OK, just a few more things I didn't see anywhere else.

     This might not apply if you are using IBM's disaster recovery package
     but I'm not and here are a few things to worry about (not a complete
     set, just a subset dealing with tapes)

     Make sure you pull the volume history off your adsm server(s) often.
     The number one reason is so you will know what volume contains your
     data base backup!
     Another reason is because after you perform a restore you will need to
     perform an "audit volume" against any and all volumes that were
     accessed after the data base backup (from which you are restoring) was
     made to bring the DB back into sync with the physical data on hand.

     I pointed out in my example earlier that I keep the last 7 days of
     full data base backup tapes. THIS IS ONLY TO COVER MY @$$. With the
     amount of activity any one of my adsm servers has, it would be
     improbable to rebuild the environment from anything older than the
     most recent dbbackup.  Say on average I have 20 - 3590 tapes that get
     altered each day.  This is for migration of my disks to tape and then
     tape reclamation once expiration has run.  Each tape will have from
     10-1,000,000 files on it anywhere from 10GB - 45GB (due to
     compression).  If on the average it takes 1 hour to perform an audit
     against each of these volumes each day further back I have to restore
     my data base will require me an additional 20 hours of audit volume
     time.  This is also 20 hours of time I really shouldn't be backing up
     any new data which is also 20 hours of potential data loss for the
     company, which is bad enough but when 40 or 60 hours comes into
     play... well, uhmmm.  BUT I keep enough data base backup tapes to
     where if for what ever reason a copy(ies) goes bad, I should have at
     least ONE that is good (ie I went the extra mile to ensure minimal
     data loss).

     Oh pardon me, I just double checked... last Friday's average unique
     volser tape mounts was 44... So each day's activity lost would take me
     aprox two days to recover.  So if I had to go back 3 data base backups
     I would not be backing up anything for about a week.

     And this is only an average for the environments here at Amoco, most
     of you will probably see a shorter "audit volume" time and others
     might see longer ones...  but you should be aware of such things.

     I've exceeded one screen so I'll shut up now.

     later,
           Dwight