2003 (07D3) (RC2003): MQRC_BACKED_OUT
Explanation
The current unit of work encountered
an unrecoverable error or was backed out. This reason code is issued
in the following cases:
- On an MQCMIT or MQDISC call, when the commit operation fails and
the unit of work is backed out. All resources that participated in
the unit of work are returned to their state at the start of the unit
of work. The MQCMIT or MQDISC call completes with MQCC_WARNING in
this case.
- On z/OS®, this reason code occurs only for batch applications.
- On an MQGET, MQPUT, or MQPUT1 call that is operating within a
unit of work, when the unit of work already encountered an error that
prevents the unit of work from being committed (for example, when
the log space is exhausted). The application must issue the appropriate
call to back out the unit of work. (For a unit of work that is coordinated
by the queue manager, this call is the MQBACK call, although the
MQCMIT call has the same effect in these circumstances.) The MQGET,
MQPUT, or MQPUT1 call completes with MQCC_FAILED in this case.
- On z/OS, this case does not occur.
- On an asynchronous consumption callback (registered by an MQCB
call), the unit of work is backed out and the asynchronous consumer
should call MQBACK.
- On z/OS, this case does not occur.
- For the IBM® WebSphere® MQ client on HP Integrity NonStop Server using TMF,
this return code can occur:
- For MQGET, MQPUT, and MQPUT1 calls, if you have an active transaction that is being coordinated by TMF, but the IBM WebSphere MQ part of the transaction is rolled back because of inactivity on the transaction.
- If the TMF/Gateway detects that TMF is rolling back the current transaction before the application finishes with it.
Completion Code
MQCC_WARNING or MQCC_FAILED
Programmer response
Check the returns from previous calls to the queue manager. For example, a previous MQPUT call might have failed.