I routinely counsel people not to freak out over RC 8, but that advice is
based on how SMPE actually works, not on how I would like it to work. Kurt
has opened up a review of the process. Assuming that the latest HOLDDATA has
been pulled just before APPLY--best practice--then unresolved hold errors
are par for the course and probably deserve only a 4. There is nothing for
the user to do except to re-verify that yup, a needed PTF is not available.
Unless it's available but not RECEIVEd. The cases where user action is
mandatory involve errors such as

-- Insufficient space (allocation or directory blocks)
-- Missing DDDEF
-- Actual compile or bind error

In these situations, RC 8 is certainly in order. Nonetheless, I maintain
that examination of the CAUSER report will quickly distinguish the oh-darn
cases from the oh-sh*ts. 

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

> -----Original Message-----
> From: IBM Mainframe Discussion List [mailto:[email protected]]
> On Behalf Of Pommier, Rex
> Sent: Wednesday, December 23, 2015 09:27 AM
> To: [email protected]
> Subject: [Bulk] Re: PTF error clarification
> 
> Chris,
> 
> I was having similar thoughts as you, although not to the extent of yours.
I tend
> to think along the first of your ideas about the purpose of the RC, that
being to
> indicate the success of the process doing what it is supposed to do.  But
I would
> disagree with your idea of everything being RC=0.  In my mind, if SMP/E
decides
> a PTF shouldn't go on because of a hold (whether error hold or other
reason like
> ACTION), it shouldn't throw a RC=8, because that is SMP/E doing what it
should.
> However, if a utility fails, then SMP/E did NOT do what it was supposed to
do,
> that is install the software or maintenance or whatever.  In any case of
SMP/E
> deciding that something should APPLY/ACCEPT/whatever, and it doesn't, a
RC=8
> (or higher) is warranted.
> 
> Maybe in way of compromise, SMP/E should set a RC=6 instead of 8 where
> maintenance is stopped due to an error hold.  :-)
> 
> Rex
> 
> -----Original Message-----
> From: IBM Mainframe Discussion List [mailto:[email protected]]
> On Behalf Of Chris Hoelscher
> Sent: Wednesday, December 23, 2015 10:12 AM
> To: [email protected]
> Subject: Re: PTF error clarification
> 
> I guess I have a basic question - is a return code intended to indicate
that
> success of a process doing what it is supposed to do? Or the
meaningfulness of
> the results of the process to the person who makes uses of the results of
the
> process?
> 
> If the former - then should EVERY run of APPLY return a RC=0 unless there
is
> malformed input where SMP/e could NOT do what it is supposed to do (on the
> assumption of well-formed input)
> 
> If the latter - then are return codes ever meaningful - every result set
should be
> reviewed with equal diligence without regard to the RC (smp/e can't guess
what
> is acceptable or meaningful to YOU)
> 
> If seems that in either case - the RC is not as important as the results;
does a RC-
> 9 mean that SMP/E did what YOU wanted it to? Or just that the process
ended
> without what SMP/E considers to be problems ...
> 
> 
> Back to my hole ....
> 
> Chris Hoelscher
> Technology Architect, Database Infrastructure Services Technology Solution
> Services
> : humana.com
> 123 East Main Street
> Louisville, KY 40202
> Humana.com
> (502) 714-8615, (502) 476-2538

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

Reply via email to