Originally posted by: alethad

I am configuring a new cluster. HA6.1 AIX 6.1
Prod app runs on Node1 and test app runs on Node2.

They cannot run on the same node at the same time. On the old boxes I had to configure the cluster where test app is brought down in a pre-event(acquire_service_addr) in order to bring it down before Prod RG is acquired.
Test will never failover. But I would like to have it in a RG instead of handling it thru any HA event to be brought down if Node1 fails.

I was told this is possible but I can't figure out how to set the fail over policies correctly. I simply would like to have it go offline when Prod fails over but I get an error when I set the "Bring Offline" policy. Doesn't seem to matter what combination I try.

Anyone ever done this before that can tell me how to set up the RG the way I want it?

Thanks.
You've got to continue to grow, or you're just like last night's cornbread -- stale & dry Loretta Lynn alethad

Originally posted by: Holgervk

use rg dependencies

smit cm_rg_odn_dependencies.dialog

Originally posted by: alethad

How do RG dependencies help with this?
From what I understood this tells HA to keep an RG online somewhere so how does that work for bringing it offline?

You've got to continue to grow, or you're just like last night's cornbread -- stale & dry Loretta Lynn alethad

Originally posted by: Holgervk

if a RG with a higher prio wants to be brought online on a node where a RG with a lower prio, hacmp shut that lower-prio down before starting the higher-prio

you can use the command
/usr/es/sbin/cluster/utilities/clrgdependency -t'ANTICOLLOCATION' -u -hp'name_of_your_higher-prio-rg' -ip'' -lp'name_of_your_lower-prio-rg'
to configure this.
I use this on many clusters.

Originally posted by: alethad

I'll take a look.
thanks.

You've got to continue to grow, or you're just like last night's cornbread -- stale & dry Loretta Lynn alethad