避免regionServer宕机

因为regionserver 的管理信息主要记录在zookeeper，regionserver的宕机判断依据是session expired。ok

那么regionserver 和Zookeeper的session expired原因有哪些尼？

1. 网络不好。

2. Java full GC， 这会block所有的线程。如果时间比较长，也会导致session expired.

解决办法：

1. 将Zookeeper的timeout时间加长。

2. 配置“hbase.regionserver.restart.on.zk.expire” 为true。 这样子，遇到ZooKeeper session expired ， regionserver将选择 restart 而不是 abort

具体的配置是，在hbase-site.xml中加入

<property>

<name>zookeeper.session.timeout</name>

<value>90000</value>

<description>ZooKeeper session timeout.

HBase passes this to the zk quorum as suggested maximum time for a

session.  See http://hadoop.apache.org/zookeeper/docs/current/zookeeperProgrammers.html#ch_zkSessions

“The client sends a requested timeout, the server responds with the

timeout that it can give the client. The current implementation

requires that the timeout be a minimum of 2 times the tickTime

(as set in the server configuration) and a maximum of 20 times

the tickTime.” Set the zk ticktime with hbase.zookeeper.property.tickTime.

In milliseconds.

</description>

</property>

<property>

<name>hbase.regionserver.restart.on.zk.expire</name>

<value>true</value>

<description>

Zookeeper session expired will force regionserver exit.

Enable this will make the regionserver restart.

</description>

</property>

3、为了避免java full GC suspend thread 对Zookeeper heartbeat的影响，我们还需要对hbase-env.sh进行配置。

  设置jvm的内存回收算法：

     -XX:+CMSParallelRemarkEnabled。

如下所示：

export HBASE_OPTS="-Xms16g -Xmx16g -Xmn2g -Xss200k -XX:MaxNewSize=2g -XX:SurvivorRatio=2 -XX:+UseCMSInitiatingOccupancyOnly -XX:CMSInitiatingOccupancyFraction=70 -XX:+UseConcMarkSweepGC -XX:+DisableExplicitGC  -XX:+CMSParallelRemarkEnabled   -XX:+UseFastAccessorMethods  -XX:+UseParNewGC -XX:MaxPermSize=300m -XX:MaxTenuringThreshold=5  -XX:GCTimeRatio=19 -XX:ParallelGCThreads=10 -XX:+UseCMSCompactAtFullCollection -XX:CMSFullGCsBeforeCompaction=0 -XX:-UseGCOverheadLimit "

最后启动  regionserver：
命令：hbase-daemon.sh start regionserver

开启balance命令：balance_switch true