From another customer - your thoughts?
Our thinking (for a specific product & maintenance level): deploy to alternate - update names and run there until satisfied. Then deploy to primary - update names and run there until satisfied. Before applying any maintenance - deploy to previous to have a backup - (you still have primary - I know) , apply maintenance - deploy to alternate update names and run there until have. Repeat prior steps.
Thanks for your input.
We have a central repository and perform all our maintenance and testing there so we use only the Primary set of disks.
When our testing is completed the production data is deployed to a production distribution system that has all the production disks for IBM and any 3rd party vendors on a single 3390 mod-9 then we use CMSDDR to package up the disk and send it out to about 40 z/VM across our data centers. Each of our systems maintains a GOLD disk and a ALT disk as part of the system. We lay down the new code disk on the ALT disk and do any local testing at that site. On cutover day we just have MAINT point to the new disks and build all the save segments. then finally swap the labels on the GOLD and ALT disk and SHUTDOWN REIPL to implement.
If all we need is a fix we send the affected files or disks out to our systems and placed them as needed and cycle a service machine as needed.