Re: Question - Your IPL Frequency

["Jim Essinger" <dilbernator@xxxxxxxxx>]

We do run a few jobs overnight, but they don't have to be.  We would have to
do some planning, but for the most part the system is idle all night.

We are on maintenance. We upgraded to a 520 with 1 terabyte available,
mirrored, plus some RAID protected user ASPs using the old disks. I IPL once
a week, and can't remember the last time a disk failure had us down.  We
have replaced a few drives with hot swap, but they did not cause us any down
time or data lose.  I have not replaced a disk since October of 2005, when
we did our upgrade.

We did test the Service Agent after upgrading to V5R4, we had an optical
drive fail two weeks ago in our jukebox, and used WRKPRB to place the call.

I have been trying to order PTFs this week via the service agent, and have
had problems.  Another thread on this list mentioned that Fix Central might
be doing some changes.  I sure wish they would let us know before hand, so
we don't spin our wheels trying to get fixes downloaded.

Good thoughts to tell the boss, though.   Better to find out in an IPL, when
it will still come up, than to lose the drive with a full load on the
system.


On 8/30/06, rob@xxxxxxxxx <rob@xxxxxxxxx> wrote:
> I don't think we'll ever do that.  We do a lot of nightly batch jobs,
> backups etc.  In addition to off shift work.  I really didn't think I'd
> find anyone who was.  Rather surprised.
>
> Jim,
> If you have maintenance, and are RAID protected, then convince your boss
> that a stress test on the disks shouldn't have more than one fail at a
> time.  Better to find out at IPL time, after your nightly backup, then
> during prime shift.
>
> Oh, by the way; test service agent after every OS VRM upgrade.  Sometimes
> a release upgrade hoses it up and it no longer calls IBM.  Losing one disk
> drive and having the machine call IBM and they replace it and your users
> never notice is no big deal - been there done that.  However, screwing up
> service agent and not having it call IBM and you not finding out until the
> second drive in the raid set fails and the whole system goes casters up is
> not a good idea.
>
> Rob Berendt
> --
> Group Dekko Services, LLC
> Dept 01.073
> PO Box 2000
> Dock 108
> 6928N 400E
> Kendallville, IN 46755
> http://www.dekko.com