IBM Support

PH46414: MQ Z/OS :MQRC=2082 MQOPEN ERROR DUE TO FAILURE TO WAIT UP TO 10 SECONDS TO ALLOW ALL FULL REPOSITORIES TO RESPOND

A fix is available

Subscribe

You can track all active APARs for this component.

 

APAR status

  • Closed as program error.

Error description

  • For a QALIAS pointing to a new cluster queue, 2082
    MQRC_UNKNOWN_ALIAS_BASE_Q is received the first time the queue
    is opened. The second time, the MQOPEN works. Similarly, 2085
    MQRC_UNKNOWN_OBJECT_NAME occurs the first time a cluster queue
    is opened directly.
    
    In order for the problem with "MQRC 2082 Unknown alias base
    queue" to occur, it is necessary for the queue manager where
    the queue is being opened to be in more than one cluster. When
    the unknown queue is opened for the first time, subscription
    requests are sent to all full repositories for all clusters in
    which the queue manager is present. The queue manager then
    waits for acknowledgement of the subscription. This wait will
    initially be for one second. After this initial one-second
    wait, the ongoing MQOPEN processing will check to see if the
    subscription request has been acknowledged by a full
    repository. If no response has been received in this time then
    it will continue to wait for a maximum of 10 seconds before
    eventually reporting MQRC_CLUSTER_RESOLUTION_ERROR.
    If however, any of the full repositories acknowledge one of the
    subscription requests, then MQOPEN processing will proceed and
    we will check if the requested queue has been inserted into the
    cluster cache as a result of an update from one of the full
    repositories. If the FR that initially responds to the
    subscription request has a definition for the queue within its
    cluster then the MQOPEN will succeed. If a full repository that
    does not host any instances of the queue within its cluster is
    the first to respond and the FR that does host instances the
    queue does not respond during the initial one second wait then
    we will fail to open the queue and the problem reported by the
    customer can occur.
    

Local fix

  • N/A
    

Problem summary

  • ****************************************************************
    * USERS AFFECTED: All users of IBM MQ for z/OS Version 9       *
    *                 Release 1 Modification 0 and                 *
    *                 Release 2 Modification 0 and                 *
    *                 Release 3 Modification 0.                    *
    ****************************************************************
    * PROBLEM DESCRIPTION: MQRC=2085 (MQRC_UNKNOWN_OBJECT_NAME) or *
    *                      MQRC=2082 (MQRC_UNKNOWN_ALIAS_BASE_Q)   *
    *                      is returned from an MQOPEN of a         *
    *                      clustered queue that is being opened    *
    *                      for the first time on a clustered Queue *
    *                      Manger even though the queue exists in  *
    *                      the cluster. This occurs when the Queue *
    *                      Manager is part of multiple clusters    *
    *                      and the clustered queue only exists in  *
    *                      a subset of the clusters.               *
    ****************************************************************
    MQRC=2085 (MQRC_UNKNOWN_OBJECT_NAME) or MQRC=2082
    (MQRC_UNKNOWN_ALIAS_BASE_Q) is returned from an MQOPEN of a
    clustered queue that is being opened for the first time on a
    clustered Queue manager even though the queue exists in the
    cluster. This occurs when the Queue Manager is part of multiple
    clusters and the clustered queue only exists in a subset of the
    clusters.
    
    MQ is failing to wait up to 10 seconds for a cluster
    subscription to a clustered queue to be acknowledged by a full
    repository that has knowledge of the location of the clustered
    queue. In this case only the clustered subscriptions for the
    clustered queue in a cluster where the clustered queue does not
    exist has been acknowledged by a full repository.
    

Problem conclusion

  • The code has been changed so that when clustered subscriptions
    for a clustered queue have been created as part of the first
    MQOPEN of a clustered queue, MQ will correctly wait up to 10
    seconds for a clustered subscription to be acknowledged by a
    full repository that has knowledge of the location of the
    clustered queue. Alternatively, MQ will stop waiting if all
    clustered subscriptions are acknowledged but no information
    about the clustered queue is published to the Queue Manager
    where the MQOPEN was attempted, this is a valid case where the
    queue does not exist in the cluster.
    

Temporary fix

Comments

APAR Information

  • APAR number

    PH46414

  • Reported component name

    IBM MQ Z/OS V9

  • Reported component ID

    5655MQ900

  • Reported release

    100

  • Status

    CLOSED PER

  • PE

    NoPE

  • HIPER

    NoHIPER

  • Special Attention

    NoSpecatt / Xsystem

  • Submitted date

    2022-05-12

  • Closed date

    2023-03-20

  • Last modified date

    2023-07-03

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

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

    UI91036 UI91037 UI91038

Modules/Macros

  • CSQMZLOO
    

Fix information

  • Fixed component name

    IBM MQ Z/OS V9

  • Fixed component ID

    5655MQ900

Applicable component levels

  • R100 PSY UI91935

       UP23/06/16 P F306

  • R200 PSY UI91934

       UP23/06/16 P F306

  • R300 PSY UI91933

       UP23/06/16 P F306

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":"BU011","label":"Systems - zSystems software"},"Product":{"code":"SG19M","label":"IBM MQ"},"Platform":[{"code":"PF054","label":"z Systems"}],"Version":"100"}]

Document Information

Modified date:
03 July 2023