I'm having some difficulty getting the clustering setup correctly. We have both servers fully setup and everything appears to be in order but when I start the rgmanager it gets stuck in a loop.
It gets to the point where it runs the "/opt/zimbra-cluster/bin/zmcluctl status" command then receives 1 instead of 0 for the return value. It then thinks the node has failed and stops all of the services. Once they are stopped it tries to bring them back up and it repeats.
I'm guessing one of two things is happening. Either the IP is for some reason not being bound to the interface or there is some environment variable that is not set correctly. When we run the command manually it returns this error:
Missing service name (OCF_RESKEY_service_name env var not set)
Once we set that variable and ran the command manually it returned nothing but there is no documentation about setting it nor does it work when the command is run by the cluster software.
Does anyone have any insight into this matter?
FYI our setup is a simple active/passive 2 node cluster with a separate LDAP. There is no MTA installed anywhere since we will only be using contacts and calendar.