IBM Support

OA57267: AFTER PUTTING CF INTO MAINTMODE, LOCK STR IN ASYNC SM DUPLEX WAS UNDERGOING REBUILD AND HANG. IXC571I QUIESCE

A fix is available

Subscribe

You can track all active APARs for this component.

 

APAR status

  • Closed as program error.

Error description

  • Customer has 3 CFs (i.e CF1, CF2 and CF3) in configuration.
    While a LOCK str was in SM Async Duplex mode between CF1 and
    CF2, CF2
    was put into MAINTMODE, then SETXCF START,REALLOCATE was issued
    to perform rebuild. The reallocate processing was hang after
    IXC571I SYSTEM-MANAGED DUPLEXING REBUILD FOR STRUCTURE
    xxx_LOCK1 HAS COMPLETED THE STARTUP PHASE        AND IS ENTERING
    THE QUIESCE PHASE.
    D XCF,STR,STRNM=xxx_LOCK1  output shows the connector with
    ..
     * ASTERISK DENOTES OUTSTANDING REBUILD PROCESSI
     CONNECTION NAME  ID VERSION  SYSNAME  JOBNAME
     ---------------- -- -------- -------- --------
    *DXRDD2L$$IDL1001 02 000200BA D005     DDL1IRLM
     DXRDD2L$$IDL2002 01 000100B1 D004     DDL2IRLM  This REALLOCATE
    processing must be stopped via SETXCF STOP,REALLOCATE  then
    SETXCF STOP,REALLOCATE,FORCE
    There was no IXL041E as the rebuild was not hang waiting for
    event respond.
    Console dump of this connector address space and all the
    dataspaces should be taken prior canceling this connector.
    This connector must be canceled before this LOCK
    str can be establishing its SM Duplex between CF1 and CF3.
    
    VERIFICATION STEPS:
    From the console dump of the connector address space, issue :
    XESDATA CONNECTION STR(XXX_LOCK1) DETAIL,
    
    It shows :
     System-managed duplexing rebuild is in progress
     Duplexing protocol.......... (Could be Synchronous or
    Asynchronous)
     Auto Version................ ...
     Connection is quiesced for a system-managed pro
     System-managed quiesce phase completed
     ************************
    ..
       Global Data Space   Name: nnnnnIXL Alet:
    ..
    Connection Related Recovery in Progress:
      Diag008: 000530B0 nnnnnIXL
      Requests waiting for Async Completion... 0
      Auto Quiesce Recovery Processing:
         Connection Identifier.. 00
         Connection Version..... 00000000
         Recovery Function...... 001C       <-
         Auto Quiesce Sequence.. 000000nn
         Event Sequence......... 000000nn
         Auto Quiesce Indicated. Yes
                 ***
         Resume After Quiesce... Yes
         Auto Version........... ,,,,,,,, ,,,,,,,,
         Diag009................ 00125028 nnnnnIXL
    ..
    Browse the storage in the dump using 10%+8C?+9C?+FC to find the
    address at this location (ie, say it is xxxxxxxx), Then issue :
    L xxxxxxxx+34 asid(aa) DSP(nnnnnIXL) ( where the aa in the asid
    is the decimal asid# of this lock str connecntor).
    The first byte at this storage location should have x04.
    

Local fix

  • BYPASS/CIRCUMVENTION:
    No local fix is available
    
    RECOVERY ACTION:
    Taking the console dump of the hang connector address space and
    dataspaces, then cancel this
    connector address space.
    

Problem summary

  • ****************************************************************
    * USERS AFFECTED:                                              *
    * Users who are running in a Parallel Sysplex                  *
    * environment making use of system-managed                     *
    * asynchronously duplexed coupling facility                    *
    * (CF) lock structures for sysplex-scope                       *
    * serialization functions.                                     *
    ****************************************************************
    * PROBLEM DESCRIPTION:                                         *
    * Due to a programming error, when a                           *
    * system-managed asynchronously duplexed                       *
    * structure is stopped to switch to                            *
    * the new structure, subsequent                                *
    * system-managed rebuild                                       *
    * processing may hang while                                    *
    * attempting to quiesce                                        *
    * activity to the structure.                                   *
    *                                                              *
    * SYSPLEXDS                                                    *
    ****************************************************************
    * RECOMMENDATION:                                              *
    * PTFs should be applied to all systems in the                 *
    * sysplex via rolling IPL.                                     *
    ****************************************************************
    Due to a programming error, an internal XES indication used to
    indicate the structure is quiesced may not be properly reset
    when a system-managed asynchronously duplexed structure is
    stopped to switch to the new structure.  This timing
    dependent problem prevents subsequent system-managed rebuild
    processing to start a rebuild from progressing past
    the quiesce phase as indicated by MSGIXC571I.  Displaying
    structure status via the D XCF,STR,STRNAME=xxx command shows
    a connector with outstanding rebuild processing in MSGIXC360I.
    
    While attempts to stop the rebuild in this state may cause
    MSGIXC367I to be issued indicating the rebuild request has been
    accepted, no further progress is made to stop the rebuild due to
    the hang condition.
    

Problem conclusion

  • Modifications made to ensure internal XES indication used to
    indicate the structure is quiesced gets properly reset after a
    system-managed asynchronously duplexed structure is stopped to
    switch to the new structure.  This prevents the hang
    condition and allows subsequent system-managed rebuild
    processing that starts or stops a rebuild to properly
    complete.
    

Temporary fix

  • *********
    * HIPER *
    *********
    

Comments

APAR Information

  • APAR number

    OA57267

  • Reported component name

    CROSS SYS.EXT.S

  • Reported component ID

    5752SCIXL

  • Reported release

    7A0

  • Status

    CLOSED PER

  • PE

    NoPE

  • HIPER

    YesHIPER

  • Special Attention

    NoSpecatt / Xsystem

  • Submitted date

    2019-04-05

  • Closed date

    2019-09-12

  • Last modified date

    2019-10-01

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

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

    UJ00787 UJ00789 UJ00790

Modules/Macros

  • IXLR2PLA
    

Fix information

  • Fixed component name

    CROSS SYS.EXT.S

  • Fixed component ID

    5752SCIXL

Applicable component levels

  • R7A0 PSY UJ00790

       UP19/09/25 P F909 ¢

  • R7B0 PSY UJ00787

       UP19/09/25 P F909 ¢

  • R7C0 PSY UJ00789

       UP19/09/25 P F909 ¢

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":"BU054","label":"Systems w\/TPS"},"Product":{"code":"SG19M","label":"APARs - z\/OS environment"},"Component":"","ARM Category":[],"Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"7A0","Edition":"","Line of Business":{"code":"","label":""}},{"Business Unit":{"code":null,"label":null},"Product":{"code":"SG19O","label":"APARs - MVS environment"},"Component":"","ARM Category":[],"Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"7A0","Edition":"","Line of Business":{"code":"","label":""}}]

Document Information

Modified date:
01 October 2019