CA Endevor concurrent processing job abends with a s213
search cancel

CA Endevor concurrent processing job abends with a s213

book

Article ID: 14771

calendar_today

Updated On:

Products

Endevor Endevor Natural Integration Endevor - ECLIPSE Plugin Endevor - Enterprise Workbench

Issue/Introduction



Running CAP (Concurrent Action Processing) I will get sporatically a S213-74.  If the job abends we usually just resubmit
the job and it will run to completion.  Lately we get this error but the restart does not work.  We have implemented the
KB article TEC481968 - How to avoid an S213-74 abend when using a PDS/e in a processor, but the CAP still abends.  

How can I fix the problem?

Environment

Release:
Component: ENDBAS

Resolution

The CAP should be running on one LPAR not on multiple LPARS.  

The following is the analysis that was done by Support:

    XCF structures found in the dump as well as the S213-74 abend itself suggested that
    the use of extended PDS/E between two systems  SYSAA and SYSBB is being done.

    As per the output provided, the processrs involved contain a single CONWRITE step that copies  the element
    into a common library with FOOTPRNT=CREATE.  That library raises the S213-74 after CONWRITE has 
    completed (that is, while we are creating the footrpint).

    Also, the outputs provided shows that the process is sometimes executed in SYSAA and sometimes in SYSBB

    Altogether this is problematic because:

    -  We need to open (connect to) the member in update mode in order to replace it's directory entry when
       footprinting the member.  That's what the IBM documentation calls - update in-place.

    -  PDS/E extended sharing at member level is good when several processes are creating members.  It is very
       restrictive for update in-place.  The IBM documentation seems to suggest that when update in-place is done
       from one system, opens from the other systems in the plex will get the S213-74.

    -  Due to a simple processor running, we are updating in-place may members per second which increases the chances
       of the abend.

    Coding DISP=OLD in the processor DD (together with correct GRS setting for CTLIPROC major name) as
    suggested in TEC481968 will also clear the abends (unless any non-Endevor process is updating the library) 
    as Endevor will single-thread updating the library.