Shoot. I thought I was super clear. (I mean I really hoped I was clear; I don't
mean you are criticizing me unjustly.)
I understand I think the basics of what is in the I&T Reference under
Parameters for SMFPRMxx. I understand the basic concept of SYS controlling
things as a whole and being overridden on a subsystem by subsystem basis by
SUBSYS. I get EXITS/NOEXITS and TYPE/NOTYPE.
What I don't understand is the exact mapping of SUBSYS names to record-level
functionality.
Let's for argument's sake say I wanted to get every possible SMF Type 30
record, and let's say I wanted to do that entirely with SUBSYS statements. I
would code SUBSYS(xxx,EXIT(IEFU83)) and ditto for IEFU84 and IEFU85. I would
code SUBSYS(xxx,(TYPE(30)). I assume I would have to do that with xxx equal to
JESn, TSO, STC, and ASCH. Am I correct? (Or I could omit TYPE entirely and let
it default to 0:255, but let's ignore that for now.)
But let's for argument's sake say I wanted to get every possible SMF Type 80
record, and I wanted to do that entirely with SUBSYS statements. I would code
SUBSYS(xxx,EXIT(IEFU83)) and ditto for IEFU84 and IEFU85. I would code
SUBSYS(xxx,(TYPE(80)). What would xxx be? Would it be JESn and TSO and
everything else that might generate RACF records, or what? (I have some
evidence that this is the case; one would have to code every possible
RACF-affected category.)
What about SMF 100, 101, and 102 records? xxx would have to be every DB2
subsystem name?
What about SMF 119 records? xxx would be STC? Or what? Or would they be
controlled exclusively by any SYS(TYPE and EXIT specifications?
Am I making a particle of sense?
I'm not trying to come up with the world's most obscure SMFPRMxx member. I am
trying to learn to interpret the effect of what customers have coded in their
SMFPRMxx members, and trust me, many have layers and layers of SYS and SUBSYS
and TYPE this and NOTYPE that. I'm trying to understand the atomic effect of
each statement so I can say "well, you have to change X to Y."
Comprende?
Charles
-----Original Message-----
From: IBM Mainframe Discussion List [mailto:IBM-MAIN@bama.ua.edu] On Behalf Of
Lizette Koehler
Sent: Wednesday, January 04, 2012 3:41 PM
To: IBM-MAIN@bama.ua.edu
Subject: Re: Calling all experts on SMFPRMxx SUBSYS
>
>I assume this august group is filled with people who have forgotten
>more about SMFPRMxx than I will ever know.
>
>I am trying to determine the exact relationship between various SUBSYS(
>specifications and the writing of specific SMF record types. (And yes,
>I know there are OTHER variables involved, but I'm focused on this one
>variable at the moment.)
>
>I would assume that the writing of SMF Type 30 records would be
>controlled by SUBSYS(TSO and SUBSYS(JESn statements -- is that correct?
>
>The writing of SMF Type 80 records: controlled by SUBSYS(STC ? Or by
>SUBSYS for the particular type of "event" -- for example, the writing
>of TSO logon violations would be controlled by SUBSYS(TSO ?
>
>The writing SMF Type 100-102 (DB2) records: controlled by SUBSYS(xxxx
>where xxxx is the name of the particular DB2 subsystem?
>
>The writing of SMT Type 119 (TCP/IP) records: controlled by SUBSYS(what?
>
>Many thanks.
>
>Charles Mills
>
Charles,
If you are asking about the teh following control cards that can be placed in
the SMFPRMxx
SYS(...EXITS(IEFU83,IEFU84,IEFUJI)...)
SUBSYS(STC,...EXITS(IEFU83,IEFU85)...)
SUBSYS(TSO,...)
SUBSYS(JES3,...EXITS(IEFUJI)...)
According to the SMF manual for z/OS V1.11
On the SUBSYS parameter, specify the exits that are to affect work processed by
a particular SMF-defined subsystem (JES2, JES3, STC, ASCH, or TSO) and no other
subsystem specific exit points will be taken.
Is that what you are asking about? Or are you asking what functions control
the writing of SMF records?
For example SMF Type 118 and 119 are produced from the Communication Server
(VTAM). But there is no SMF SUBSYS exit that could be used for it (I think).
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
