hbase集群部分节点HRegionServer启动后自动关闭的问题

参考链接

http://f.dataguru.cn/thread-209058-1-1.html

我有４ＨRegionServer节点，1个master,其中３个是unbuntu 系统，２个节点是centos 6.5，

启动过程都很正常，但是一会后slave3 的HRegionServer会自动关闭.

查看tail -n100 hbase-hadoop-regionserver-Slave3.log日志如下:

-- ::, WARN  [regionserver/Slave3/192.168.2.38:] zookeeper.RecoverableZooKeeper: Possibly transient ZooKeeper, quorum=Master:,Slave1:,Slave2:,Slave3:,Slavrg.apache.zookeeper.KeeperException$OperationTimeoutException: KeeperErrorCode = OperationTimeout
-- ::, ERROR [regionserver/Slave3/192.168.2.38:] zookeeper.RecoverableZooKeeper: ZooKeeper delete failed after  attempts
-- ::, WARN  [regionserver/Slave3/192.168.2.38:] regionserver.HRegionServer: Failed deleting my ephemeral node
org.apache.zookeeper.KeeperException$OperationTimeoutException: KeeperErrorCode = OperationTimeout
        at org.apache.hadoop.hbase.zookeeper.RecoverableZooKeeper.checkZk(RecoverableZooKeeper.java:)
        at org.apache.hadoop.hbase.zookeeper.RecoverableZooKeeper.delete(RecoverableZooKeeper.java:)
        at org.apache.hadoop.hbase.zookeeper.ZKUtil.deleteNode(ZKUtil.java:)
        at org.apache.hadoop.hbase.zookeeper.ZKUtil.deleteNode(ZKUtil.java:)
        at org.apache.hadoop.hbase.regionserver.HRegionServer.deleteMyEphemeralNode(HRegionServer.java:)
        at org.apache.hadoop.hbase.regionserver.HRegionServer.run(HRegionServer.java:)
        at java.lang.Thread.run(Thread.java:)
-- ::, INFO  [regionserver/Slave3/192.168.2.38:] regionserver.HRegionServer: stopping server Slave3,,; zookeeper connection close

通过调整系统时间解决问题了.参考内容复制如下:

2、问题原因是时间不致造成的，解决方法如下：
1）在hbase-site.xml文件中 修改增加 ，将时间改大点
<property>
<name>hbase.master.maxclockskew</name>
<value>150000</value>
</property>
2）修改系统时间，将时间改为一致（建议采用本方法）：
修改日期
date -s 11/23/2013
修改时间
date -s 15:14:00
检查硬件（CMOS）时间
clock -r
将系统时间写入CMOS
clock -w

3、修改完成后单独启动HRegionServer节点即可：
启动集群中所有的regionserver
./hbase-daemons.sh start regionserver
启动某个regionserver
./hbase-daemon.sh start regionserver

其实最好关闭hbase和hadoop之后重启,才能浏览器http://192.168.2.35:16010/查看到结果.