "hostname1.example.org:1234"));
  8.  assertEquals(1, rt.getListOfRegionServers().size());
  9.  // 2 region servers
  10.  ZKUtil.createWithParents(zkw,
  11.   ZNodePaths.joinZNode(zkw.getZNodePaths().rsZNode, "hostname2.example.org:1234"));
  12.  assertEquals(2, rt.getListOfRegionServers().size());
  13.  // 1 region server
  14.  ZKUtil.deleteNode(zkw,
  15.   ZNodePaths.joinZNode(zkw.getZNodePaths().rsZNode, "hostname2.example.org:1234"));
  16.  assertEquals(1, rt.getListOfRegionServers().size());
  17.  // 0 region server
  18.  ZKUtil.deleteNode(zkw,
  19.   ZNodePaths.joinZNode(zkw.getZNodePaths().rsZNode, "hostname1.example.org:1234"));
  20.  assertEquals(0, rt.getListOfRegionServers().size());
  21. }

代码示例来源：origin: apache/hbase

  1. "testActiveMasterManagerFromZK", null, true);
  2. try {
  3.  ZKUtil.deleteNode(zk, zk.getZNodePaths().masterAddressZNode);
  4.  ZKUtil.deleteNode(zk, zk.getZNodePaths().clusterStateZNode);
  5. } catch(KeeperException.NoNodeException nne) {}

代码示例来源：origin: apache/hbase

  1. ZKUtil.deleteNode(UTIL.getZooKeeperWatcher(), STORAGE.getSyncReplicationStateNode(peerId));
  2. ZKUtil.deleteNode(UTIL.getZooKeeperWatcher(), STORAGE.getNewSyncReplicationStateNode(peerId));

代码示例来源：origin: apache/hbase

  1. @Ignore @Test // The close silently doesn't work any more since HBASE-14614. Fix.
  2. public void testHBaseFsckWithFewerMetaReplicaZnodes() throws Exception {
  3.  ClusterConnection c = (ClusterConnection)ConnectionFactory.createConnection(
  4.    TEST_UTIL.getConfiguration());
  5.  RegionLocations rl = c.locateRegion(TableName.META_TABLE_NAME, HConstants.EMPTY_START_ROW,
  6.    false, false);
  7.  HBaseFsckRepair.closeRegionSilentlyAndWait(c,
  8.    rl.getRegionLocation(2).getServerName(), rl.getRegionLocation(2).getRegionInfo());
  9.  ZKWatcher zkw = TEST_UTIL.getZooKeeperWatcher();
  10.  ZKUtil.deleteNode(zkw, zkw.getZNodePaths().getZNodeForReplica(2));
  11.  // check that problem exists
  12.  HBaseFsck hbck = doFsck(TEST_UTIL.getConfiguration(), false);
  13.  assertErrors(hbck, new ERROR_CODE[]{ERROR_CODE.UNKNOWN,ERROR_CODE.NO_META_REGION});
  14.  // fix the problem
  15.  hbck = doFsck(TEST_UTIL.getConfiguration(), true);
  16.  // run hbck again to make sure we don't see any errors
  17.  hbck = doFsck(TEST_UTIL.getConfiguration(), false);
  18.  assertErrors(hbck, new ERROR_CODE[]{});
  19. }

内容来源于网络，如有侵权，请联系作者删除！

ZKUtil

相关文章

    查看更多

    热门标签

    更多
    JavaquerypythonNode开发语言requestUtil数据库Table后端算法LoggerMessageElementParser

    最新文章

    更多

    ZKUtil类方法

    • 蜀ICP备13028337号-1 大数据知识库 https://www.saoniuhuo.com © All rights reserved
    • 本站内容来源互联网,如果侵犯您的权益请联系我们删除, 联系方式：448109455@qq.com