You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@tomcat.apache.org by Dan <li...@dankeeley.co.uk> on 2007/12/14 18:00:20 UTC
Question about tomcat bugzilla which is resolved but not fixed.
Hi,
We've come across a problem in Tomcat 5.5.17 not completely dissimilar to
the following issue:
http://issues.apache.org/bugzilla/show_bug.cgi?id=33374
I notice that this issue has been marked fixed, but the comments on there
indicate that the bug isnt actually fixed, and that the problem remains.
Indeed, looking in the source of 5.5.25 it does not seem to contain the code
in the patch in this issue.
Can anyone answer why not? I wonder if i should attempt to patch the
current 5.5.25 and try this fix to see if it solves our problem? Any other
suggestions?
( The problem we have is threads in tomcat stuck in socketRead up to the
maxThreads, yet no similar connections on apache webserver. (modjk))
Thanks,
Dan
________________________________________________
Message sent using UebiMiau 2.7.10
---------------------------------------------------------------------
To start a new topic, e-mail: users@tomcat.apache.org
To unsubscribe, e-mail: users-unsubscribe@tomcat.apache.org
For additional commands, e-mail: users-help@tomcat.apache.org
Re: Question about tomcat bugzilla which is resolved but not fixed.
Posted by Bill Barker <wb...@wilshire.com>.
"Mark Thomas" <ma...@apache.org> wrote in message
news:4762CFD8.8080103@apache.org...
> Dan wrote:
>> Hi,
>> I notice that this issue has been marked fixed, but the comments on there
>> indicate that the bug isnt actually fixed, and that the problem remains.
>> Indeed, looking in the source of 5.5.25 it does not seem to contain the
>> code
>> in the patch in this issue.
>
> The actual patch applied is here.
> http://marc.info/?l=tomcat-dev&m=110740182826539&w=2
>
>> Can anyone answer why not?
>
> Only generally. I haven't looked at the patch in detail but Bill knows
> what
> he is doing. I suspect the proposed patch worked but didn't do things in
> the correct manner given how the Handler was designed.
>
The committed patch actually errors out sooner then the patch in 33374.
Also, the committed patch works for e.g. ChannelNioSocket as well.
Also, the failure to throw an IOException back to the servlet was fixed in a
later patch.
>> I wonder if i should attempt to patch the
>> current 5.5.25 and try this fix to see if it solves our problem? Any
>> other
>> suggestions?
>
> You could try but I doubt it would work any better than Bill's patch.
>
Actually, my patch should prevent that patch from doing much of anything at
all. But feel free to knock yourself out ;).
>> ( The problem we have is threads in tomcat stuck in socketRead up to the
>> maxThreads, yet no similar connections on apache webserver. (modjk))
>
Some firewalls forget to close the socket to the backend when it is closed
on the frontend. A work-around is to set a connectionTimeout on the
<Connector /> element similar to what is configured on the Apache side.
Personally, on a system that I actually care about the maxThreads value, I
prefer to set the connectionTimeout only on the Tomcat side and configure to
use CPING/CPONG on the Apache side.
> Can you reproduce this for one thread with a single request? A reliable
> test case is what is really needed to track any further problems down.
>
> Mark
>
>
> ---------------------------------------------------------------------
> To start a new topic, e-mail: users@tomcat.apache.org
> To unsubscribe, e-mail: users-unsubscribe@tomcat.apache.org
> For additional commands, e-mail: users-help@tomcat.apache.org
>
>
---------------------------------------------------------------------
To start a new topic, e-mail: users@tomcat.apache.org
To unsubscribe, e-mail: users-unsubscribe@tomcat.apache.org
For additional commands, e-mail: users-help@tomcat.apache.org
Re: Question about tomcat bugzilla which is resolved but not fixed.
Posted by Mark Thomas <ma...@apache.org>.
Dan wrote:
> Hi,
> I notice that this issue has been marked fixed, but the comments on there
> indicate that the bug isnt actually fixed, and that the problem remains.
> Indeed, looking in the source of 5.5.25 it does not seem to contain the code
> in the patch in this issue.
The actual patch applied is here.
http://marc.info/?l=tomcat-dev&m=110740182826539&w=2
> Can anyone answer why not?
Only generally. I haven't looked at the patch in detail but Bill knows what
he is doing. I suspect the proposed patch worked but didn't do things in
the correct manner given how the Handler was designed.
> I wonder if i should attempt to patch the
> current 5.5.25 and try this fix to see if it solves our problem? Any other
> suggestions?
You could try but I doubt it would work any better than Bill's patch.
> ( The problem we have is threads in tomcat stuck in socketRead up to the
> maxThreads, yet no similar connections on apache webserver. (modjk))
Can you reproduce this for one thread with a single request? A reliable
test case is what is really needed to track any further problems down.
Mark
---------------------------------------------------------------------
To start a new topic, e-mail: users@tomcat.apache.org
To unsubscribe, e-mail: users-unsubscribe@tomcat.apache.org
For additional commands, e-mail: users-help@tomcat.apache.org