I've seen the questions about backing up cluster resources, and while I think
that the answer to my question is to upgrade to 3.4, I'd like to avoid that if
possible.
I have 2 NT machines
NT1
NT2
They are cluster, and I have 2 aliases, NT-ALIAS-1 and NT-ALIAS-2.
ALIAS1 owns G:\
ALIAS2 owns H:\
Right now, I have 3 NB classes configured.
Class NT backs up NT1 and NT2 but excludes G:\ and H:\
Class NT-1 backs up G:\
Class NT-2 backs up H:\
Should either of the cluster resources fail over, my backup will restart on
the new node, and my backups will proceed properly.
I want to configure the machines to do SSO. The robot control host is on
another machine entirely.
I create 2 storage units:
ST1 lives on NT1
ST2 lives on NT2
I want to make sure that my backups always happen via a local tape drive
across my SAN, rather than over the network. In addition, I don't want my
machines backing up other clients.
I make ST1 and ST2 "on demand only" storage units.
I make class NT-1 force itself to use ST1 and NT-2 to use ST2. This works
great, as long as my cluster resources don't fail over between nodes. If they
do, then I either won't get my backup, or it will go across the network, which
I don't want. So, the question is:
1. Can I force local backups using the MUST_USE_LOCAL_DRIVE entry here? Does
the backup software realize that my storage unit is local, even though the
client its backing up is a cluster resource name, not an actual node name? If
I set this, but mark the storage unit to "on demand only" will the storage
unit still get used?
2. Or, do I need to go to 3.4 and create clustered media servers on the
machines so that the storage units themselves can fail between the machines?
Thanks.
--
Andy Steingruebl | e-mail: steingra AT pprd.abbott DOT com
Unix Systems Admin/Programmer | phone: (847) 935-4728
Unix/Network Security | fax: (847) 935-0142
Abbott Laboratories, PPDR&D | post: 100 Abbott Park Road, D472 - AP9A
| Abbott Park, IL 60064-6115
|