Clustering-Issues

[Christian Wurbs <Ch...@itemic.com>]

Hello together,

 

 

I'm using Jackrabbit 1.5.6 with clustering. The clustering itself works
without any problem.

 

But there is a very annoying issue especially during development:

 

.         There are - let's say - two cluster nodes working for a while.
These have already produced 100000 revisions in the datastore.

.         When I'm going to add a third cluster node to the cluster it looks
like it's replaying 100000 journal entries - which really takes its time for
the third node to come up.

 

Is there a way of having the third (new) cluster node start at the latest
Global-Revision immediately?

 

And second:

 

If I temporarily shut down the second cluster node I receive the following
error messages during synchronization at restarting this second node:

 

2009-10-26 13:09:04,797 [main] INFO
org.apache.jackrabbit.core.cluster.ClusterNode - Processing revision: 897

2009-10-26 13:09:04,812 [main] WARN
org.apache.jackrabbit.core.cluster.ClusterNode - Error making update
listener for workspace trgWorkSpace online: trgWorkSpace

2009-10-26 13:09:04,812 [main] ERROR
org.apache.jackrabbit.core.cluster.ClusterNode - Update listener unavailable
for workspace: trgWorkSpace

 

How can I avoid this?

 

Any help would be appreciated since I couldn't find anything regarding these
questions?

 

 

Christian Wurbs