I've been reading over the Sept. 2003 AppNote entitled "Running Novell
BorderManager on Novell Cluster Services" and have a couple of questions I
hope the author(s) or someone else familiar with the procedure can answer:

1. The AppNote is written for BM 3.7 & 3.8 in a NCS version 1.6
environment. Is there any additional considerations/changes for BM 3.8 on
NCS version 1.7?

2. I'm a little fuzzy on the placement of the Proxy cache volume(s). The
AppNote refers to a shared volume object called "BM_POOL_SERVER" for the
load and unload scripts (essentially the failover stuff from what I
gather). Then later in the article it states not to use NSS for the proxy
cache volume(s). But from what I understand of NCS (I admit somewhat
limited at this time...but growing rapidly)...the shared volume object can
only be an NSS volume. So how does a Traditional File System (TFS)
volume(s) play into this? Can a TFS be failed over in a cluster safely
without using a pool? Also, if NSS can't be used...what's the point of the
shared volume object if the volume can't be used for anything? Wouldn't
another resource object type make more sense?

The reason I go on like this is because it's not very clear to me in the
AppNote...my apologies. ;)

For instance...on page 38 of the AppNote the 'Load Script' refers to
activating a pool and mounting two volumes. It's not clear where those
volumes are (ie. part of the pool or not). If they are part of the pool,
then what are they used for?

Thanks for your help...and feel free to smack me if it's clearly laid out
and I just don't get it. :D


BTW. Other then my confusion on the the volume stuff...the AppNote is
awesome! Major kudos to the authors!