We've done a whole lot of ServerPacs over the years. In every case up to 
now, we've done full replace (FSR) even though we have a perfectly usable 
set of 'operational data sets' in every sysplex that we end up using 
anyway as we migrate across the Enterprise. One reason for FSR is that 
it's kind of comforting to know that any previous sins (commission or 
omission) will get washed away in a fresh ground-up install. That comfort 
level comes at the price of creating a number of throw-away data sets like 
SPOOL, RACF, PAGE, etc. that we never intend to use. We minimize those 
data sets and selectively skip jobs that prepare them as SOP. 

I've come back from SHARE more than once with the ambition to try software 
upgrade (SU). I'm all ready to shift gears for 2.1 except for one sticking 
point. We depend heavily on a brand new level-specific catalog not just 
during ServerPac install but throughout the life of the z/OS release. That 
is, a data set created on sysres looks like this: 

  SYS1.library 

SMP/E finds SYS1.library via the name OSRnn.SYS1.library, where OSRnn is a 
master catalog alias that points to MVSRnn.ICF.MASTER (our name), a user 
catalog created during FSR of release nn. Only the driving system needs 
this user catalog for maintenance and migration purposes because in 
production, sysres data sets are cataloged directly in master as 
SYS1.library . 

So here's the question. Having never gone the SU route, I don't know 
whether it will yield the expected MVSRnn.ICF.MASTER . If so, I'm off to 
the races. (Too late to enter Triple Crown.) Otherwise having to rework a 
number of tried-and-true processes would surely be more work than tweaking 
FSR yet again, a task we have practiced many times. 

So the 64 dollar question: does SU create an OS-level specific user 
catalog that we can continue to use for the life of the release?

.
.
J.O.Skip Robinson
Southern California Edison Company
Electric Dragon Team Paddler 
SHARE MVS Program Co-Manager
626-302-7535 Office
323-715-0595 Mobile
[email protected]

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to