jackrabbit-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Edward Winston <ewins...@periscopeholdings.com>
Subject Invalid cluster node removes data
Date Tue, 04 Mar 2014 17:22:09 GMT
We have seen an interesting and a little disturbing behavior with
Jackrabbit.  If a cluster configuration is incorrect and the journal points
to a directory that does not exist, a new journal is created.  This node
then begins to remove all of the data in the repository, I am guessing,
because the journal appears out of sync with the database and it starts
trying to reconcile everything.

When all is finished the database tables are mostly empty and the data
directory has been cleared out.

We are using Jackrabbit 2.6.4, with an Oracle or SQLServer database.  The
actual data is stored on the filesystem, and the journal is file based.

Is there a configuration that can be set to throw an error or terminate the
process if the journal doesn't exist?  I can see some chicken and egg
issues here but I can see where this may be useful feature.

Edward Winston
That money talks,
I'll not deny,
I heard it once,
It said "Good-bye.
		-- Richard Armour

View raw message