name node process is exiting with Detected pause in JVM or host machine

[#1]

Name node process is crashing few moments after restarting with the following exception

2017-03-07 11:54:15,997 INFO org.apache.hadoop.util.JvmPauseMonitor: Detected pause in JVM or host machine (eg GC): pause of approximately 1013ms
GC pool 'ConcurrentMarkSweep' had collection(s): count=1 time=1445ms
2017-03-07 11:54:17,877 INFO org.apache.hadoop.util.JvmPauseMonitor: Detected pause in JVM or host machine (eg GC): pause of approximately 1229ms
GC pool 'ConcurrentMarkSweep' had collection(s): count=1 time=1409ms
2017-03-07 11:54:19,913 INFO org.apache.hadoop.util.JvmPauseMonitor: Detected pause in JVM or host machine (eg GC): pause of approximately 1535ms
GC pool 'ConcurrentMarkSweep' had collection(s): count=1 time=1870ms
2017-03-07 11:54:23,297 INFO org.apache.hadoop.util.JvmPauseMonitor: Detected pause in JVM or host machine (eg GC): pause of approximately 1266ms
GC pool 'ConcurrentMarkSweep' had collection(s): count=1 time=1658ms
2017-03-07 11:54:26,528 INFO org.apache.hadoop.util.JvmPauseMonitor: Detected pause in JVM or host machine (eg GC): pause of approximately 1382ms
GC pool 'ConcurrentMarkSweep' had collection(s): count=1 time=1861ms
2017-03-07 11:54:29,996 INFO org.apache.hadoop.util.JvmPauseMonitor: Detected pause in JVM or host machine (eg GC): pause of approximately 1345ms
GC pool 'ConcurrentMarkSweep' had collection(s): count=1 time=1837ms
2017-03-07 11:54:34,016 INFO org.apache.hadoop.util.JvmPauseMonitor: Detected pause in JVM or host machine (eg GC): pause of approximately 1142ms
GC pool 'ConcurrentMarkSweep' had collection(s): count=2 time=1640ms 

A few days back, the machine got shut down forcefully due to power failure. getting this issue after that. Please suggest or let me know if any further details are required.

[#2 · Accepted solution]

Worked after increaing the Java Heap size of name node

Name node process was exiting while loading the inodes. Since my last restart, I had created a lot of files on hdfs which lead to significant increase in inodes. But since my java heap size was not comaprable to that so JVM was pausing.