Thanks. Standing up a test cluster in VMs was actually next on my list to try
but it wasn't required.
The DHCP was pretty simple, one subnet only and a few reservations, and there
was actually another server that had the configuration from before the cluster
was created so that's good. A little more digging revealed they only had 3
printers to deal with and all the drivers were already there so I just
recreated them.
The fun part was that the failed SAN volumes were preventing me from moving
quorum. I finally just removed it forcibly and changed it to point elsewhere.
At that point the cluster was up, but it wouldn't service any of the shares
because of the dead volumes. I removed the dead volumes and everything else is
available.
Since we're tearing down the cluster anyway as part of our migration I think
we're good at this point, but I have learned a lot of things about what NOT to
do. :) Fortunately I didn't do them to begin with, and wouldn't have made
those choices (I don't think) so I guess it always helps to be able to learn
from someone else's mistakes.
--
There are 10 kinds of people in the world...
those who understand binary and those who don't.
From: [email protected] [mailto:[email protected]] On
Behalf Of Brian Desmond
Sent: Wednesday, June 17, 2015 6:05 PM
To: [email protected]
Subject: [NTSysADM] RE: What if - Destroying a 2008R2 cluster
You should be able to force it to start - there are some command line switches
for the service to force quorum and such.
The print config is stored in the registry and my gut says that it's in the
Cluster replicated part which would make me think the queues might go away. The
DHCP database is stored in a JET database so presumably you could restore that
elsewhere.
This would be easy to test in a VM - create a one node cluster, add DHCP and
Print with a scope and a single print queue and then nuke the cluster and see
what happens.
Thanks,
Brian Desmond
[email protected]<mailto:[email protected]>
w - 312.625.1438 | c - 312.731.3132
From: [email protected]<mailto:[email protected]>
[mailto:[email protected]] On Behalf Of Melvin Backus
Sent: Wednesday, June 17, 2015 6:25 AM
To: [email protected]<mailto:[email protected]>
Subject: [NTSysADM] What if - Destroying a 2008R2 cluster
We had an OMG event in a remote office yesterday when one of the local "IT
guys" rebooted a clustered server while the other node was disabled for
maintenance. The maintenance event was to correct a problem on a SAN unit
which is mirrored via Veritas Enterprise Administrator. Things were running
normally until the reboot, but when they rebooted the server it lost
connectivity to the witness disk for the cluster as well as some other data
volumes. Since the witness disk is unavailable the cluster server is offline
so DHCP and Print services won't run.
Since we're going to have to destroy the cluster anyway as part of the
migration, we're thinking about doing it now to simply some of what needs to be
done, but we don't have the DHCP/Print configurations readily available. (Yes,
it's all backed up, but restoring is taking longer than anticipated) If we
destroy the cluster, does anyone know if the DHCP and Print configuration data
gets retained by the host server? If so it should be trivial to recreate on
another server as long as we can see it.
--------------------
Melvin Backus | Sr. Systems Engineer | Byers Engineering Company | 404.497.1565
Service Desk | 404-497-1599 | http://servicedesk.byers.com
--
There are 10 kinds of people in the world...
those who understand binary and those who don't.
