Michael,

Rick

On 11/27/2012 8:03 AM, Michael Leone wrote:

OK, so I admit I am having a mentally slow day, and I should probably know
this, or be able to figure it out. But I'm not, so I figured I would come,
cap in hand, and beg for some information and understanding ...

I have a number of MS clusters,. These clusters have virtual clients,
which are then mounted on a physical node. These virtual cluster clients
have other cluster resources associated with them, specifically drives
mapped to SAN storage. Fine so far. But here is where I seem to be falling
down. I have the cluster client defined as a Networker client, as well as
the physical nodes, and I have jobs that backup saveset ALL of the
physical nodes, and other jobs that backup saveset ALL of the cluster
client. Again, all well and good.

Since the cluster client is virtual, there is no console for it; what you
actually connect to is the console of the physical node. And when I bring
up NWUser on that node to do a recover, the only choice of clients I get
is that node I am on. And that node then only shows me the local, physical
drives - it is not showing me the mapped SAN drives, because the SAN
drives belong to the named cluster client (obviously).

If I try the recover from the NW server, I do see the cluster client name
as the client to recover from, and then  I do see the mapped SAN drives to
use as sources for the recover. Which means if I wanted to restore from
those mapped drives, I would have to do it from the NW server, and
re-direct over to the cluster client. Again, I understand all that.

But I seem to remember that I used to be able to see the cluster client as
a source on the actual physical node. And I don't remember what I might
have done to change that.

So, in effect, what I want: at the moment, the only place I see those
cluster clients when doing a recover is if I do it from the NW server.
What I would like (if I could, and I seem to remember being able to do
this) is to be able to also see the cluster client as a source for a
recover on the actual physical node that the cluster client is mounted on.
If I ever needed to do recovers of the cluster client, it would be a lot
easier for me to do it from one of the physical nodes, as opposed to
having multiple re-directed recovers that have to be run only from the NW
server.

Am I just mis-remembering, or did I change something that limited me like
that? I feel sure I used to be able to see it. I have a feeling (altho I
can't articulate exactly why) that it has something to do the security of
the ADMINISTRATORS group in NMC.

Can anyone clue me up? Thanks.