Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

IMS v9 on z/os 1.13

145 views
Skip to first unread message

Vince Getgood

unread,
Aug 20, 2012, 11:10:17 AM8/20/12
to
Hi all,
Is anyone out there running IMS V9 on z/OS 1.13?

I know IMS v9 has been out of support since 2010, but that's where we are. I can see from the IBM site that V10, 11 & 12 of IMS will run on z/os 1.13, but not v9.

I can't see why it wouldn't run, but would like some confirmation (or otherwise!) before I suggest it.

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to list...@listserv.ua.edu with the message: INFO IBM-MAIN

Bob Shannon

unread,
Aug 20, 2012, 11:13:49 AM8/20/12
to
> Is anyone out there running IMS V9 on z/OS 1.13?
We run lots of versions on 1.13 including V9.

Bob Shannon
Rocket Software

John Gilmore

unread,
Aug 20, 2012, 11:24:40 AM8/20/12
to
I know of a client site where V9, v10, V11, and V12 are all in
concurrent use under z/OS 1.13. It has been unproblematic except on
the occasion of one operator error, which has not been repeated.

--jg

Daniel Allen

unread,
Aug 20, 2012, 11:30:14 AM8/20/12
to
We run IMS v7, v8, v9 (unsupported) and v10, v11 and v12 (supported).

Mark Zelden

unread,
Aug 20, 2012, 12:01:08 PM8/20/12
to
On Mon, 20 Aug 2012 10:10:10 -0500, Vince Getgood <getg...@HOTMAIL.COM> wrote:

>Hi all,
>Is anyone out there running IMS V9 on z/OS 1.13?
>
>I know IMS v9 has been out of support since 2010, but that's where we are. I can see from the IBM site that V10, 11 & 12 of IMS will run on z/os 1.13, but not v9.
>
>I can't see why it wouldn't run, but would like some confirmation (or otherwise!) before I suggest it.


Unfortunately I have a client in that boat. Make sure you have PM46805 / UK71184
applied since z/OS 1.13 base has code that could lead to a problem described by PM46805.

My brief notes / timeline from my client below should help explain:

OA32318 / UA57797 - z/OS 1.11 support for IMS V12 applied 04/26/2011
OA32318 / UA57797 - goes PE on 06/28/2011
OA36933 / UA62725, - new APAR created to fix PE, but exposes a bug in IMS V9 fixed by PM46805 / UK71884.

Regards,

Mark
--
Mark Zelden - Zelden Consulting Services - z/OS, OS/390 and MVS
mailto:ma...@mzelden.com
Mark's MVS Utilities: http://www.mzelden.com/mvsutil.html
Systems Programming expert at http://expertanswercenter.techtarget.com/

Clark Morris

unread,
Aug 20, 2012, 1:55:41 PM8/20/12
to
On 20 Aug 2012 08:30:14 -0700, in bit.listserv.ibm-main you wrote:

>We run IMS v7, v8, v9 (unsupported) and v10, v11 and v12 (supported).

Is there a reason for doing so for other than vendors supporting
multiple releases.

Clark Morris

Brian Westerman

unread,
Aug 21, 2012, 12:07:59 AM8/21/12
to
As long as you have the most current maintenance available for your IMS release, it should work without problems. We currently have customers all the way back to V8 running under z/OS 1.12 and 1.13. None of them have any release related issues with IMS.

Brian.

Timothy Sipples

unread,
Aug 21, 2012, 3:46:28 AM8/21/12
to
I'm a bit puzzled why you'd be running IMS V9. It's generally quite
straightforward to upgrade versions, you can skip versions when you
upgrade, the newer versions pick up substantial CPU efficiency
improvements, they're supported, and the new functions are often
interesting. What's not to like?

--------------------------------------------------------------------------------------------------------
Timothy Sipples
Resident Enterprise Architect (Based in Singapore)
E-Mail: timothy...@us.ibm.com

Vince Getgood

unread,
Aug 21, 2012, 6:35:30 AM8/21/12
to
Thanks to all those that replied. Good to know we aren't the only ones!

We lost our IMS guy (retired), and are unlikely to replace him. I've never done an IMS upgrade, so rather than rush into it, I'm probably going to upgrade z/OS 1st, and approach IMS with caution!

Thanks again.

Avram Friedman

unread,
Aug 21, 2012, 8:14:36 AM8/21/12
to
I would suggest checking the IMS-L Listserv and the history of this sort of question.

There have been changes to both z/os and IMS access methods over the years.
IMS makes heavy use of VSAM shared services once called the VSAM DC option.
IMS also has its own zOS access method known as OSAM
(Originally Overflow Sequencial Access Method designed to replace ISAM overflow but now used by IMS for something else)
These access methods are very z/os service sensitive.
The more you use IMS the more you are exposed.
I would not pay much attention to ISVs who claim that they run many historicial version of IMS under zos 1.13.
They tend not to run them with high volume complex workloads and also do not really do any mesurable work on a day to day basis.

Avram Friedman

Avram Friedman

unread,
Aug 21, 2012, 8:46:41 AM8/21/12
to
It all depends but
In general IMS upgrades are easy as IMS is not tightly bound to the data.

The first upgrade beyond V9 is often not easy.
It was at V10 that IMS required security to be managed by an external security package like RACF.
Prior to that IMS internal security called SMU was an option as well as default security.
As an historical asside many early IMS shops did not use default security because it was too restrictive.

An IMS upgrade beyond V9 is difficult to the extent that SMU is used.

Avram Friedman

On Tue, 21 Aug 2012 05:35:23 -0500, Vince Getgood <getg...@HOTMAIL.COM> wrote:

>Thanks to all those that replied. Good to know we aren't the only ones!
>
>We lost our IMS guy (retired), and are unlikely to replace him. I've never done an IMS upgrade, so rather than rush into it, I'm probably going to upgrade z/OS 1st, and approach IMS with caution!
>
>Thanks again.
>

Mark Zelden

unread,
Aug 21, 2012, 1:09:05 PM8/21/12
to
On Tue, 21 Aug 2012 15:46:17 +0800, Timothy Sipples <timothy...@US.IBM.COM> wrote:

>I'm a bit puzzled why you'd be running IMS V9. It's generally quite
>straightforward to upgrade versions, you can skip versions when you
>upgrade, the newer versions pick up substantial CPU efficiency
>improvements, they're supported, and the new functions are often
>interesting. What's not to like?

My 2 guesses:

1) IMS is behind along with other software (so you might as well ask why
all that other software is behind also).

2) "Red Headed step child" - I suspect in shops where IMS is as important
as DB2 and CICS they don't tend to run (as far) behind.

My experience at several shops has been #2. Not much production IMS,
not much expertise left or people who care to get it upgraded, so it
is an after thought.

Mark
--
Mark Zelden - Zelden Consulting Services - z/OS, OS/390 and MVS
mailto:ma...@mzelden.com
Mark's MVS Utilities: http://www.mzelden.com/mvsutil.html
Systems Programming expert at http://expertanswercenter.techtarget.com/

0 new messages