ADSM-L

Re: Version 2 install

1996-06-28 17:15:00
Subject: Re: Version 2 install
From: Bill Colwell <bcolwell AT CCLINK.DRAPER DOT COM>
Date: Fri, 28 Jun 1996 17:15:00 -0400
Text item: Body.822
By all means, test the heck out of it!

I think a better test scenario would be to setup a version 1 test server.
Backup a few clients, migrate to some tapes etc.  Then do the smp for the
v2 product and start it on the v1 test server with parm='/upgradedb'.  Then
 exploit the new features of v2 like db backup, copy storagepools etc.
This will be a better dry run of your eventual production  upgrade (we have
to be careful about saying 'install', since that has a special meaning for
ADSM).
_________________________Reply Header_________________________
Author: ADSM-L AT vm.marist DOT edu
Subject: Version 2 install
06-28-96 03:33 PM

Date:         Fri, 28 Jun 1996 15:33:15 -0600
From:         Jane Dagostino-Snyder <j-dago AT ngwmail.ais.umn DOT edu>
Subject:      Version 2 install
To:           Multiple recipients of list ADSM-L <ADSM-L AT vm.marist DOT edu>

Hi,
Thanks to those who asked my pervious question on this same subject.
I am using MVS as my server:
The read me doc states you can install a separate Version 2 server and
test before installing it over your Version 1 database.  I am considering
that option.   Has anyone else done this first or did most install over
your
Version 1 database right away?  This is my First install so these may be
basic questions.

Now say that a separate Version 2 was installed on a test LPAR.  Not
where the Version1 is and new pools were defined,  can you define
new pools there or not??  I still need the old ones of course.  Then once
you test Version 2, install Version 2 over Version 1.  Is this what they
are referring to or what is the recommendation?

This is what I am thinking of:

Version 2, new install on CPUD receive/ apply/accept with new pools,
logs, DB.    Set up test enviroment...

Next:  Install Version 2 over Version 1:  Do not install or run jobs to
define
new pools, ect.   upgradedb, I have read page 51 in install book.

Usually when installs that place, we use different volumes, CPU's
TCPIP's ect before going live.   Can this not be done this way?  Looks like
since Version 1 is in production that you are basically overlaying Version
1 on a production system.

Comments, Ideas,  ALL welcome, well, maybe not all?  Be kind !!

Thanks, Jane

Jane Dagostino-Snyder
University of Minnesota
Storage Administration
(612)  624-4883
<Prev in Thread] Current Thread [Next in Thread>