I didn't get futher reports concerning bug 21763 (and related), so I assume this means things improved. It would be a good idea to release a new 4.1.28 build once this is confirmed to also provide a fix for Tomcat 4.1.x.
I've had some updates on this:
- Rex Young confirmed than bug 21763 has been fixed (very cool ! Thanks for testing).
- Ryan Lissack still experiences problems with the almost-latest (I think it doesn't include the latest catch Throwable change) tomcat-util from TC 5 (no thread is listening on the server socket anymore - for an unknown reason - after a period of time, under low to medium load; debugging traces show that the management of control runnables inside the pool is not at fault). Ryan is using Sun 1.4.1 Server JVM under Linux 2.4 on a 4 CPU server. I had doubts on the Sun Server VM, so he's going to test with the IBM VM (to which he'll switch to, as it performs faster, apparently). I hope the VM switch fixes the issue, because I don't see any problem with the socket handling code (unless the catch Throwable is somehow the needed fix, but there's apparently no trace in the logs about a problem like that).
Remy
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
