[
https://issues.apache.org/jira/browse/JCR-2433?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12872812#action_12872812
]
Paul Jara commented on JCR-2433:
--------------------------------
The original problem seems to be addressed which is good news. Another problem,
which might not be related to this one, appears to have cropped up in its place
however. Our CMS grinds to a halt and stops serving up requests. The problem
seems to have started since I deployed the new JackRabbit JARs to the
production servers. In the past couple of weeks, our servers have only been
able to run for a few days at most before one of them just stops serving
requests (Tomcat takes the request but never actually sends a response). Is
there anything in those nightly 1.6.2-SNAPSHOTs that may have contributed to
this problem? There doesn't appear to be anything in the server logs that is
helping me diagnose the problem. But when I go to shutdown Tomcat, I get a
bunch of messages like this in the catalina.out:
"INFO: Waiting for N instance(s) to be deallocated" where N is some large
number like >100.
I was notified tonight that our website was "down". We have a couple of servers
and each Tomcat instance had stopped serving requests. So on each server, I
stopped Tomcat, deleted all the JackRabbit indexes for each workspace
associated with our CMS, and restarted. Upon the restart on both servers, I got
a slew of these messages:
"A property claimed to start before zero, at -512! Resetting it to zero, and
hoping for the best"
However, I restarted again and everything seems back to normal. Do you think
this fix and the problem I'm now seeing are related? Or is it just some bad
luck that the problem seems to have begun around the time that we found a fix
here?
> NPE when copying nodes with Workspace.copy()
> --------------------------------------------
>
> Key: JCR-2433
> URL: https://issues.apache.org/jira/browse/JCR-2433
> Project: Jackrabbit Content Repository
> Issue Type: Bug
> Components: jackrabbit-core
> Affects Versions: 1.6.0
> Reporter: Sascha Theves
> Assignee: Jukka Zitting
> Priority: Blocker
> Fix For: 1.6.2
>
>
> I get a NullpointerException when using Workspace.copy():
> java.lang.NullPointerException
> at
> org.apache.jackrabbit.core.BatchedItemOperations.copyNodeState(BatchedItemOperations.java:1834)
> at
> org.apache.jackrabbit.core.BatchedItemOperations.copyNodeState(BatchedItemOperations.java:1806)
> at
> org.apache.jackrabbit.core.BatchedItemOperations.copyNodeState(BatchedItemOperations.java:1806)
> at
> org.apache.jackrabbit.core.BatchedItemOperations.copy(BatchedItemOperations.java:423)
> at
> org.apache.jackrabbit.core.WorkspaceImpl.internalCopy(WorkspaceImpl.java:444)
> at org.apache.jackrabbit.core.WorkspaceImpl.copy(WorkspaceImpl.java:666)
> at xxx.MyClass.myMeth(MyClass.java)
> It seems that it happens not all the time. The error occurs since we use
> Jackrabbit 1.6.0. We do not get the error with previous versions. It seems
> that we only get the error when trying to copy nodes that were created with
> Jackrabbit 1.4 and copied with Jackrabbit 1.6.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
