Greetings MVS folks.
A few SHARE’s ago, several of us did a presentation on Successful Parmlib Practices. You can see it here https://share.confex.com/share/118/webprogram/Session10648.html
I have to admit I was a bit surprised at the interest. There were 85 people in attendance in Atlanta! Ok, one of them was my husband. But it got me to thinking…What other ‘normal’ topics are there that people might want to hear other perspectives on? Part of the Parmlib presentation mentioned maintenance strategy, but just peripherally. So I thought I’d explain our maintenance strategy to you, and hopefully you can SHARE yours in the comments section.
Without further ado, here’s what we do:
We apply maintenance bi-monthly, and include everything up to two months back. So, in December, I put on maintenance up to October 2012. So that’s PUT1210 and RSU1210, and PUT1209 and RSU1209. We then go through hipers up to that day and apply any that are particularly evil looking. These are few and far between. We generally like to say that we only put on a bleeding edge hiper is if we’ve actually encountered the issue, but on a couple of occasions it looks so nasty and we’re using all the functions, so I’ve put them on.
All that said, I often wonder about putting PUTs on. I’d have to say we do it ‘because that’s the way we’ve always done it’. Personally I’m not too keen on putting PUTs on, but I haven’t been able to pinpoint one that went bad and hurt us.
TIP: A nightly or weekly batch job to receive PTFs from IBM is a beautiful thing. You just run GIMSMP, and I have a sample at the end of the blog.
After the apply, we clone a new res pack and IPL the system programmer LPAR. Though this sounds pretty small, it actually gets some good use. At 11AM on Thursday, it has 13 people logged on to TSO. The maintenance stays there for two weeks, then moves to Developer and QA LPARs. It shakes out there for two weeks under pretty heavy usage then moves to the first production system. From then, it takes us four weeks to IPL the other 8 production LPARs. Which brings us right back to the next round of maintenance!
So now you have the ‘What’ and the ‘Why’. My next entry will be the ‘How’ and ‘Where’ of maintenance for us. Stay tuned!
In the meantime, how often do you do maintenance, or do you do it all? Some people just do MVS release upgrades rather than maintenance. Do you have a fairly rigorous clone procedure set up, or is it a struggle every time? (I ask because z/OSMF and CA MSM might help with that).
Your link today is to the CBTTape Top Ten. Did you know that it was still around? It is around, active, and relevant!
Till next time,
After much effort and consternation, I finally decided to screen print these, because the >'s were messing up the html. (sad) I realize that makes it difficult for you to cut/paste them. If you need them, email me at firstname.lastname@example.org.
If you don't have an SMPE Service Certificate, you can get one on SHOPZ: