rhcs rgmanager service group state not in sync
Have rhcs setup on two nodes. I have a service group configured "service:svc-pgsql-instance1" when I run "clustat" on node1 the service state is showing (owner: node1, state: started). on node2 clustat is showing (owner: none, state: disabled).
I am trying to get rgmanager back in sync, preferably without having to interrupt service.
troubleshooting steps:
1. I have tried rebooting node2 (the standby node), but it comes back up, but continues to say (owner: none, state: disabled) when I run clustat.
2. I've viewed the corosync,rgmanager logs, and it does seem that rgmanager is communicating with one another at some level, and I am not seeing anything out of the ordinary in the logs.
3. I also tried freezing the resource group on the active node (node1) using "clusvcadm -Z <svc name>" but it just hangs until I eventually hit <CTRL+c>.
Any suggestions on how I can clear this up with out having to reboot the active node and interrupt service would be much appreciated.
|