IBM Support

PK99102: EYUCL0105E ESSS RECEIVE LINK TASK TERMINATED ABNORMALLY COI0

A fix is available

Subscribe

You can track all active APARs for this component.

 

APAR status

  • Closed as program error.

Error description

  • A MAS stayed up for several months.
    Then, without previous errors, it got:
    .
    EYUCL0105E ESSS Receive Link Task terminated abnormally.
    EYUNL0905E An unrecoverable error has occurred.
            The MAS agent is being terminated.
    EYUXZ0910I EYU0XZSD Dump ... LMAS,COI0,0000079,TRAC,EYU0NLSD...
    EYUCL0005I ESSS Receive Link Task terminated.
    .
    The MAS agent does come back up automatically.
    .
    Two trace entries that were taken for the error:
    .
    Task:79 ...
        Method:CTES   Prior:CLEV  Debug:EXCEPT    Point-Id:8
     Major-Obj:COM    Envrn:LMAS   Type:EXCEPTION  Tran_Id:COI0
     CICS-Task:COM                 CMAS:
    Keyword           Data Queue   Req Data      Data
    Value             Type Dir     Opt Address   Value
       In: *FUNCTION     FUN  .  1D2A0C28  ENDSINK
      Out: *RESPONSE     RSP  .  1D2A0C2A  EXCEPTION
           *REASON       RSN  .  1D2A0C2B  METHOD_FAILED
    .
    Task:79
        Method:CLEV   Prior:XLOP  Debug:EXCEPT    Point-Id:3
     Major-Obj:COM    Envrn:LMAS   Type:EXCEPTION  Tran_Id:COI0
     CICS-Task:COM                 CMAS:
    Keyword           Data Queue   Req Data      Data
    Value             Type Dir     Opt Address   Value
       In: *FUNCTION       FUN  .  1D2A1D23  MASRLT
      Out: *RESPONSE       RSP  .  1D2A1D25  00
           *REASON         RSN  .  1D2A1D26  00
    .
      Exception Point-ID number 8 in module DFH0CTES
    happens in label CTES_PCKT_TYPE:
    .
    *-----------------------------------------------------------*
    *   Format MALRL Element and add to the MALRL List          *
    *-----------------------------------------------------------*
    ...
        EYUQXCLA CLEAR,
              ...
              FUNCTION(ADDCLST),
              DEBUG('CTESXCLA'),
              ELEMENT(#WRK_MALRL,CMRL_SIZE,*),
              LIST_ID(TRNSPT_MASRL),
              ...
        LA    R4,CMPI_XCPT_XCLA          Cache List Add Exception
    .
    This is the method that failed.  It failed because
    the transmision counter in the CMAS overflowed
    and restarted from 0.
      This counter is used to build the key for MAxRL elements.
    If it overflows and restarts, then we risk it not being unique.
    EYUQXCLA failed with a duplicate key status
    .
      The transmision counter overflowed
    because both the CMAS and the MAS had been running continuously
    and were connected to each other for over five months.
      The wrapping will only be a problem for a request
    that would stay around for the life of the MAS.
    There is one such request
    - the request to start the heartbeat task
    (method EYU0NLHD running under a COIE task).
    This APAR is being taken to investigate whether
    we can do something about that
    and eliminate the problem when a wrap occurs.
    Additional Symptom(s) Search Keyword(s):
    KIXREVRBD
    

Local fix

Problem summary

  • ****************************************************************
    * USERS AFFECTED: All CICSPlex SM V4R1M0 Users                 *
    ****************************************************************
    * PROBLEM DESCRIPTION: If a MAS has been connected to a CMAS   *
    *                      for a long period of time (generally a  *
    *                      number of months), the MAS agent may    *
    *                      terminate with messages EYUCL0105E and  *
    *                      EYUNL0905E:                             *
    *                                                              *
    *                        EYUCL0105E ESSS Receive Link Task     *
    *                                   terminated abnormally.     *
    *                                                              *
    *                        EYUNL0905E An unrecoverable error has *
    *                                   occurred.  The MAS agent   *
    *                                   is being terminated.       *
    *                                                              *
    *                      After the messages are issued, a dump   *
    *                      will be taken.  The title of the dump   *
    *                      will be similar to the following:       *
    *                                                              *
    *                        EYU0XZSD Dump,jobname,masname,lpar,   *
    *                                 LMAS,COI0,tasknum,TRAC,      *
    *                                 EYU0NLSD,mm/dd/yy,hh:mm:ss   *
    *                                                              *
    *                      Examination of the MAS's trace records, *
    *                      either in the auxtrace datasets or the  *
    *                      dump, will show the following exception *
    *                      trace entries from the time of the      *
    *                      messages:                               *
    *                                                              *
    *                           Method  TPID  Debug text           *
    *                           ------  ----  ----------           *
    *                           CTES       8  EXCEPT               *
    *                           CLEV       3  EXCEPT               *
    ****************************************************************
    * RECOMMENDATION: After applying the PTF that resolves this    *
    *                 APAR, all MASes must be restarted.  Note     *
    *                 that the restarts do not need to occur at    *
    *                 the same time.                               *
    ****************************************************************
    When a CMAS sends a request to a MAS, method EYU0CTES (CTES)
    running in the MAS, places an element in its MAS response list
    (CMRL) for the request.  The element will be removed from the
    CMRL when the request ends.
    
    The key for the list element, called the packet set ID (PSID),
    is the four byte SYSIDNT of the CMAS followed by a four byte
    numeric field.  The numeric field is a counter maintained by the
    CMAS.  It starts with a value of one (1), and is incremented by
    one (1) for each request the CMAS sends.  When the counter
    reaches 2,147,483,647, it will wrap to zero (0) for the next
    request, and continue to be incremented for subsequent requests.
    
    When a MAS starts, the CMAS sends a request to the MAS to start
    the hearbeat task (method EYU0NLHD (NLHD) running under
    transaction COIE).  Since this task runs for the life of the
    MAS, its CMRL element will remain until termination.
    
    If the counter wraps while the MAS stays connected to the CMAS,
    it is possible that a subsequent request sent to the MAS will
    have a PSID with the same four byte numeric value as the PSID of
    the hearbeat task.  When this occurs, CTES will fail while
    adding a duplicate key to the CMRL, resulting in the errors
    documented above.
    

Problem conclusion

  • CTES has been updated to turn on the high order bit in the
    numeric portion of the PSID before adding the CMRL element for
    the heartbeat task.  Since the hearbeat task is only sent to
    a MAS once during its run, and since no other CMRL entries
    last over the entire run of a MAS, this will ensure that CTES
    does not encounter a duplicate key while adding CMRL elements.
    

Temporary fix

  • FIX AVAILABLE BY PTF ONLY
    

Comments

APAR Information

  • APAR number

    PK99102

  • Reported component name

    CICS TS Z/OS V4

  • Reported component ID

    5655S9700

  • Reported release

    60M

  • Status

    CLOSED PER

  • PE

    NoPE

  • HIPER

    NoHIPER

  • Special Attention

    NoSpecatt

  • Submitted date

    2009-10-20

  • Closed date

    2009-10-22

  • Last modified date

    2009-11-04

  • APAR is sysrouted FROM one or more of the following:

    PK97508

  • APAR is sysrouted TO one or more of the following:

    UK51222

Modules/Macros

  • EYU0CTES
    

Fix information

  • Fixed component name

    CICS TS Z/OS V4

  • Fixed component ID

    5655S9700

Applicable component levels

  • R60M PSY UK51222

       UP09/10/24 P F910

Fix is available

  • Select the PTF appropriate for your component level. You will be required to sign in. Distribution on physical media is not available in all countries.

[{"Business Unit":{"code":"BU058","label":"IBM Infrastructure w\/TPS"},"Product":{"code":"SSGMGV","label":"CICS Transaction Server"},"Component":"","ARM Category":[],"Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"4.1","Edition":"","Line of Business":{"code":"LOB35","label":"Mainframe SW"}},{"Business Unit":{"code":"BU054","label":"Systems w\/TPS"},"Product":{"code":"SG19M","label":"APARs - z\/OS environment"},"Component":"","ARM Category":[],"Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"4.1","Edition":"","Line of Business":{"code":"","label":""}}]

Document Information

Modified date:
04 November 2009