PK98544: MDB WITH BEAN MANAGED TRANSACTION AND USING MESSAGE LISTENER SERVICE IN NON-ASF MODE DOESNOT REQUEUE BACKOUT MESSAGES.

Fixes are available

APAR status

Closed as program error.

Error description

Backout message queue handling not properly working in non-ASF
mode for message-driven beans using bean-managed transactions
(BMT) with the NON.ASF.BMT.ROLLBACK.ENABLED message listener
service custom property enabled.

In the BMT case with NON.ASF.BMT.ROLLBACK.ENABLED set there
seems to be a need for another successful message to be
processed by the message-driven bean, before the failing
message is re-queued to the backout queue. If no message
arrives, the message is never committed to the backout queue,
and the uncommitted status of both the source queue and
backout queue is UNCOM(YES).

Whereas in the container-managed transaction (CMT) case
messages requeued to their backout queues without another
message being successfully processed.
However, there might be a delay of up to the configured
NON.ASF.RECEIVE.TIMEOUT timeout before the message arrives.

In CMT case the only mechanism available to reduce the amount
of time it takes to move messages to the backout queue or
exception destination in non-ASF mode is to reduce the
NON.ASF.RECEIVE.TIMEOUT.
For example, setting this to 1000 milliseconds would mean in
most scenarios  it takes a up to 1 second to move messages to
the backout queue after a rollback.

Local fix

use ASF mode or use CMT with acceptable NON.ASF.RECEIVE.TIMEOUT

Problem summary

****************************************************************
* USERS AFFECTED:  Users of IBM WebSphere Application Server   *
*                  Version 6.1 or Version 7.0, using           *
*                  WebSphere MQ as a messaging provider        *
****************************************************************
* PROBLEM DESCRIPTION: Messages not delivered to the           *
*                      WebSphere MQ backout queue for          *
*                      a message-driven bean using             *
*                      bean-managed transactions and a         *
*                      non-ASF message listener port           *
****************************************************************
* RECOMMENDATION:                                              *
****************************************************************
A delay moving messages to the backout queue when in non-ASF
mode, up to the configured non-ASF receive timout
(NON.ASF.RECEIVE.TIMEOUT), is expected for all transaction
modes. This is due the fact that the delivery of messages to
the backout queue does not occur as part of the rollback.
Instead, it occurs silently within the WebSphere MQ JMS client
the next time a non-ASF receive thread calls attempts to
receive a message from the queue.

The move of the invalid message (that has reached the backout
threshold) to the backout queue occurs within the same unit of
work as the synchronous receive that is looking for new
messages to pass to the onMessage method of the message-driven
bean. When there are no valid messages available on the queue
(only invalid messages that have already reached the backout
threshold) the synchronous receive returns null to the non-ASF
listener after the timeout configured in the
NON.ASF.RECEIVE.TIMEOUT property has been reached.

The move of the message to the backout queue remains indoubt
within WebSphere MQ until the unit of work is completed.

In the container-managed transaction (CMT) case, this occurs
successfully because the container commits the transaction that
includes the WebSphere MQ unit of work, even if the receive
call returned null (and the MDB was not called).

However, if bean-managed transactions are used with a non-ASF
listener port and the NON.ASF.BMT.ROLLBACK.ENABLED property
enabled, then the JMS client-acknowledgment interface
currently used does not provide any mechanism to commit the
unit of work under which the message was moved to the backout
queue (which is internal to the WebSphere MQ JMS client).

This leaves the unit of work indoubt indefinitely, until
another message arrives on the queue and is processed
successfully by the message-driven bean, on the same non-ASF
receive thread where WebSphere MQ performed the backout requeue.

Problem conclusion

The fix for this APAR changes the internal implementation of
the BMT rollback behavior (enabled with the
NON.ASF.BMT.ROLLBACK.ENABLED property) to use a
transacted JMS session, rather than client-acknowledgment, to
complete or rollback consumption of messages
As a result, message are correctly moved to the backout queue
within the non-ASF receive timeout specified by the
NON.ASF.RECEIVE.TIMEOUT property.

The fix for this APAR is currently targeted for inclusion in
fix packs 6.1.0.31 and 7.0.0.9.  Please refer to the
Recommended Updates page for delivery information:
http://www.ibm.com/support/docview.wss?rs=180&uid=swg27004980

Temporary fix

Comments

APAR Information

APAR number
PK98544
Reported component name
WEBS APP SERV N
Reported component ID
5724H8800
Reported release
61S
Status
CLOSED PER
PE
NoPE
HIPER
NoHIPER
Special Attention
NoSpecatt
Submitted date
2009-10-14
Closed date
2009-11-16
Last modified date
2009-11-16

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

Fix information

Fixed component name
WEBS APP SERV N
Fixed component ID
5724H8800

Applicable component levels

R61A PSY
UP
R61H PSY
UP
R61I PSY
UP
R61P PSY
UP
R61S PSY
UP
R61W PSY
UP
R61Z PSY
UP
R700 PSY
UP

[{"Business Unit":{"code":"BU059","label":"IBM Software w\/o TPS"},"Product":{"code":"SSEQTP","label":"WebSphere Application Server"},"Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"6.1","Line of Business":{"code":"LOB45","label":"Automation"}}]

Document Information

Modified date:
29 December 2021

Tips

PK98544: MDB WITH BEAN MANAGED TRANSACTION AND USING MESSAGE LISTENER SERVICE IN NON-ASF MODE DOESNOT REQUEUE BACKOUT MESSAGES.

Fixes are available

Subscribe

APAR status

Closed as program error.

Error description

Local fix

Problem summary

Problem conclusion

Temporary fix

Comments

APAR Information

APAR number

Reported component name

Reported component ID

Reported release

Status

PE

HIPER

Special Attention

Submitted date

Closed date

Last modified date

APAR is sysrouted FROM one or more of the following:

APAR is sysrouted TO one or more of the following:

Fix information

Fixed component name

Fixed component ID

Applicable component levels

R61A PSY

R61H PSY

R61I PSY

R61P PSY

R61S PSY

R61W PSY

R61Z PSY

R700 PSY

Document Information

Share your feedback

Need support?