If you are using ACF2 you can set LOG to the program on the PROGRAM class, I 
think RACF doesn’t log FASTAUTH requests but ACF2 does.

T R(PGM)
RECKEY myprog ADD(UID(*) LOG)

We actually use a homegrown ICHRTX00 exit to log program usage which works very 
well and it's "free" but like Charles said, it's not for the faint-hearted.

Regards,
Leo



-----Original Message-----
From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf 
Of Charles Mills
Sent: Wednesday, November 13, 2019 11:30 AM
To: IBM-MAIN@LISTSERV.UA.EDU
Subject: Re: Tracking called (non-main) programs using SMF records

Similar topics have been kicked around here several times. You are basically 
correct -- SMF will not do the job you want it to do.

SMF 30 reports the "high CPU usage" program of the jobstep. That might be a 
different program than the jobstep program. It's an additional clue, but it is 
not the solution to your query.

I think the conclusion here was that there were some third-party products that 
do this. IIRC there is a CSV exit point that audits LOADs, but writing exits is 
not for the faint-hearted, and performance impact is a concern.

There is always the "you bet your job" approach. Move any questionable programs 
into a shadow copy of the load library and wait for the screams.

Charles


-----Original Message-----
From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf 
Of Steff Gladstone
Sent: Wednesday, November 13, 2019 8:06 AM
To: IBM-MAIN@LISTSERV.UA.EDU
Subject: Tracking called (non-main) programs using SMF records

We would like to clean up our load libraries by deleting unused programs.

Is there any way to use SMF data to track real-time usage of programs which
are not main programs but are called by other programs?

We were under the impression  that only executions of main programs (PGM=)
 were recorded with SMF.  Scanning program sources for called programs is
unsatisfactory for us since the call could be dynamic (name of program
contained in a variable) or conditional on a particular set of
circumstances that never occurs in practice.

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Reply via email to