1 of 1 people found this helpful
As you use EPP 5.2, you will need to:
- Configure a shared FS at least an NFS v3 with locking enabled ideally NFS v4
- Set in you configuration.properties the property gatein.jcr.data.dir to the path of your shared FS
- Set in you configuration.properties the property gatein.jcr.config.type to cluster
- Set in you configuration.properties the property gatein.jcr.index.changefilterclass to org.exoplatform.services.jcr.impl.core.query.jbosscache.JBossCacheIndexChangesFilter
- Launch your nodes one after the other with the System properties exo.profiles set to cluster the launch command should be something like ./run.sh -c cluster -Dexo.profiles=cluster
Thanks for the response, i already have EPP configurated on cluster. But i want to have the databases you configurate (JCR and IDM) on cluster or disaster recovery.
Thanks and Regards.
I'm very sorry but AFAIK it is not covered by the JCR neither by IDM, IMHO HA to be efficient should be managed natively at your infrastructure level (DB + hardware). For the db part as you obviously use PGSQL you should read this http://www.postgresql.org/docs/8.3/interactive/high-availability.html and this http://wiki.postgresql.org/wiki/Replication,_Clustering,_and_Connection_Pooling. For the FS part, consider using approach like RAID (http://en.wikipedia.org/wiki/RAID) or SAN (http://en.wikipedia.org/wiki/Storage_area_network).