Re: HACMP failover service IP issue

a_g_lyph_at_hotmail.com
Date: 07/08/04


Date: 8 Jul 2004 07:38:41 -0700

When I say, 'took the secondary node out of the cluster,' I mean that
on the secondary (failover) node I did:

smit hacmp->Cluster Services->Stop Cluster Services

The primary node was rebooted, but the secondary node was not. After
the primary node came back up, I had to manually mount some filesystems
(previous admin did not set them up to be mounted automagically at
boot). After I got everything running on the primary node, I then
re-joined the secondary node to the cluster:

smit hacmp->Cluster Services->Start Cluster Services

The service label is part of the resource group and has the proper
'production' IP associated. It did not, however, migrate to the
primary node on reboot, even though I ran the 'Bring a resource group
online' on the primary node.

FWIW, I just ran `clconfig -v '-t' -m 'All' -R
'/home/aglyph/clverify.out'` and there were no errors reported. This
only verifies topology, though, from what I can see, but I would think
that should show any problems.



Relevant Pages