[
https://issues.apache.org/jira/browse/CASSANDRA-4387?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13402686#comment-13402686
]
Jonathan Ellis commented on CASSANDRA-4387:
-------------------------------------------
Odd. Sounds like a problem with SynchronousQueue internals to me. Only thing
I could suggest trying is replacing it (in AbstractCassandraDaemon) with the
java7 TransferQueue, which is supposed to be significantly more performant.
http://docs.oracle.com/javase/7/docs/api/java/util/concurrent/TransferQueue.html
> Multiple RPC threads waiting on a single thread
> -----------------------------------------------
>
> Key: CASSANDRA-4387
> URL: https://issues.apache.org/jira/browse/CASSANDRA-4387
> Project: Cassandra
> Issue Type: Bug
> Affects Versions: 1.0.7
> Environment: sun java 6.24
> Reporter: Jason Harvey
> Priority: Minor
> Attachments: output.log
>
>
> I'm investigating a load spike issue on our cluster. A thread dump that I
> collected during a spike shows a peculiar symptom. There are 162 RPC threads
> all parked waiting on a single
> java.util.concurrent.SynchronousQueue$TransferStack thread. Brandon took a
> peek and recommended I create a ticket for investigation.
> Full thread dump attached. The thread which is being waited on is
> 0x00000006ac1aba60.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
