[
https://issues.apache.org/jira/browse/ZOOKEEPER-1140?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13082146#comment-13082146
]
Laxman commented on ZOOKEEPER-1140:
-----------------------------------
Thread [Sender-/127.0.0.1:3084] (Running)
When I observed QuorumZxidSyncTest, the above thread is also leaking in a
different scenario apart from ReadOnlyZKServer. Above mentioned thread will be
shutdown only if it receives proposalOfDeath.
We are queuing proposalOfDeath in finally block of LearnerHandler.run().
To summarize, proposalOfDeath may not queued up when LearnerHandler receives
IOException and other thread calling LearnerHandler.shutdown(). This leads to
failure of queuing the proposalOfDeath.
To solve this, can we add the proposalOfDeath in shutdown() rather in finally
block.
Anyways, this finding will solve one of the leaks. We still need to fix other
leaks caused by ROZK.
{quote}Just to see if I understand now, are you saying that the test is
spawning over one hundred servers because we are shutting before the server
actually starts? If so, then it is certainly a problem.{quote}
Yes. I will add more analysis and possibly patch today.
> server shutdown is not stopping threads
> ---------------------------------------
>
> Key: ZOOKEEPER-1140
> URL: https://issues.apache.org/jira/browse/ZOOKEEPER-1140
> Project: ZooKeeper
> Issue Type: Bug
> Components: server, tests
> Affects Versions: 3.4.0
> Reporter: Patrick Hunt
> Priority: Blocker
> Fix For: 3.4.0
>
>
> Near the end of QuorumZxidSyncTest there are tons of threads running - 115
> "ProcessThread" threads, similar numbers of SessionTracker.
> Also I see ~100 ReadOnlyRequestProcessor - why is this running as a separate
> thread? (henry/flavio?)
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
