Easy, but it has become a bit more difficult with the latest networker
versions.
Stop the networker services.
find the default pool resource.
use an editor to change "recycle to/from" for the default pool
start networker services.
Networker stops you from changing the attribute from the gui/nwadmin, but it
does accept the change if you change the resource by hand.
That is the answer to your first question.
I can't answer your second question because it requires more thought. My
first thought is that it does not make sense because you will need tapes in
the pools on servers AND storage nodes. Also you will need a lot of tape
changes if you ever have to recover the networker server.
On thing you should consider is that the index is backed up after completing
the group because it ensures that you have the latest index on tape just in
case the server happens to go up in smoke. But the indexes have become
"normal" files with the new 6.x index structure. This means that you can
also do a normal backup of the indexes instead.
Bye
Ernie
-----Original Message-----
From: Oscar NOSPAM Olsson [mailto:spam1 AT QBRANCH DOT SE]
Sent: Friday, April 25, 2003 10:16 AM
To: NETWORKER AT LISTMAIL.TEMPLE DOT EDU
Subject: [Networker] NDMP clients indexes gets backed up to pool Default
- how to solve?
We have recently begun backing up a netapp filer via NDMP. We run
snapimage on the backup server, and one jukebox tape on the backup server
has been reassigned as a dedicated NDMP device. However, the indexes that
belong to the NDMP clients gets backed up to the pool Default. This is
undesirable since that pool can't be edited, and thus recycle to/from
other pools can't be enabled.
As far as I have understood, only NDMP data may be backed up to the NDMP
drive, and only NDMP data should be written to the pool that is specificly
assigned for NDMP clients, correct?
Our normal clients are grouped in different groups, and the pools match
the client group. Then the indexes for a specific client gets written down
to the same pool as the client. The NDMP clients are grouped in a specific
group that is dedicated for only NDMP clients. The NDMP dedicated pool
matches only client names. Why not match by group this time as well, one
may ask? Well, its because I have understood that one needs to separate
NDMP data from other normal legato data, which the indexes still are,
correct?
So what I have tried is creating a new pool that matches the client names
and saveset data in the format (without ")
"therealnsrservername:therealclientname:index" - however it still wishes
to write indexes to the default pool. The backupserver itself, when backed
up, is matched by group membership, as everything else, when networker
decides to what pool the backup should be written to.
So how can I solve this without creating a tape or file device that has a
tape beloning to group Default mounted? Perhaps matching the nsrservername
and the specific index savesets to a specific pool?
I'm asking because I'd like to select the best solution from the
beginning. So, any ideas? Huh? Huh? ;)
//Oscar
--
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.
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=
--
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.
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=
|
