[
https://issues.apache.org/jira/browse/HADOOP-15081?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16274433#comment-16274433
]
Jason Lowe commented on HADOOP-15081:
-------------------------------------
The JvmPauseMonitor does not call System.exit (or proxies to it like ExitUtil),
so I don't see how JvmPauseMonitor could be responsible for the process
exiting. Do you have evidence other than the log message appears towards the
end of the logs? What seems more likely is that there is memory pressure to
the point that out of memory errors are being thrown and caught by the
YarnUncaughtExceptionHandler. Check the stderr log to see if there's a message
about halting the system due to out of memory errors.
> org.apache.hadoop.util.JvmPauseMonitor Detected pause in JVM or host
> machine (eg GC) cause ResourceManager exit
> -----------------------------------------------------------------------------------------------------------------------
>
> Key: HADOOP-15081
> URL: https://issues.apache.org/jira/browse/HADOOP-15081
> Project: Hadoop Common
> Issue Type: Bug
> Components: common
> Affects Versions: 2.6.3
> Reporter: liuxiaobin
>
> org.apache.hadoop.util.JvmPauseMonitor
> Detected pause in JVM or host machine (eg GC): pause of approximately 2562ms
> GC pool 'ConcurrentMarkSweep' had collection(s): count=4 time=3040ms
> ResourceManager NodeManager exit .....
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
