Re: 2nd level install - duplicate volsers

Fri, 04 Nov 2005 12:01:36 -0800

You can't always count on the lower address volsers being found first.  That's the way it used to work before the XA/ESA I/O subsystem came into existence.  I always thought that was still the case but John Franciscovich or one of the IBM Endicott VM gurus set the record straight a year or two ago on this list.  Now, the order of what is found first is, at least to CP, random in that the interrupts that are returned aren't necessarily in ascending device number sequence. 

If you install this way onto duplicate volsers, you're only going to be safe if you set the test addresses offline at ipl in SYSTEM CONFIG and then when you bring up the test system, vary the addresses online and attach them to the guest.  That would be the thing to do until you get the volids changed.

Jim

At 02:51 PM 11/4/2005, you wrote:
Or you can indeed install using full-pack minidisks (cylinder 0-END) and change the second level system volsers once you are up and running.  My DASD naming convention includes a bit to indicate TEST or PRODUCTION (i.e. VMPRES for the VM Production RESidence volume, VMTRES for the VM Test RESidence volume, etc. - and could be expanded to include multiple test systems by using other flags such as VM0RES for Test System 0, VM1RES for Test System 1, etc. etc.).  That will allow your 2nd level system to have full access to the spindle (including cylinder 0) if that's more desirable.
 
When installing using this method, place your test system volumes on spindles at physicall addressing HIGHER than your production system spindles having the same names.  VM will take the FIRST spindle it finds at IPL time in the event of duplicate DASD labels, so your production system volumes should be those with the lower addresses to ensure your production system gets the correct spindles should you IPL (planned or otherwise) prior to changing your second level system labels.
 
David's method (using cylinder 1-END) is arguably easier at the cost of a cylinder, and loss of access to real cylinder 0 by the second level guest (should that be needed).
 
Michael Coffin, President
MC Consulting Company, Inc.
PMB 123
289 Park Street
Stoughton, Massachusetts  02072
 
Voice: (781) 344-9837    FAX: (781) 344-7683
 
[EMAIL PROTECTED]
www.mccci.com

We employ aggressive SPAM filters.  If you cannot reply or send email to mccci.com go to www.mccci.com/spamblockremove.php

 
-----Original Message-----
From: VM/ESA and z/VM Discussions [ mailto:[email protected]] On Behalf Of David Boyes
Sent: Friday, November 04, 2005 10:19 AM
To: [email protected]
Subject: Re: 2nd level install - duplicate volsers

When you define the minidisks for the 2nd level system, make them cyl 1 to end of the volumes. The installation images are deliberately 1 cyl short to allow for this to happen. If you do that, then the real (1st level) system sees unique volsers, and the 2nd level system sees the same volsers as production.
 
Ain't virtualization lovely?
 
-- db
 

From: VM/ESA and z/VM Discussions [ mailto:[email protected]] On Behalf Of Clark, Carl
Sent: Friday, November 04, 2005 9:58 AM
To: [email protected]
Subject: 2nd level install - duplicate volsers

VM enthusiasts:

   Have any of you installed a 2nd level VM by using the Guide for Automated Installation and Service?  It uses the same volsers as our 1st level system. 

Does anyone know how to use this install method but specify different volsers?

Your help is appreciated,
Carl Clark
BCBST 

Please see the following link for the BlueCross BlueShield of
Tennessee E-mail


disclaimer: 

http://www.bcbst.com/email_disclaimer.shtm

Jim Bohnsack
Cornell Univ.
(607) 255-1760

Reply via email to