ADSM-L

schedule problems

1997-04-18 19:04:19
Subject: schedule problems
From: Leonard Boyle <SNOLEN AT VM.SAS DOT COM>
Date: Fri, 18 Apr 1997 19:04:19 EDT
Last night a severe problem with our scheduled backups. If
fact it appears that tonight we still have the problem.

We had about 400 clients that normally complete come thru the night
with a schedule event status of missing.

This clients are novell 3.12 and windows nt 4.0 srv pack 2 systems.
with the client code at the ptf 5 level for the most part with a few at ptf6.
The systems should all be defined with prompted mode.
The Windows NT systems are set up with the adsm central scheduer service.
The server is the adsm ver 2 server at the latest code level (.9).
The server was upgraded from ver 1.

We spot checked a few of the nodes. We found no messages in the server log
and no messages on the client logs. For these machines.
That are no windows nt event logs for the covered period of time,
and no updates in the dsmsched.log or dsmerror.log files.

It appears that the server never contacted the clients that were
ready and able to be contacted.

With one node it reported an eventid of 4100 of next scheduled event...
Server Window stat 17:10 on 04/17/1997 at 4/17/97 9:50:16am.
But on 4/18 at 10:15 it was after the window ended and marked as missed.
I enrolled it in a midday schedule and the user rebooted his machine.
it then started the midday schedule.

From a problem machine the user could run a manual backup.
Novell and windows servers in another adsm server had no problem.

I turned on the server trace for the sched class and every 600 secs
I see the following:

   schedule manager awake
   pending table scan not done
   next deadline: 04/19/1997 03:30:00
   sleeping 600 secs


The "pending table scan not done" seems a little strange to me. At this
moment there are 5 scheduled session active and a max session setting of
20. And normally at this time of the day there are 20 running.

Can anyone shine any light on this problem? Anyone seen this problem
with any server platform?


At this time I will be restarting the server in the morning. I suspect
that it will clear the problem. But I have two open questions.
Was this happing before in smaller amounts and we just did not know.
And will it happen again.



Thanks for you time len

PS I have an ETR opened with IBM

-------------------------------------------------------------------------
Leonard Boyle, Mainframe support            snolen AT vm.sas DOT com
Leonard Boyle, Mainframe support            snolen AT vm.sas DOT com
SAS Institute Inc.                          ussas4hs@ibmmail
Room E206                                   (919) 677-8000 ext 6241
203 SAS Campus Drive
Cary NC 27513
<Prev in Thread] Current Thread [Next in Thread>
  • schedule problems, Leonard Boyle <=