I-STAR Problem

We have a weird problem with an I-Star configuration: The users are experiencing really slow response times since moving one database to a separate machine - the versions of Online and I-Star on both (HP-UX) machines are 5.01.UD3. The log file on the server machine is showing the following errors: 23:57:21 Checkpoint Completed Thu Jan 20 00:02:44 1994 00:02:44 Checkpoint Completed 00:08:00 Process Aborted Abnormally: pid=9041 user=305 us=c0018f2c flags=1 00:08:32 Checkpoint Completed 00:09:04 Process Aborted Abnormally: pid=9180 user=314 us=c0018f2c flags=1 00:14:00 Checkpoint Completed 00:14:25 Process Aborted Abnormally: pid=10506 user=314 us=c0018f2c flags=1 00:17:05 Process Aborted Abnormally: pid=11766 user=305 us=c0018f2c flags=1 The tbstat profile is showing the following: RSAM Version 5.01.UD3 -- On-Line -- Up 16:16:53 -- 29984 Kbytes Profile dskreads pagreads bufreads %cached dskwrits pagwrits bufwrits %cached 1012683 1020386 18518671 94.53 48375 111291 221183 78.13 isamtot open start read write rewrite delete commit rollbk 14771825 349273 2429145 5390859 63967 0 0 148579 0 ovtbls ovlock ovuser ovbuff usercpu syscpu numckpts flushes 0 0 0 0 2136.92 835.26 88 176 bufwaits lokwaits lockreqs deadlks dltouts lchwaits ckpwaits compress 114198 39 24088220 0 0 10451 212 5705 The application on the client machine does not show any errors - in fact, it gets the correct data from the server machine, just apparently very slowly (Looks like a new sqlturbo is automatically re-forked when the one dies abnormally ??) The data access to the server is done using sysnonyms - if you log into the client machine, run isql and do a query through the synonym, the data is returned with no problem. Any ideas on this would be greatly appreciated. Regards, Richard Ridley DHL Asia Pacific Singapore