tomcat hangging problem

[jo...@bnpparibas.com]

Hello...

Here is the description of current problem I try to solve...
I have already mailed to tomcat-user, but since nobody has a beginning
of an answer, and since we can't go on production launch anymore, I try
here.

environment:
- AIX 4.3.3 SMP,
- JRE 1.2.2 (several builds), then 1.3.0
- tomcat 3.2.1, 3.2.3 then 3.3.b2 (sorry, did not try 3.3.rc1 yet!)

problem is: tomcat "hangs" after some time (may be 3 minutes or some
hours). I mean tomcat is running ("ps" shows it), but no answer is
received to requests when tomcat hangs (telnetting it with a http get
request results in no response). Even a kill -3 does nothing.

Strangely, problem occurs only on _some_ of AIX 4.3.3 machines where I run
tomcat.
Some machines, with same jre/tomcat/code, run with no problem (well,
maybe problem would occur after a much longer time, I don't really know).
And I did not met that on winNT4 (with a Sun jdk: I got to try with
an ibm jvm).
Note that I got a javacore twice, if somebody can interpret it...

I wrote a simple servlet which shows tomcat hangging, if that can help:
a singleton deals with doGet: first time it launches 'n' threads, each of
them will http-request same singleton every 't' seconds and print result to stdout.
other request to singleton only prints text/plain string to response's writer.

Thanks
--
Joseph Vallot




This message and any attachments (the "message") is
intended solely for the addressees and is confidential. 
If you receive this message in error, please delete it and 
immediately notify the sender. Any use not in accord with 
its purpose, any dissemination or disclosure, either whole 
or partial, is prohibited except formal approval. The internet
can not guarantee the integrity of this message. 
BNP PARIBAS (and its subsidiaries) shall (will) not 
therefore be liable for the message if modified. 

                ---------------------------------------------

Ce message et toutes les pieces jointes (ci-apres le 
"message") sont etablis a l'intention exclusive de ses 
destinataires et sont confidentiels. Si vous recevez ce 
message par erreur, merci de le detruire et d'en avertir 
immediatement l'expediteur. Toute utilisation de ce 
message non conforme a sa destination, toute diffusion 
ou toute publication, totale ou partielle, est interdite, sauf 
autorisation expresse. L'internet ne permettant pas 
d'assurer l'integrite de ce message, BNP PARIBAS (et ses
filiales) decline(nt) toute responsabilite au titre de ce 
message, dans l'hypothese ou il aurait ete modifie.

Re: tomcat hangging problem

[cm...@yahoo.com]

Hi Joseph,

I assume this is happening only on SMP machines ( i.e. with multiple
CPUs).

My bet is that this is related with the VM/libraries. It can of course be
a bug in tomcat, like a deadlock or something - but then it would fail
on all SMP machines.

Could you check with the 'latest' 1.3.0 and make sure you have all the
patches installed  ( pthred.so, etc ). Also, check the version of the
libraries between the AIX machines that work and those that don't.

If you get a core dump - again, this is a clear sign something is wrong
with the VM/libraries. A java program shouldn't be able to coredump
regardless of the code inside ( unless JNI is used ), and there's nothing
we can do in tomcat ( even if we find a workaround, some user code can
crash the server as well ).

Is anyone else running SMP machines ? I remember long time ago a problem
on linux/smp, but was solved by using a newer VM.

Costin

>
> Here is the description of current problem I try to solve...
> I have already mailed to tomcat-user, but since nobody has a beginning
> of an answer, and since we can't go on production launch anymore, I try
> here.
>
> environment:
> - AIX 4.3.3 SMP,
> - JRE 1.2.2 (several builds), then 1.3.0
> - tomcat 3.2.1, 3.2.3 then 3.3.b2 (sorry, did not try 3.3.rc1 yet!)
>
> problem is: tomcat "hangs" after some time (may be 3 minutes or some
> hours). I mean tomcat is running ("ps" shows it), but no answer is
> received to requests when tomcat hangs (telnetting it with a http get
> request results in no response). Even a kill -3 does nothing.
>
> Strangely, problem occurs only on _some_ of AIX 4.3.3 machines where I run
> tomcat.
> Some machines, with same jre/tomcat/code, run with no problem (well,
> maybe problem would occur after a much longer time, I don't really know).
> And I did not met that on winNT4 (with a Sun jdk: I got to try with
> an ibm jvm).
> Note that I got a javacore twice, if somebody can interpret it...
>
> I wrote a simple servlet which shows tomcat hangging, if that can help:
> a singleton deals with doGet: first time it launches 'n' threads, each of
> them will http-request same singleton every 't' seconds and print result to stdout.
> other request to singleton only prints text/plain string to response's writer.
>
> Thanks
> --
> Joseph Vallot
>
>
>
>
> This message and any attachments (the "message") is
> intended solely for the addressees and is confidential.
> If you receive this message in error, please delete it and
> immediately notify the sender. Any use not in accord with
> its purpose, any dissemination or disclosure, either whole
> or partial, is prohibited except formal approval. The internet
> can not guarantee the integrity of this message.
> BNP PARIBAS (and its subsidiaries) shall (will) not
> therefore be liable for the message if modified.
>
>                 ---------------------------------------------
>
> Ce message et toutes les pieces jointes (ci-apres le
> "message") sont etablis a l'intention exclusive de ses
> destinataires et sont confidentiels. Si vous recevez ce
> message par erreur, merci de le detruire et d'en avertir
> immediatement l'expediteur. Toute utilisation de ce
> message non conforme a sa destination, toute diffusion
> ou toute publication, totale ou partielle, est interdite, sauf
> autorisation expresse. L'internet ne permettant pas
> d'assurer l'integrite de ce message, BNP PARIBAS (et ses
> filiales) decline(nt) toute responsabilite au titre de ce
> message, dans l'hypothese ou il aurait ete modifie.
>