tomcat-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Noam Camiel" <>
Subject Re: tomcat jitters, then hangs - please help
Date Tue, 21 Oct 2003 10:38:39 GMT
Very interesting bug report....
I will currently stick with my current configuration, but if this confirms
to solve the prolem, it could explain some oddities Remy wondered upon....

----- Original Message -----
From: "Euan Guttridge" <>
To: "'Tomcat Users List'" <>
Sent: Tuesday, October 21, 2003 10:54 AM
Subject: RE: tomcat jitters, then hangs - please help

Excellent news Noam.. Our testing (LD_ASSUME_KERNEL=2.4.1) is in progress
but too early to confirm anything. Another option is to stick with the NTPL,
using an updated glibc.

Check out RH errata ;
"A bug in pthread_cond_wait() which could cause programs hangs in some

If anyone tests this please drop the list a mail..


-----Original Message-----
From: Remy Maucherat []
Sent: 20 October 2003 14:32
To: Tomcat Users List
Subject: Re: tomcat jitters, then hangs - please help

Noam Camiel wrote:

> Hi folks,
> A week's gone by and all seems fine.  Server is stable and working
> properly. Thanks for all who helped out! Ryan, Remi Yoav and Arthur,
> many thanks. It may be a good idea to document this for others who may
> stumble over this problem. All the best
> Noam Camiel
> Details:
> Problem: Tomcat Server becomes unresponsive to new requests after
> several hours, regardless of load. OS RH9; Tomcat 4 or 5; VM Sun or
> IBM (JDK 1.4)
> Solve: (tcsh): setenv LD_ASSUME_KERNEL 2.2.5

Thanks :)
Very good news :)

I think I'll need to add that in the release notes (and putting it in
the FAQ would be good as well). Very weird symptoms, though. The only
explanation for the bug is if the thread waiting on the socket just
vanishes (there's no way it can go out of the blocking accept loop
otherwise - I reviewed that quite extensively ;-)).

Q: would the assume kernel 2.4 work as well ? I believe it would, but it
could be worth confirming it eventually. This is not critical, obviously.

Rémy Maucherat
Senior Developer & Consultant
JBoss Group (Europe) SàRL

To unsubscribe, e-mail:
For additional commands, e-mail:

To unsubscribe, e-mail:
For additional commands, e-mail:

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message