[Pacemaker] Time out issue while stopping resource in pacemaker

[Lax]

Andrew Beekhof <andrew at ...> writes:


> I'm guessing you don't have stonith?
> 
> The underlying philosophy is that the services pacemaker manages need to
exit before pacemaker can.
> If the service can't stop, it would be dishonest of pacemaker to do so.
> 
> If you had fencing, it would have been able to clean up after a failed
stop and allow the rest of the cluster to continue.

Thanks Andrew. I have a 2 node setup so had to turn off stonith. 

One more thing, on another setup with same configuration while running
pacemaker I keep getting 'gfs_controld[10744]: daemon cpg_join error
retrying'. Even after I force kill the pacemaker processes and reboot the
server and bring the pacemaker back up, it keeps giving cpg_join error. Is
there any way to fix this issue?  

Thanks
Lax