Hi,

I'm confused here about the savesets field for pools.

Let's suppose I have a client named fred, and I want to backup
everything on fred (except /disk2) to pool A only, and I want to back up
fred:/disk2 to pool B only. I would have thought that I would need two
client instances: instance_1 lists 'All', and uses a server side
directive to skip /disk2, and instance_2 specifically lists /disk2. Each
client instance would be in a separate group, and then I'd list each
group under its respective pool. But I see that I can actually list a
file system under a pool and not select a group. So what will happen if
instead of creating client instance_2, I instead just list /disk2 under
pool B? When the first group runs it will obviously back up 'All' but
will skip /disk2. Will /disk2 then get picked up at the same time by
pool B even though I don't have an active group selected for it? Seems
very foreign to me to not invoke a backup via a group.

I guess I'm trying to figure out what the difference is between these
two approaches. Which is better? Also, I'd like to grab a copy of the
client index to pool B, too. If I list it under the savesets for pool B
will this just be a copy, or will this force the client index to ONLY go
to pool B?

Currently, indexes go to pool A, and I'd like to keep it that way. Just
wanted a copy to go to pool B. Guess I could clone a copy to a clone
pool (have considered that) but would be nice if I could have a copy
automatically go to pool B. To be clear, though, the reason why I want a
copy of the index on pool B is because the file system that's going to
pool B is an important one, and I want it on its own physically separate
set of tapes (yes, I do make non-Legato tar archives, too) so it won't
be subjugated to the normal recycle policies affecting the normal pool
of tapes (pool A), along with the respective index so I can always
recover an older version of the index to be able to browse this data.
The indexes and the savsets will never be recycled on this pool. I'm not
concerned with being able to browse the other data that's going to pool
A any further back than the 2 month browse policy now set for the client
so I could care less about recovering any indexes from more than 2
months back that would contain that other data.

Any help or advice would be appreciated.

George
George.Sinclair AT noaa DOT gov

George

--
Note: To sign off this list, send a "signoff networker" command via email
to listserv AT listmail.temple DOT edu or visit the list's Web site at
http://listmail.temple.edu/archives/networker.html where you can
also view and post messages to the list.
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=