IBM Support

SE61936: MQM400-MSGCPF7003 MQ JOURNAL CPF7003 RC4 'MAX SEQ NUMBER' HIT CAN CRASH QUEUE MANAGER

Subscribe

You can track all active APARs for this component.

 

APAR status

  • Closed as program error.

Error description

  • MQ Queue Managers use a journal for logging. OS400 journals use
    sequence numbers to track entries. The sequence number is
    occasionally reset by various operations, like OS400 IPL. MQ
    defaults to using option *MAXOPT2, which has a limit of
    9,999,999,999 for max sequence.
    
    If a Queue Manager happens to hit max sequence, the Queue
    Manager will crash. FDC will show mainly CPF7003 rc4. QHST will
    show CPF7091 rc4. A cold start is needed to restart the Queue
    Manager (which will start the joural at seq# 1)
    
    Example FDC for MQ Agent encountering maxseq when putting
    persistent message (could be hit at other times, get persistent
    message, create queue, record image etc)
    
    Date/Time :- Sunday May 10 2015
    Operating System :- OS400 V6R1M0
    PIDS :- 5724H7226
    LVLS :- 7.1.0.4
    Product Long Name :- WebSphere MQ for IBM i
    Vendor :- IBM
    Installation Path :- /QIBM/ProdData/mqm
    Probe Id :- AL028001
    Application Name :- MQM
    Component :- alsWriteLog
    SCCS Info :- lib/lqm/unix/as400/amqalwcx.c, 1.40.1.1
    Line Number :- 564
    Build Date :- Nov 7 2013
    CMVC level :- p710-004-131106
    UserID :- 00000288 (QMQM)
    Job Name :- 136156/QMQM/AMQZLAA0 (AMQZLAA0)
    Job Description :- QMQM/AMQZLAA0
    Submitted By :- 130419/QMQM/AMQZXMA0
    Activation Group :- 17 (QMQM) (QMQM/AMQZLAA0)
    Max File Handles :- 2048
    Process :- 00004524
    Thread :- 00001758
    Major Errorcode :- STOP_ALL
    Minor Errorcode :- OK
    Probe Type :- HALT6110
    Probe Severity :- 1
    Probe Description :- AMQ6110: An internal WebSphere MQ error
    has occurred.
    FDCSequenceNumber :- 0
    Arith1 :- 19 0x'13'
    Arith2 :- 264 0x'108'
    |
    IBM i Function Stack
    LE_Create_Thread2__FP12crtth_parm_t
    pthread_create_part2
    ThreadMain
    zlaCallMainThread
    zlaMainThread
    zlaProcessMessage
    zlaProcessMQIRequest
    zlaMQPUT
    zsqMQPUT
    kpiMQPUT
    kqiPutIt
    kqiPutMsgSegments
    kqiTickleEarly
    apiLockExclusive
    aqmDoRecoverQueue
    aqsDoStartQOp
    aqsRecoverQOp
    aqhRecoverQueue
    aqqRecoverQueue
    aqlLogCLR 0000000064 QMQM LIBMQML_R QMQM
    almLogIt 0000000038 QMQM LIBMQML_R QMQM
    alsWriteLog 0000000049 QMQM LIBMQML_R QMQM
    xcsFFSTSFn 0000000022 QMQM LIBMQMCS_R QMQM
    

Local fix

  • Occasionally reset the journal sequence number. This is
    automatically done with an IPL. Or, CHGJRN qmgrlib/amqajrn
    jrnrcv(*GEN) seqopt(*reset).
    

Problem summary

  • ****************************************************************
    USERS AFFECTED:
    IBM MQ on IBM i users who do not perform tasks which would allow
    for a queue manager journal sequence number to be reset.
    
    
    Platforms affected:
    IBM iSeries
    
    ****************************************************************
    PROBLEM DESCRIPTION:
    If an MQ Queue Manager Journal on IBM i is allowed to reach
    maximum journal sequence numbers, the Queue Manager may crash as
    no new journal entries will be allowed to be written.
    

Problem conclusion

  • IBM MQ code has been modified to monitor for message CPI70E7 in
    the AMQAJRNMSG message queue for the queue manager journal.
    According to IBM i Knowledge Center:
    
    The system sends a warning message (CPI70E7) to the journal's
    message queue when the sequence
    number exceeds 2 147 000 000. If you specified
    RCVSIZOPT(*MAXOPT1), RCVSIZOPT(*MAXOPT2), or RCVSIZOPT(*SYSDFT)
    for the journal that you attached the receiver to, the system
    sends the warning
    message when the sequence number exceeds 9 900 000 000. If you
    specified RCVSIZOPT(*MAXOPT3), the
    system sends the warning message when the sequence number
    exceeds 18 446 644 000 000 000 000. If you use system
    change-journal management support (MNGRCV(*SYSTEM)) for the
    journal, the system attempts to change the journal and reset the
    sequence number one time. The message is sent only if the
    attempt is not successful.
    
    Hence, if the CPI70E7 message is written to the journal message
    queue, then MQ will attempt to perform a CHGJRN JRNRCV(*GEN)
    SEQOPT(*RESET)
    
    However, users should be aware of warning messages in QSYSOPR,
    such as CPF7019, which are indicative of the journal approaching
    maximum sequence numbers, and take appropriate measures.
    
    Actions that will reset the journal sequence number include:
    CHGJRN JRNRCV(*GEN) SEQOPT(*RESET)
    An IPL of the system
    Vary on of the independent disk pool containing the AMQARN
    journal, if applicable.
    
    Refer to the IBM i Knowledge Center for more information on
    journal maintenance.
    
    Additionally, this APAR changes the default for the Queue
    Manager Journal Receiver size options (RCVSIZOPT) from MAXOPT2
    to MAXOPT3 at IBM MQ V8. This will automatically be applied to
    all queue managers created after this APAR has been applied.
    Existing queue managers can also be modified to use MAXOPT3, if
    needed.
    
    Previous versions of MQ will continue to utilize MAXOPT2.
    MAXOPT3 will not be supported at pre-V8 releases.
    
    MAXOPT3 provides a maximum sequence number of 18 446 744 073 709
    551 600 vs 9 999 999 999 for MAXOPT2.
    
    ---------------------------------------------------------------
    The fix is targeted for delivery in the following PTFs:
    
    Version    Maintenance Level
    v7.1       7.1.0.7
    v8.0       8.0.0.4
    
    The latest available maintenance can be obtained from
    'WebSphere MQ Recommended Fixes'
    http://www-1.ibm.com/support/docview.wss?rs=171&uid=swg27006037
    
    If the maintenance level is not yet available information on
    its planned availability can be found in 'WebSphere MQ
    Planned Maintenance Release Dates'
    http://www-1.ibm.com/support/docview.wss?rs=171&uid=swg27006309
    ---------------------------------------------------------------
    

Temporary fix

Comments

APAR Information

  • APAR number

    SE61936

  • Reported component name

    WMQ ISERIES V7

  • Reported component ID

    5724H7226

  • Reported release

    710

  • Status

    CLOSED PER

  • PE

    NoPE

  • HIPER

    NoHIPER

  • Special Attention

    NoSpecatt / Xsystem

  • Submitted date

    2015-05-13

  • Closed date

    2015-06-26

  • Last modified date

    2015-07-02

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

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

Modules/Macros

  • RCHMGR
    

Fix information

  • Fixed component name

    WMQ ISERIES V7

  • Fixed component ID

    5724H7226

Applicable component levels

[{"Business Unit":{"code":"BU048","label":"IBM Software"},"Product":{"code":"SSCPQ63","label":"APAR \/ Maintenance"},"Platform":[{"code":"PF025","label":"Platform Independent"}],"Line of Business":{"code":"","label":""}}]

Document Information

Modified date:
27 April 2020