SMF is back!

[sysprog]

The SMF is finally back again.
The SMF dump datasets are in LOG.SMF.*.
The parameters may be tuned in next few days. Some SMF types may not be recorded.

[steve-myers]

I've just reviewed the SMF data, both LOG.SMF(0) (G0003V00) and the live SMF as of about noon system time Apr 6.

I'd like to suggest the addition of types 14 and 15 so we can see who is accessing data if necessary.

On the other hand, I'm not sure type 42 is very useful in our environment, and it is the largest space user at present.

[steve-myers]

Thanks for adding type 14 & 15 records.

I'm working on a program to print some information from them, and I've run into a problem. Most likely it's my program, but a lot of datasets seem to be missing. Datasets I know I was using, both personal and system, in the period from 9:30 April 11 through 9:30 April 14, are simply not there.

Like I said, it's most likely my program, but I can't see the problem.

[steve-myers]

Yep, it was my program. Fairly dumb mistake, too, but it was hard to localize. There are a couple of SYSFAN datasets being updated by users, but I'm not sure this is a concern.

[prino]

The 'SYSFAN.USER.*' libs are OK, anything else should be off-limits, except for a few people with somewhat higher authority.

[steve-myers]

SYSFAN.CICS.LOAD, SYSFAN.CICS.SDFHMAC