Hi Maria. Yes, I do believe that I can create the next z/os software instance very quickly now that we have a v2.5 instance. We use what the Serverpac dialog called ‘system upgrade’ , retain the existing master catalog and use indirect cataloging.
In fact I was so confident that I could do this I re-ordered the Serverpac because the first order had two target zones (I had assumed that this was because I had accidentally ordered 2 releases of the same product - node.js IIRC - but it turned out to be because I received ISPF-case and upper case panels). Incidentally my shopz orders were shipped within 24 hours - it used to take a few days. This turned out to be a mistake because I had not actually finished the first install so it was not available as a model so I had to use the 2.4 instance. But with my experience of the first installation tailoring the dataset names and volumes was very quick. On reflection, using my first 2.5 instance as a model may not have made much difference if only SMPE-managed datasets are modelled. I much prefer z/osmf to the previous method. I never did learn all of the commands to edit the dataset names and volume layout. (Laying down the files in a couple of hours is a bit optimistic for me. I had to spend a lot of time freeing up disk space for the order etc and downloading the order and running the ‘unpack’ job took several hours. But the z/osmf dialogue part was quick and easy). BTW I am the odd-ball who uses 2-level aliases for the ZFS datasets. I think the reason may be that when we moved to ZFS we could no longer use indirect cataloging for file systems so we added a second qualifier after “OMVS’ to create unique names. Then , when I realised that it was easier to put the ZFS datasets on the single large sysres, I needed to catalog them on a catalog residing on sysres to make cloning easier. We ended up with 3 “qualifiers” associated with a particular software instance on sysres: the volser, the target zone name and the qualifier in the “OMVS” datasets. I know that it would be simpler to use the volser as the HLQ but we have to maintain and run several back levels of z/os and I am reluctant to change. I may however do that even though it is a minor ‘post install’ step to move the entries out of the ‘OMVS’ catalog to a catalog Keith Gooding ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
