I work in a large enterprise IT shop, and we are running our Java applications on WebSphere's application server.

We have observed a behavior that I want to bring to an end: If a thread times out --- most notably while making an RMI call --- the application server appears to respawn the thread and send it again, at least once if not multiple times. This behavior is difficult to reproduce, but we have been able to observe it in Production logs of requests that repeat at exactly 60-second intervals (which matches with our timeout settings).

My research has only turned up one possible parameter to adjust that might end this behavior: - com.ibm.websphere.webservices.http.requestResendEnabled

This is a tightly controlled shop, so I don't want to adjust that parameter without being fairly certain it will have the intended effect.

Has anyone else observed this behavior, and is this the right path to correct it?

Thanks, Mike

  • Have you contacted IBM Support? – dbreaux Mar 28 '13 at 14:22

We contacted IBM support and they were unable to answer this question.

After performing additional research in IBM's documentation, we realized this is being caused by the HTTP plugin, which had a default setting added with the move to WebSphere 7: PostBufferSize.

This behavior (resending threads) is potentially very harmful in data entry applications and can be disabled by setting PostBufferSize = 0.

Refer to this article for more directly from IBM: http://www-01.ibm.com/support/docview.wss?uid=swg21450051

| improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.