> If what you say is true, then IBM has to update its claim about sequential > retrieval. No where in that claims does it make exceptions due to the > physical network. I would have thought that TCP/IP would reassemble the > packets in their proper order regardless how they traversed the network. > But, may IBM can clarify that.
I think this discussion is getting a little off base. Certainly TCP/IP may have to deal with IP packets arriving out of sequence because of different routes etc etc but TCP/IP guarantees to return bytes to the user of the socket in the same order. All hell would break loose if it didn't. Just imagine your average FTP or TelNet session. Besides MQ checks the sequence number of each message it receives and if it gets one in a different order than was sent if will complain, surely you've all seen this error :-) If you are getting messages in a different order then TCP/IP is not to blame. A more likely explanation is one of those listed in the manuals such as DLQ, different persistence attributes, different priorities or multiple routes etc etc. If you are certain it is not one of these then I would be interested in a trace. Cheers, P. Paul G Clarke WebSphere MQ Development IBM Hursley (although currently in Rochester,MN,USA) 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
