Expand all | Collapse all

Capturing CA-7 Abends in CA-OPS/MVS

Jump to Best Answer
  • 1.  Capturing CA-7 Abends in CA-OPS/MVS

    Posted 05-01-2015 10:05 AM

    Currently we have a copy of CA-7 running on an LPAR that is kind of our sandbox/maintenance partition. It is generally used for our systems programmers to test things out and the first LPAr to get the new maintenance so as you can imagine there isn't much being scheduled through CA-7 and thus very little monitoring. To that end I was hoping to write a rule to capture abend messages via the CA7BRWSE interface and notify via e-mail the system programmers since they are really the only ones that would be scheduling jobs in the first place. Here is what I have/know so far:


    A. There are two message ID's that appear to be abend messages from CA-7, SCRJ-14 and SMF0-19. The information in the manual for SCRJ-14 points me to SMF0-19 and SMF0-19 is the only message I see in the OPSLOG on our production LPAR which I can guarentee has abends. So I am looking at SMF0-19. Of course if anyone can tell me what happened to SCRJ-14 I'd appreciate it because I only some minor formatting differences between these two messages and to have both messages with the same information seems redundant.


    B. SMF0-19 is a two line message but the second line of the message does not use the same message ID. The second line is under CA7BRWSE message ID. I do not see any connection between the two messages except that one immediately follows the other in the log.


    C. I found a web site or information in a manual the other day that mentioned that to get both lines for the abend, you would need to code two rules. Unfortunately, I can't remember where I found it and of course forgot to bookmark it.


    So here is my question:


    Because I will essentially need two rules to process one message ID (annoying, CA you may want to fix that), how do I ensure that the CA7BRWSE message I am processing is the one attached to the SMF0-19 message I originally processed?


    P.S. There is also a SCRJ-13 message ID indicating a JCL error which I will be processing as well but it only has one line.

  • 2.  Re: Capturing CA-7 Abends in CA-OPS/MVS

    Posted 05-04-2015 04:35 PM

    Is anyone able to offer any guidance on this question?


    Thank you

  • 3.  Re: Capturing CA-7 Abends in CA-OPS/MVS
    Best Answer

    Posted 05-05-2015 12:42 PM

    Because of the internal delivery mechanism of CA-7 browse messages issued within the original method of CA-7 msgs=>ENF=>OPS/MVS , subsequent 'data lines' for all messages begin with CA7BRWSE. So for the CA-7 messages that were multi segmented , you'll need a )MSG xxxx (such as )MSG SMF0-19, ) and )MSG CA7BRWSE...Adding logic in each rule to 'communicate' between the events. Refer to your opsmvshlq.CCLXRULS DSN. There are two samples that will assist you with this effort. Members are CA7ABEND, and CA7BRWSE. Both are fully commented with details of how they work.


    If you are setting up CA-7 automation for the first time within OPS/MVS  (or looking to improve overall efficiency of existing) and you are at CA-7 r11.3 or higher, you may want to look at setting up the CA7 Master Station Message Routing (MSMR) to directly issue internal OPS/MVS API events (drives an OPS/MVS )API CA7* rule). This method of getting CA7 Browse messages to OPS/MVS to process, eliminates ENF queueing logic as well as the need to set CA7BRWSE for message id of the segmented lines (data lines will have same msg.id as primary just like a regular MLWTO). Its also more efficient in that only the 'desired' CA7 Browse messages specified on the CA7 MSMR will be forwarded, unlike the original CA7=>ENF=>OPS/MVS delivery method which sends 'all' CA-7 msgs.


    Refer to the CA-7 doc for complete details on setting up the MSMR control file (Systems Programmers Guide) to cause CA7 to create API events.