jackrabbit-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Shaun Barriball" <sbarr...@yahoo.co.uk>
Subject JackRabbit clustering : "No log file found containing revision: 1111540"
Date Mon, 23 Apr 2007 09:58:22 GMT
Hi all,
We've been playing with the JackRabbit clustering features using version
1.2.3.
After running a few simple tests the main issue is that the nodes appear to
get 'out of sync' and then we've found no doc on how to bring them back into
sync. One node now fails to start up and reports:

org.apache.jackrabbit.core.cluster.JournalException: Unable to iterate over
modified records.
        at
org.apache.jackrabbit.core.cluster.FileJournal.sync(FileJournal.java:237)
        [big stack]
Caused by: java.io.IOException: No log file found containing revision:
1111540
        at
org.apache.jackrabbit.core.cluster.FileRecordCursor.getRecordLog(FileRecordC
ursor.java:122)
        at
org.apache.jackrabbit.core.cluster.FileRecordCursor.next(FileRecordCursor.ja
va:98)
        at
org.apache.jackrabbit.core.cluster.FileJournal.sync(FileJournal.java:227)

So I'd appreciate any feedback from those in the know on:

 * how can the above issue be resolved?

 * is it more reliable to use the Database Journal (with which we've also
had issues)?

We'd really like to use the clustering so hopefully we can iron out the
above.
Regards,
Shaun.


PS> The clustering config we're using is as follows:

<Cluster id="[nodeName]" >
  <Journal class="org.apache.jackrabbit.core.cluster.FileJournal">
    <param name="directory"  value="/opt/jcrrepositories/cluster/journal"/>
    <param name="revision" value="${rep.home}/repository/revision" />
  </Journal>
 </Cluster>


-----Original Message-----
From: Shaun Barriball [mailto:sbarriba@yahoo.co.uk] 
Sent: 05 April 2007 15:09
To: 'users@jackrabbit.apache.org'
Subject: JackRabbit clustering documentation

Hi all,
Can anyone point in the direction of any documentation on the clustering
features other than:

http://comments.gmane.org/gmane.comp.apache.jackrabbit.devel/9381

I'm after info on:
 * the relationship between the persist store, filesystem and the database
journal e.g. I assume all cluster nodes share the persistent store?
 * the procedure to bring another node into the cluster. Do you need to copy
the filesystem of an existing cluster node etc?
 * any other recommendations

Regards,
Shaun.



Mime
View raw message