Re: PWRDWNSYS

Not even controlled works as it should.
Machine: 9406-515, low end 600CPW
RAM: 4Gb
Disks: 8 4327
Problem:
Any, and all shutdowns be they controlled or not, end in an Abnormal Shutdown.
Reason:
Some Genius between Rochester and the customer (in Argentina) made the
old style LPAR (via SST) and assigned only 1Gb of the 4 to the main
partition. So crippled the system had Gbs upon Gbs of pagefiles and
even with 1hour to flush it would still fail to flush. Ah, that was
also the reason that after installing a print group PTF that changed
the behaviour when restarting a spoolfile from page >1000 the system
took more than 2 minutes to restart. With 1Gb ram v5r3 DID NOT work.
So i would say:

If you are going to play with the machine, either set caches to failed
or powerdown from the HMC with the option "Are you going to replace a
cache battery? YES"

Best Regards,


On Wed, Jan 16, 2013 at 1:24 PM, <rob@xxxxxxxxx> wrote:

<snip>
that 30 seconds are enough time to flush the cache?
</snip>

That was rather my point questioning the poster who said the 'trick' was
to use a delay, even as low as 5 seconds, on a controlled and you'd never
have to use *IMMED. Really, the trick is, after you've brought the system
into restricted state, and you've used the controlled with a time limit,
is to sign off your system console. Because, with what they've said then
all would be well with PWRDWNSYS *CNTRLD DELAY(1) right?


Rob Berendt
--
IBM Certified System Administrator - IBM i 6.1
Group Dekko
Dept 1600
Mail to: 2505 Dekko Drive
Garrett, IN 46738
Ship to: Dock 108
6928N 400E
Kendallville, IN 46755
http://www.dekko.com





From: Gqcy <gmufasa01@xxxxxxxxx>
To: midrange-l@xxxxxxxxxxxx,
Date: 01/16/2013 11:12 AM
Subject: Re: PWRDWNSYS
Sent by: midrange-l-bounces@xxxxxxxxxxxx



Is there a document proving this?
if you would do a PWRDWNSYS *CNTRLD 30,
how do you know that 30 seconds are enough time to flush the cache?

I would think something as important as this, there would be some parm
on PWRDWNSYS to flush the cache...

I understood the *CNTRLD to be for those times you do not wait for the
"SYSTEM IN RESTRICTED STATE" message on the ENDSBS *ALL command. That is
what the help makes it sound as well...






On 1/16/2013 8:49 AM, Patrick.Bingham@xxxxxxxxxxxxx wrote:

I would have to agree with what the consultant said. When working with

disk you should always end the system controlled because it does flush the
cache in the disk controller.

--
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.


--
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.