IBM used to document how it assigned the prefix in sysmod ids, but its been 
decades since I've seen documentation that matched practice. What I've never 
seen violated is that the sysmod ids associated with an APAR all have the same 
numeric portion.

The simplest situation is when the error only exists in one sysmod, only one 
sysmod id is needed in the error hold and a single PTF supercedes. the error 
sysmod. When the error exists in multiple PTFs, IBM may need to clone the error 
sysmod, in order to provide ++APAR fixes for each PTF.


-- 
Shmuel (Seymour J.) Metz
http://mason.gmu.edu/~smetz3
עַם יִשְׂרָאֵל חַי



________________________________________
From: IBM Mainframe Discussion List <[email protected]> on behalf of 
Phil Smith III <[email protected]>
Sent: Friday, November 3, 2023 8:51 PM
To: [email protected]
Subject: APAR theology (was: IBM APAR Names)

Without wanting to start a war, I’m interested in how this works. I’ve worked 
with IBM stuff for over four decades, but mostly with VM until the last 15 or 
so years.



My understanding is:

*       PMR: represents a customer issue, which may end there.
*       APAR: represents a customer issue that at least seems to indicate a 
code problem. Existence of an APAR does not guarantee a fix will ever be 
created; it’s more a recognition that there’s more to it than just “customer 
needs an existing fix”, “is doing something dumb”, etc. One APAR might thus map 
to multiple PMRs, if multiple customers have the same problem.
*       PTF: An actual fix



Now y’all are talking about “APAR fixes”, which I’ve never heard of. And I know 
there’s more to APARs than I wrote, just can’t remember it. Having worked for 
vendors for the last 37 years means I’ve been out of the normal flow of such 
things, too—when I was at UofWaterloo, we had a backlog of a couple of hundred 
things that we’d fixed in IBM code; we had an agreement with our PSR—the sort 
of “level 1 ½” that customers have [had, then, anyway] that we’d only have 5 
open at once. That was always fun: call IBM, “I want to open five problems.” 
“FIVE PROBLEMS???!” “Well, I have a couple of hundred, but I’m only opening 
five today.” “Oh, ok.”



ISTR that if a problem appears in multiple supported releases, one APAR might 
result in multiple PTFs, one per release.



I’m also interested in a definitive list of APAR closings. I’ve never seen one, 
and the lists I’ve seen have been conflicting and often included different 
interpretations of the same closing.



None of this matters a ton, of course, but it *works* when done right, which is 
more than I can say for many customer problem workflows!



I don’t see this as particularly sensitive info, but perhaps IBM will disagree, 
in which case I’ll STFU.


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

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

Reply via email to