Exception in Message Driven Bean

Question

I have a MDB deployed to a weblogic server that listens to a queue on another weblogic server. All configuration is done through a
deployment descriptor. When a message is put on the queue, my MDB picks it off, processes the message. I see the last line of debug
in my log files. Shortly thereafter, I see the following exception being recorded:
ExecuteThread: '13' for queue: 'weblogic.kernel.Default'> <<anonymous>> <> =
<BEA-010026> <Exception occurred during commit of transaction Xid=3DBEA1-00=
C5E1770B39616949EF(15171805),Status=3DRolled back. [Reason=3Djavax.ejb.EJBE=
xception: nested exception is: java.sql.SQLException: Data exception -- Dat=
a -- Input Data length 1,050,060 is greater from the length 1,048,576 speci=
fied in create table.],numRepliesOwedMe=3D0,numRepliesOwedOthers=3D0,second=
s since begin=3D27,seconds left=3D58,XAServerResourceInfo[JMS_cgJMSStore]=
=3D(ServerResourceInfo[JMS_cgJMSStore]=3D(state=3Drolledback,assigned=3DcgS=
erver),xar=3DJMS_cgJMSStore,re-Registered =3D false),XAServerResourceInfo[w=
eblogic.jdbc.wrapper.JTSXAResourceImpl]=3D(ServerResourceInfo[weblogic.jdbc=
..wrapper.JTSXAResourceImpl]=3D(state=3Drolledback,assigned=3DcgServer),xar=
=3Dweblogic.jdbc.wrapper.JTSXAResourceImpl@bd33c9,re-Registered =3D false),=
XAServerResourceInfo[rfsdev1-xa]=3D(ServerResourceInfo[rfsdev1-xa]=3D(state=
=3Drolledback,assigned=3DcgServer),xar=3Drfsdev1-xa,re-Registered =3D false=
),SCInfo[integro+cgServer]=3D(state=3Drolledback),properties=3D({weblogic.j=
dbc=3Dt3://10.246.0.119:7001}),local properties=3D({modifiedListeners=3D[we=
blogic.ejb20.internal.TxManager$TxListener@198b85b]}),OwnerTransactionManag=
er=3DServerTM[ServerCoordinatorDescriptor=3D(CoordinatorURL=3DcgServer+10.2=
46.0.119:7001+integro+t3+, XAResources=3D{JMS_FileStore, weblogic.jdbc.wrap=
per.JTSXAResourceImpl, JMS_cgJMSStore, rfsdev1-xa},NonXAResources=3D{})],Co=
ordinatorURL=3DcgServer+10.246.0.119:7001+integro+t3+): javax.transaction.SystemException:
Could not contact coordinator at .......
The thing is in my MDB, I'm not doing anything with the JTA, wha-so-ever! I've had this mdb
deployed for over a year and it was after an upgrade to the weblogic server hosting the queue
that we started seeing these exceptions.
When this started, I thought it was because I had upgraded my development box to 10.3, while
the host upgraded from 8.1 to 9.2. When they upgraded their production box, where these
exceptions were not seen prior to their upgrade, we began seeing them there, too.
Does anyone have any possible idea what may be causing this to happen? No code in my
MDB has changed!
Thanks,
John

Answer 1

thanks for the suggestion. I added it as an ActivationConfigProperty and set it to Bean, but I got the same error.
Is there any way to tell who's attempting to do the transacting? My bean doesn't have any logic to persist
anything or use any of the JTA Api. My counterpart, where the queue resides to which my MDB listens, is
as lost as I am.
John