[
https://issues.apache.org/jira/browse/CASSANDRA-5716?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13698744#comment-13698744
]
Ignace Desimpel commented on CASSANDRA-5716:
--------------------------------------------
Sorry, I forgot the fact that the thread is dead when the uncaught exception
handling is called and thus was thinking that thread would continue to run
until the system.exit would stop all the threads.
> Remark on cassandra-5273 : Hanging system after OutOfMemory. Server cannot
> die due to uncaughtException handling
> ----------------------------------------------------------------------------------------------------------------
>
> Key: CASSANDRA-5716
> URL: https://issues.apache.org/jira/browse/CASSANDRA-5716
> Project: Cassandra
> Issue Type: Bug
> Components: Core
> Affects Versions: 1.2.6
> Environment: linux
> Reporter: Ignace Desimpel
> Assignee: Jonathan Ellis
> Priority: Minor
> Fix For: 1.2.7
>
>
> Possible incorrect handling of an OOM as a result of modifications made for
> issue cassandra-5273.
> I could reproduce the OOM, with the patch of Cassandra-5273 applied.
> The good news is that, at least in my case, it works fine : the system did
> die !
>
> However, due to multiple uncaughtException handling, multiple threads are
> calling the exitThread.start() routine, causing an IllegalStateException.
> There are some other exceptions also, but that seems logical. Also, after
> calling the start() function, the thread(s) will continue to run, and that
> could not be wanted.
>
> Below I pasted the stack trace.
> Just for your information, after all this works, and I could restart the
> Cassandra server and redo the OOM
> [stack trace moved to
> http://aep.appspot.com/display/mQFNFHUh1VvQJYGcxRK0lQSM2j8/ ]
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
