hbase hregion server 异常退出排查

---HBASE日志
2017-10-25 23:01:15,788 INFO  [JvmPauseMonitor] util.JvmPauseMonitor: Detected pause in JVM or host machine (eg GC): pause of approximately 2907ms
GC pool 'ParNew' had collection(s): count=1 time=3312ms
2017-10-25 23:01:29,859 INFO  [JvmPauseMonitor] util.JvmPauseMonitor: Detected pause in JVM or host machine (eg GC): pause of approximately 4569ms
GC pool 'ParNew' had collection(s): count=1 time=4746ms
2017-10-25 23:01:37,146 INFO  [JvmPauseMonitor] util.JvmPauseMonitor: Detected pause in JVM or host machine (eg GC): pause of approximately 2285ms
GC pool 'ParNew' had collection(s): count=1 time=2530ms
2017-10-25 23:01:44,387 INFO  [JvmPauseMonitor] util.JvmPauseMonitor: Detected pause in JVM or host machine (eg GC): pause of approximately 5240ms
GC pool 'ParNew' had collection(s): count=1 time=5417ms

2017-10-25 23:02:06,587 WARN  [regionserver60020] util.Sleeper: We slept 22194ms instead of 3000ms, this is likely due to a long garbage collecting pause and it's usually bad, see http://hbase.apache.org/book.html#trouble.rs.runtime.zkexpired
2017-10-25 23:02:06,588 WARN  [JvmPauseMonitor] util.JvmPauseMonitor: Detected pause in JVM or host machine (eg GC): pause of approximately 20200ms
GC pool 'ParNew' had collection(s): count=1 time=20520ms



2017-10-25 23:02:06,587 WARN  [regionserver60020] util.Sleeper: We slept 22194ms instead of 3000ms, this is likely due to a long garbage collecting pause and it's usually bad, see http://hbase.apache.org/book.html#trouble.rs.runtime.zkexpired
2017-10-25 23:02:06,588 WARN  [JvmPauseMonitor] util.JvmPauseMonitor: Detected pause in JVM or host machine (eg GC): pause of approximately 20200ms
GC pool 'ParNew