RE: BRMS restricted system -- MIDRANGE-L

Loyd,
I did put the *Exit to end subsystems because I did this not under *SYSTEM control group but *BKUGRP, and I could not see in docs that just the *CTLSBS itself would force a restricted state. I do know this running did not go to a restricted state and I had files not saved.

Log:
2/19/08 2:05:01 Begin processing for control group *BKUGRP type *BKU.
2/19/08 2:05:01 Error occurred processing exit command for control group
2/19/08 2:05:02 Devices TAPMLB01 will be used for control group *BKUGRP t
2/19/08 2:05:03 Interactive usage check indicated no user jobs active on
2/19/08 2:05:05 Change to media controls successful.

This is the BRMS detail of the error msg:

Message ID . . . . . . . . : BRM1387 Severity . . . . . . . : 30

Job . . . : BKUGRP User . . : QSECOFR Number . . . . : 005253
Date sent . . . . . . . . : 2/19/08 Time sent . . . . . . : 2:05:01
Program . . . . . . . . . : Q1ARBK Area . . . . . . . . . : *BKU

Message . . . . : Error occurred processing exit command for control
group *BKUGRP.
Cause . . . . . : The control group *BKUGRP was processing the exit command
at sequence 0010 when an error occurred. Check the job log for more
information.

This is the actual joblog in spool file:
BRM1387 Information 30 02/19/08 02:05:01.786768 Q1ACALG QBRM *STMT Q1ACALG QBRM *STMT
From module . . . . . . . . : Q1ACALG
From procedure . . . . . . : Q1ACALG
Statement . . . . . . . . . : 229
To module . . . . . . . . . : Q1ACALG
To procedure . . . . . . . : Q1ACALG
Statement . . . . . . . . . : 229
Message . . . . : Error occurred processing exit command for control group

This is V5R4M5 w/cume 7282 and brms group level 17

Jim Franz

---- lgoodbar@xxxxxxxxxxxxxx wrote:

What are the errors? Can you share the BRMS log (DSPLOGBRM)?

V5R4 BRMS configuration:
*SYSTEM policy page 2
Controlling subsystem
Allow backups in batch: *YES
Restricted state end time: nnn
This allows BRMS to end the batch job if restricted state time in
minutes is exceeded. Otherwise, need to force DST on console and take
option to end restricted batch job.

Then in your job scheduler:
STRBKUBRM CTLGRP(WEEKLTOCG) SBMJOB(*CTLSBS)

You may already have this, just verifying.

In the backup control group, you do NOT need any commands to end to
restricted state, that is part of the BRMS pre processing.
The first item in my control group is *SAVSYS

All console sessions are signed off during the backup, including the
QCONSOLE/DSP01 you would normally use for recovery. This is why the
restricted state end time is important.

Not to tarnish this feature, I have an open issue with IBM that this
does not work correctly in my controlling partition. (9406/810,
controlling partition is a "working" partition, not dedicated)

Here is what BRMS log said.
2/17/08 3:33:18 Command ended due to error.
2/17/08 3:55:25 *SAVSYS bypassed for control group WEEKLTOCG.
2/17/08 3:55:25 *SAVSECDTA and *SAVCFG issued for control group
WEEKLTOCG
...
backup continued otherwise
...
2/17/08 6:23:42 Media information report is created.
2/17/08 6:24:16 Subsystem not started because system ending.
2/17/08 6:24:17 Subsystem not started because system ending.
2/17/08 6:24:17 Error occurred processing exit command for control
group
2/17/08 6:24:17 Control group WEEKLTOCG not successful.
2/17/08 6:24:17 Control group WEEKLTOCG not successful.
2/17/08 6:24:17 Command ended due to error.
2/17/08 6:24:17 Command ended due to error.
2/17/08 15:45:21 System start up program executed.

The *SAVSYS was bypassed because the system was not in restricted state.
Normally, BRMS waits for the system to be restricted before starting the
backup, but for us something timed out and BRMS continued anyway.

In our 2-LPAR system, the controlling LPAR/production environment was
"down" with no connectivity - Navigator, TCP/IP, no twinax devices. The
secondary/test LPAR was running fine.

IBM support said that part of the system (not necessarily BRMS) thought
it hadn't ended to restricted state yet, and the system could not come
out of restricted state because it was going into restricted state.

Some objects were not saved because they were in use. After IPLing the
system (which per IBM was a forced power down/abnormal end), many system
jobs were ended during IPL with end code 70.

IBM wants to investigate whether ending to restricted state in
production causes any issues.

I should note that this same process (restricted state BRMS backup to
batch, no interactive sessions) works flawlessly in our test LPAR.

Hope you get this working in your environment because it is a nice hands
off backup process. Troubleshooting it is more difficult, especially if
you are remote to the system.

HTH,
Loyd

Loyd Goodbar
Senior programmer/analyst
BorgWarner
TS Water Valley
662-473-5713
-----Original Message-----
From: midrange-l-bounces@xxxxxxxxxxxx
[mailto:midrange-l-bounces@xxxxxxxxxxxx] On Behalf Of
franz400@xxxxxxxxxxxx
Sent: Tuesday, February 19, 2008 11:17
To: Midrange Systems Technical Discussion
Subject: RE: BRMS restricted system

My problem is it does not work for
STRBKUBRM CTLGRP(*BKUGRP) SBMJOB(*CTLSBS)
Many errors in joblog.

Am considering making the nightly a full system save, else
tracking all the subsystems from various products and
ending/starting.
jim

--
This is the Midrange Systems Technical Discussion (MIDRANGE-L) mailing list
To post a message email: MIDRANGE-L@xxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: http://lists.midrange.com/mailman/listinfo/midrange-l
or email: MIDRANGE-L-request@xxxxxxxxxxxx
Before posting, please take a moment to review the archives
at http://archive.midrange.com/midrange-l.