----------------------------------------------<snip>----------------------------------------
I would respectfully disagree. I posed this question to the group a
while back and the consensus was to let it crash. I believe the thought
was that there was just not enough time or information to make an
informed decision. Plus, anytime someone is frantically banging on a
keyboard, screw ups are a sure bet.
I had to make this call a while back and followed the advice. All in
all, things went very smoothly and we were back in full operation with
very few issues. DB2 and JES knew what to do and did it. Batch jobs that
were running were simply treated just like any other failure. Our most
loved proprietary online hit the ground running.
I did drain initiators to prevent any new batch jobs kicking off.
I learned a few things, though. One, if you have an OS/2 HMC, shut it
down ASAP. When power comes back, it does a check disk that takes
forever. I had to IPL using the support elements.
The Shark has backup batteries so that it will come in for a graceful
landing. That worked. But the unexpected twist was that the Shark would
not go ready until the batteries were recharged. We now have a DS8100
but I would expect it would be no different.
So, my suggestion for a loss of power scenario is to immediately
evacuate personnel to a safe place. Nothing is worth getting someone
hurt. BTDT.
---------------------------------------------<unsnip>------------------------------------------
I tend to agree with your conclusions, expecially the last one. Consider
using your automation to accomplish an orderly shutdown, or
alternatively, use something like the COMMAND program from the CBTTAPE
site for same. While it's not always perfect, it will help eliminate the
"finger checks" that so often happen in times of great stress. Getting a
clean termination of batch jobs will always be problematic but
automation of some sort can be used to at least affect an orderly
shutdown of online systems, DBMS's, etc. We use COMMAND and other than
batch jobs, we are completely shut down and ready for power down, or
IPL, in about 2 minutes. (3 CICS regions, 1 DBMS, plus TSO, VTAM and
various monitoring tools.)
(We learned just how valuable this was during the Chicago Flood of 1992,
when Edison have us 10 minutes warning before cutting our power.)
--
Rick
--
Remember that if you’re not the lead dog, the view never changes.
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html
