I agree with your position about using the ADSM facilities to manage the
deletion of tapes, and how long they are kept. I would also wonder about what
would happen if ADSM returned a tape to scratch, deleted it from it's record
of "world knowledge", only to have you produce the tape and tell ADSM to use
it anyway!
Additionally, the CA-1 parm, because it covers ALL tapes in your installation,
may not be a changeable option - it is set at a value to provide all
applications with a little breathing room; it is something that might save
your @#$$ someday, but I wouldn't depend on it being there. Also, it might be
adjusted to a smaller value if thenumber of tapes being held back is too high,
causing more consternation when you don't expect it. I believe that using the
ADSM controls is the ONLY way to go here.
Jerry Lawson
jlawson AT itthartford DOT com
________________________Forward Header________________________
Author: INTERNET.OWNERAD
Subject: Re: What is volume status of Pending?
05-28-96 01:44 AM
Jerry Lawson wrote:
>Great answer. You covered it in great detail. One other variable that may
or
Thanks!
>may not be worth considering is what CA-1 does with the tape after it is
>returned by ADSM and expired. In our shop, we have a CA-1 exit coded as you
>suggest. Somewhere in this mix, we have a delay for actual reuse by CA-1 set
>to 3 days. I believe, but am not certain, that this is established in a CA-1
>parmlib (I'm not a TMS guru, hardly even a novice). The effect of this is
>that even after a tape shows an expiration date (when ADSM returned the
tape),
>it is not actually used for a scratch mount until the 3 day grace period has
>expired.
Yes, I have heard of this delay, too. When I was running HSM (well before i
got
involved with ADSM) we used this delay for HSM tapes. It effectively does the
same thing as REUSEDELAY, and I think that was the intent because HSM doesn't
provide that feature. But ADSM does provide that feature, so you can use
either
ADSM's REUSEDELAY, the CA-1 delay, or both. I can't see the point of using
both, as it's redundant. My own personal preferance would be to use ADSM's
REUSEDELAY only because ADSM is the external data manager -- once ADSM takes
the scratch tape, it has complete control over that tape until the tape is
ready to be returned to CA-1 as scratch. Same thing with DELETE VOLHIST
TODATE=-n, where 'n' is a number of days. The effective number of days would
be 'n + y' where 'y' is the CA-1 delay. Again, it's redundant since you would
hypothetically be using a value of 'n' that already meets your needs.
I like ADSM's REUSEDELAY because it seems 'purer' -- at least to me. But
my own personal preference aside, from a security standpoint, I don't think it
really matters which one you use. Some operational issues may differ, however.
For example, if you use CA-1's delay and have ADSM's REUSEDELAY=0, then once
the tapes are deleted from ADSM's inventory, you won't be able to do a QUERY
VOLUME WHERESTATUS=PENDING; you'll have to obtain that information from CA-1.
It's a nit, but one I thought I'd point out.
Hmmmmmm......... come to think of it, there may be a disadvantage to using
CA-1's delay. If you ever had to go back to a prior version of the ADSM data-
base, the "pending" volumes, from CA-1's point of view will still have expira-
tion dates. This means that they could conceivably expire out from under you.
Whereas if you used ADSM's REUSEDELAY, the restored database would still have
them in part of ADSM's inventory, and the expiration date would still be
permanent. With the CA-1 delay, you'd have to determine which tapes are
affected and mark them with permanent retention again. This process, I'm sure
you'd agree, adds complexity and more chance for error.
So I'm changing my mind. I would definitely recommend using ADSM's
REUSEDELAY instead of CA-1's delay.
Andy Raibeck
ADSM Level 2 Support
408-256-0130
|
