jackrabbit-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Sergey Podatelev <brightnesslev...@gmail.com>
Subject Re: Repository won't start after import
Date Wed, 04 Mar 2009 12:06:07 GMT
I'm sorry for the delay, I had troubles reproducing this, until I got even
worse issues.

First off, I'm sorry for this stupid question, but how do I enable
debug-level output?
I have log4j.properties file in my /src folder with the following content:

log4j.rootLogger=DEBUG, stdout
log4j.appender.stdout=org.apache.log4j.ConsoleAppender
log4j.appender.stdout.layout=org.apache.log4j.PatternLayout
log4j.appender.stdout.layout.ConversionPattern=%-4r [%t] %-5p %c %x - %m%n

However, I can only see INFO messages from the Jackrabbit's classes (my own
classes emit some DEBUG-s as well, and I can see those).

Secondly, I actually had a couple of "successful" imports, after which my
web application started normally.
However, what I'm getting now is somewhat similar to what I've found here:
http://issues.apache.org/jira/browse/JCR-160

Specifically, I get messages like "Exception retrieving Node with UUID:
fabf6884-1457-4ccf-94f3-dde9d8319f6d: javax.jcr.ItemNotFoundException:
fabf6884-1457-4ccf-94f3-dde9d8319f6d"
from time to time though I didn't do any UUID lookups.

And there're NodeIterator failures.
Specifically, NodeIterator iter = q.execute().getNodes(); iter.getSize() --
returns 1, but iter.hasNext() returns false.

What scares me a little bit is that even if I delete
"/var/jackrabbit/repository/workspaces/default/index" folder, those
ghost-UUID errors still happen.

Perhaps this is somehow related to the fact that before these UUID-errors I
also configured webapp to work through Spring's LocalTransactionManager?

On Wed, Feb 25, 2009 at 4:58 PM, Marcel Reutegger
<marcel.reutegger@gmx.net>wrote:

> Hi Sergey,
>
> On Mon, Feb 23, 2009 at 19:24, Sergey Podatelev
> <brightnesslevels@gmail.com> wrote:
> > ...*repository initialized*
> > ...*basic read: find a node, read it's name*
> > ...*repository is shutting down*
> > 5198 [main] INFO
> > org.apache.jackrabbit.core.observation.ObservationDispatcher -
> Notification
> > of EventListeners stopped.
> > 5698 [main] INFO org.apache.jackrabbit.core.query.lucene.IndexMerger -
> *Unable
> > to stop IndexMerger. Deamon is busy.*
> > 6009 [main] INFO org.apache.jackrabbit.core.query.lucene.SearchIndex -
> Index
> > closed: /var/jackrabbit/repository/workspaces/default/index
> > 6022 [main] INFO
> > org.apache.jackrabbit.core.persistence.bundle.DerbyPersistenceManager -
> > Database '/var/jackrabbit/repository/workspaces/default/db' shutdown.*
> >
> > *Is this is somehow related to the index corruption?
>
> Yes, that could be the reason. I've opened a jira issue with the above
> log information:
> https://issues.apache.org/jira/browse/JCR-1999
>
> Can you please re-run your test with debug messages enabled for the
> class IndexMerger and attach the log file to the jira issue? Thank
> you.
>
> > Also, there's a typo in
> > that warning, it's either "daemon" or "demon" (:
>
> Thanks for noticing, I'll correct it.
>
> regards
>  marcel
>



-- 
sp

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message