JBoss and Strange Looking Characters in the Message Data

Here’s an interesting issue that Lawrence Coombs had regarding porting a J2EE application from WAS to JBoss:

We ported an applictaion from WAS to jboss 5.1. It interfaces with a mainframe CICS transcation that uses MQ 7.x.

Neither the distributed nor the Z/OS queue manager changed. The CCSID of the Z/OS queue manager is 500 (did not change), the CCSID of the AIX queue manager is 819. The application did not change.

The data in question is coming from the mainframe. Any idea where to start looking?

D11¦00612730¦2012-07-29¦8807¦2012-07-30¦ ¦2012-07-24¦2012-07-25

This is the extra character at the end of each field: Â except the last.

The solution, as reported by Lawrence Coombs, was to use either of the following JVM environment variables.

-Dfile.encoding=ISO8859-1

or

-Dcom.ibm.msg.client.wmq.receiveConversionCCSID=1208

Hopefully, this posting will help other JBoss users in the future.

Regards,
Roger Lacroix
Capitalware Inc.

This entry was posted in IBM MQ, Java, JMS.

Comments are closed.