Fcuk, sorry -- filed the issue under Lucene, moved:
https://issues.apache.org/jira/browse/SOLR-3748

Dawid

On Tue, Aug 21, 2012 at 10:01 PM, Dawid Weiss
<[email protected]> wrote:
> I filed this issue for this build failure:
> https://issues.apache.org/jira/browse/LUCENE-4320
>
> Uwe took a snapshot of log files and I was able to analyze them. Conclusions:
>
> 1) Solr's BasicDistributedZkTest stalled after ~40 seconds and showed
> no sign of progress until timed out (grep for 'jstack' in the log file
> attached to the issue).
>
> 2) the timeout mechanism works fine, so does stack trace dumps it seems,
>
> 3) this particular JVM seems to be buggy because a call to interrupt()
> on those threads should be safe and resulted in end-the-world JVM
> termination, the events file is truncated in the middle of an atomic
> event write operation so I think it was very low-level abort. Uwe took
> care of this already I think by upgrading the JVM.
>
> Dawid

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to