Just a couple of questions...

Why do you do incremental DB backups instead of full backups?  Each of our ADSM
databases are between 3 and 6 gig and we run 2 full DB backups in the morning.
One goes to tape and the other to a disk copy group.  The disk copy we then
backup to tape which allows us to cram about ten db's on a tape.  It only takes
about 20 minutes to do a full backup...

Second, are you sure you want to set reclamation to 50%?  As I understand it,
that may mean you move more information than is actually useful.  If you had ten
volumes, 9 of which were full and the 10th had 50% full, ADSM would call in
another scratch tape and move that 50% to a new tape - gaining you nothing!  Am
I wrong?  We set our pools to 60% to avoid this problem...

Scott







"Moir,Betsy" <betsy.moir AT ABBOTT DOT COM> on 04/06/99 07:25:37 AM

Please respond to "ADSM: Dist Stor Manager" <ADSM-L AT VM.MARIST DOT EDU>

To:   ADSM-L AT VM.MARIST DOT EDU
cc:    (bcc: Scott Fluegge/COR/MLW/US/venatorgroup)
Subject:  Re: log growing too quickly ...




Just a couple of things to think about - we've been trying to get the balance
just right on this for a while now.  Do you do incremental backups on a
scheduled basis or just when the trigger gets set off?  We found doing a
scheduled incremental backup of the database every night helped
considerably.  We also do a full dbbackup on the weekend.

We see the problem happen most often when we're doing large reclamation
processing or filespace deletion - lots of records being written to the
recovery log.  We had our reclamation processing turned off during the week
(we're pushing our processor limits on a daily basis), but just recently
decided we should set reclamation to 50 at night during the week to help ease
the load on the weekend.  We don't want to increase our log size, because
generally it's more than enough space.

Mostly it's a fiddle and tweak kind of thing, I think.  And your particular
setup and what kind of data is being handled by each server will make a
difference.  Of course, I don't know if the fact that we're still running a
V2 server on VM has any effect on any of this.  Just some thoughts.




ADSM-L AT VM.MARIST DOT EDU on 04/05/99 08:57:54 PM
Please respond to ADSM-L AT VM.MARIST DOT EDU @ INTERNET
To: ADSM-L AT VM.MARIST DOT EDU @ INTERNET
cc:
Subject: Re: log growing too quickly ...

Hi Steve,

We've got exactly this problem on one of our four ADSM servers at the moment.
We are getting a triggered DB backup at least once per day and can't
understand why. The server with the problem doesn't do significantly more
than 2 of the other servers (the last server is a baby). We had a problem
over the weekend where the DB backup triggered (at 70%) and the incremental
backup didn't finish before utilisation reached 100%. Not pretty.


Trevor