This was our test environment - z/VM 5.3, about 120 active VMs, 24 GB
memory divided 20 central and 4 expanded, and 54 mod-3s as paging
devices.  STORBUF is 300 300 300.  We've been short on memory for quite
a while (typically 72+ GB virtual), but evidently nothing is gonna help
in a case like this.

 

Michael Czora

Wells Fargo

Mainframe Resource and Performance Management

612-667-0044

 

This message may contain confidential and/or privileged information.  If
you are not the addressee or authorized to receive this for the
addressee, you must not use, copy, disclose, or take any action based on
this message or any information herein.  If you have received this
message in error, please advise the sender immediately by reply e-mail
and delete this message.  Thank you for your cooperation.

 

-----Original Message-----
From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On
Behalf Of Rob van der Heij
Sent: Friday, October 12, 2007 3:33 AM
To: [email protected]
Subject: Re: Sanity check?

 

On 10/12/07, Marcy Cortes <[EMAIL PROTECTED]> wrote:

 

> One shouldn't define a virtual machine of size 1500G when one meant to
do

> 1500M!

 

We have been spoiled by an operating system that only uses what you

need, rather than what you give it. So the big game with Linux is to

give it just what it needs...

 

Normally when paging space fills, you overflow into spool space.

Because spool space is often small that does not help you a lot and

you abend shortly after that. That message is clear.

 

But I assume this is z/VM 5.2 and you were starting a Linux virtual

machine, right?  If so, then I would expect your system stalled

because of PGMBK fragmentation. If you still have the performance data

from that period, it would be good to look at the ESAASPC report - if

possible with 1 minute granularity.

 

Rob

-- 

Rob van der Heij

Velocity Software, Inc

http://velocitysoftware.com/

Reply via email to