[Linux-HA] Auto/Nice Failback

ha at buglecreek.com ha at buglecreek.com
Wed Jul 13 15:03:01 MDT 2005


We are upgrading a 2 node cluster with a new OS.  The plan is to do the
upgrade first on machine2 while machine1 stays online.  The new ha.cf
file on machine2 will have auto_failback off.  The issue is that
machine1 has nice_failback on.  Our desire is to have machine2 stay
active after heartbeat is up on both machines to verify all is working
o.k.  Then the upgrade will be done on machine1. I'm not sure how the
systems will react since the failback options are different on both
machines.  Which file has the priority?  Do I need to edit the ha.cf
file on machine1 to nice_failback off?  Or not?

Thanks



More information about the Linux-HA mailing list