Part of the problem here is that we have no doc (no dumps were taken).  I did
see what appeared to be a loop, but we were in such deep trouble that we
decided to wait for a few hours for a scheduled IPL, and didn't dump ADSM.
Part of the problem has been that ADSM is so reliable that when we do take a
failure, I tend to not collect data on the first failure, since the few that
I have been involved with usually don't reoccur, and are caused by something
else..... Not at all like the Bad old days of MVS.......

APAR PQ02977 has been suggested to me by a cuple of people.....do you
recommend installing it?

Jerry Lawson
jlawson AT thehartford DOT com


______________________________ Forward Header __________________________________
Subject: Re: A strange weekend
Author:  INTERNET.OWNERAD at SNADGATE
Date:    9/11/97 11:52 AM


 Yes, the scheduler code is shared by scheduled clients and scheduled
admin commands. There is problem with admin command scheduling getting into
a loop on a schedule in which the duration has elapsed, but the schedule is
still in pending status. This is fixed in PTF15. This normally only ocurrs
on very slow MVS processors, but can happen on other platforms.
 There are too many variables in determining why to be discussed here.
You really should contact ADSM Level 2 on this; they will ask for the
proper doc (lots).

Tom Brooks
ADSM Development