Archives of the TeradataForum
Message Posted: Mon, 22 Mar 2004 @ 06:00:09 GMT
This obviously isn't going to help, but as part of a performance test, I ran two versions of a macro over 1,000 times without incident. The test simulated 1 to 64 users (or sessions) all running the same macro simultaneously. Once a session finished running the macro, it simply ran it again. This continued until the macro had been run 1,000 times in total.
In my example, I used "native" Java, but since JSP compiles down to Java, that is what you would have been doing to.
Are you tracking sessions or requests from one JSP page to another? If so, is it possible that you are losing your session (or request) data in some way? Do you hold any session information in an application level bean? If so, could one of the sessions be corrupting another sessions data in some way?
Are you creating threads to submit queries on behalf of the user in the background? I have encountered a (self inflicted) problem where I didn't keep track of my running threads properly and found that all my threads always got the result set of the last threads query - despite the individual threads running different queries. You might be encountering a similar problem.
It is a little hard to identify the source of a NullPointerException without the source and knowledge of the JSP container. The message that the system produced is only slightly better than "an error has occurred".
I hope some of these thoughts might help.
|Copyright 2016 - All Rights Reserved|
|Last Modified: 27 Dec 2016|