jackrabbit-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Michael McCandless (JIRA)" <j...@apache.org>
Subject [jira] Commented: (JCR-2426) Deadlock in lucene (Jackrabbit 1.4.4)
Date Sun, 14 Feb 2010 11:30:27 GMT

    [ https://issues.apache.org/jira/browse/JCR-2426?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12833557#action_12833557

Michael McCandless commented on JCR-2426:

That's a dangerous JRE to use with Lucene -- 1.6 JREs less than _10
can cause index corruption due to a known Sun hotspot compiler bug
(see LUCENE-1282, referenced from

Also, consider switching to NIOFSDir (if you're not on windows) -- it
avoids contention when multiple threads want to read from the same

As of 2.9, Lucene has removed the
FSDirectory$FSIndexInput$Descriptor's finalize method.  But, even so,
that finalize method is only invoked when the instance is being GC'd,
which makes no sense given that there are other threads actively using
it for searching/merging.  I think we need something else to explain
why the thread dump seems to incorrectly claim that
"jmssecondaryApplnJobExecutor-7" is holding the lock...

> Deadlock in lucene (Jackrabbit 1.4.4)
> -------------------------------------
>                 Key: JCR-2426
>                 URL: https://issues.apache.org/jira/browse/JCR-2426
>             Project: Jackrabbit Content Repository
>          Issue Type: Bug
>          Components: indexing
>    Affects Versions: core 1.4.4
>            Reporter: Antonio Martinez
>            Priority: Blocker
>         Attachments: deadlock_summary.txt
> We get a deadlock in lucene part of jackrabbit (see deadlock_summary.txt)
> This issue has been observed in a production setup running Jackrabbit 1.4.4 in cluster

This message is automatically generated by JIRA.
You can reply to this email to add a comment to the issue online.

View raw message