I have seen the problem too, and I am on level 13, so it would appear that
the fix wasn't too good.  I too have an open problem (but no APAR yet) with
Level 2.  I recently saw a 4 hour migration process move about 6GB of data to
tape, and when I saw it (at the end of the process - last reel I think - it
said that the wait time for the current mount was in excess of 14,400
seconds.  That would indicate that the process had been waiting for 4 hours,
which was the overall length of time the process had been active.  I figured
we either had a bug, or I had a very efficient STK Silo that had written 6 GB
(to a 480 MB 3490 tape) in a very few seconds!

I agree that this is not a serious problem, but when I am trying to figure
out if I need to go feed my operators some more bananas to get them to put a
tape back in the silo, or if I am just waiting for a drive, it gets to be a
real pain - it's like having no information at all.

Jerry Lawson


______________________________ Forward Header __________________________________
Subject: Re: migration/mount waits
Author:  INTERNET.OWNERAD at SNADGATE
Date:    10/30/97 11:23 AM


Yes, I have seen this, but it was quit some time ago.  I reported it to
IBM, and they opened apar PN76397, which was fixed, for the MVS server, by
level 2.1.07.  What level of server are you on?  It should go away if you
upgrade to the current level.

Bill Colwell
The Charles Stark Draper Laboratory
Cambridge Ma.

_________________________Reply Header_________________________
Author: ADSM-L AT vm.marist DOT edu
Subject: migration/mount waits
10-30-1997 03:20 PM

When migration from a diskpool to a collocated tapepool is running,
several output-volumes will be needed.
For each volume that needs to be mountet the process will wait for the
tape to be mounted. (we have a 3494 with 4 3590's, and the average
mount-time is about 60 seconds)

During this wait-time the show process command (admin CLI) will at the end
of its output say for example

Waiting for mount of output volume BCK326 (7521 seconds)

(in this case the migration-process had been running for about two hours
an was wating for it's 22nd tape to be mounted)

If this number (7521) was right, something wold be terribly wrong with the
library. (the other tapedrives were not bysy)
But the tape was mounted in less than one miunte...

I then thougt the number could be the accumulated wait-time for the
 process. But it's not.

It's actually the total of all the wait-time plus the time each tape has
been mounted. Which adds up to be about the total running time of the
process....

NOt a major problem, but it's quite annoying.. ANyone else seen this?
Comments?