Just a stab but maybe the below applies....

Record Type 7 (07) - Data Lost

z/OS V1R12.0 MVS System Management Facilities (SMF)
SA22-7630-22 

 
 

If all SMF buffers become full (either a result of no available output data 
sets for SMF to write to OR the system generating records at a rate faster than 
SMF can physically write them), SMF data will be lost. When this condition 
occurs, record type 7 tracks the number of lost records. It contains a count of 
the SMF records that were not written and the start and end times of the period 
when data was lost. (The end time is the time recorded in SMF7TME at offset 6).

Record type 7 is not built until SMF buffers become available again. Data 
existing in the SMF buffer, prior to data lost, is written to available SMF 
data sets before record type 7 is written to a data set.



-----Original Message-----
From: IBM Mainframe Discussion List [mailto:[email protected]] On Behalf 
Of Greg Shirey
Sent: Wednesday, August 07, 2013 12:33 PM
To: [email protected]
Subject: RACF User ID resumed without an SMF record?

Hello group,

Does anyone know of a method to resume a RACF revoked ID without having an SMF 
record be written?  

We produce a daily listing of RACF commands from our SMF type 80s (using 
RACFRW) and we list ADDUSER ADDGROUP ALTUSER ALTGROUP CONNECT DELUSER DELGROUP 
PASSWORD PERMIT RALTER RDEFINE REMOVE.  

We also produce a daily listing of our CICS user IDs and their RACF status.  On 
July 8 we had a user ID on our report that was listed as REVOKED and a 
LAST-ACCESS date and time of 07/17/07 17:01:28. 

On July 9, the report showed the ID was no longer revoked and the LAST-ACCESS 
reported as 07/08/13   19:24:14.  However, our SMF report listed no ALTUSER 
command or any other command against this ID.  (No DELUSER or ADDUSER, for 
instance).  

I dumped the SMF records for both July 7 and July 8 and ran a RACFRW to list 
all the records and there is no reference to this User ID.   

I'm a sysprog, so I can't blame it on magic or elves - I could try blaming it 
on the software, but I'm finding that hard to believe - so I have to think 
there's something I'm missing.  I've just looked at everything I know to look 
at.  (Did someone modify SMF for a period?  No.  Does the COBOL program that 
lists the RACF users have a bug in it?  No.)  

If anyone has a suggestion for what to look for, I'd appreciate hearing about 
it.   

Thanks,
Greg Shirey
Ben E. Keith Company 

----------------------------------------------------------------------
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