>Hi, >I also noticed that in the documentation there is a mention that the >previous exitBuffer address is returned in the next invocation of the >message exit, so I guess I can free the memory then...That should be >fine...but in any case if there is something wrong with this approach...or >if there is any other way ...that would be nice to know.
>Thanks >Kind Regards >Aby Philip Aby, No this isn't really the right solution. You are not guaranteed to be called again with MQXR_MSG and therefore this could leak. Exits are always passed a Channel Exit Parameter structure containing a 16 byte field ExitUserArea. This field should be used to store a pointer to any state information the exit may have. This is really true for all exits. The reason it is 16 bytes is that OS/400 has 16 byte pointers. In your case the state information is the Exit buffer address. I would recommend you allocate a structure for all your state information at MQXR_INIT time. One of the fields in this structure would be a place for you to store your exit buffer pointer if you've allocated one. Perhaps you might also want to store things like the length of the buffer etc. At MQXR_TERM time you should look to see whether this buffer pointer has been set and if so free it and then free the state structure itself. Hope this helps, P. Paul G Clarke WebSphere MQ Development IBM Hursley Instructions for managing your mailing list subscription are provided in the Listserv General Users Guide available at http://www.lsoft.com Archive: http://vm.akh-wien.ac.at/MQSeries.archive
