You are viewing a plain text version of this content. The canonical link for it is here.
Posted to modperl@perl.apache.org by E Kolve <ek...@corp.classmates.com> on 2002/07/02 00:15:48 UTC
apache 1.3.26 reverse proxy
Has anyone noticed any performance problems using 1.3.26 as the front
end proxy to a backend mod_perl server?
I upgraded a box running apache 1.3.22 as the frontend proxy to 1.3.26.
Prior to upgrading the load was ~2.0 - 3.0. After upgrading, the load
went up to around 21 - 25. I then downgraded and the load went back to
normal. Could this have anything to do with the changes to mod_proxy
between 1.3.22 and 1.3.26? Any ideas?
--eric
Re: apache 1.3.26 reverse proxy
Posted by Igor Sysoev <is...@rambler-co.ru>.
On Mon, 1 Jul 2002, E Kolve wrote:
> I was watching the apache scoreboard file and it appeared the the
> mod_perl process was not being immediately freed by the proxy. Normally
> there will be 3 or 4 mod_perl procs in mode "W" Sending Reply, but after
> around 20 - 30 seconds on 1.3.26 as the proxy all 30 (that is my
> MaxClients for mod_perl) were in "W". This seems like a problem with
> either the IOBufferSize that was recently added or possibly something to
> do with the proxy not releasing the mod_perl process once it has gotten
> the last byte...
You can try mod_accel. Incomplete English documentation is available here:
http://dapi.chaz.ru/articles/mod_accel.xml
Also feel free to ask me directly.
Igor Sysoev
http://sysoev.ru
Re: apache 1.3.26 reverse proxy
Posted by Roger Pettett <rm...@sanger.ac.uk>.
Could this be a keepalive issue? I think keepalive support was fiddled
with in recent mod_proxies.
R.
Re: apache 1.3.26 reverse proxy
Posted by E Kolve <ek...@corp.classmates.com>.
I was watching the apache scoreboard file and it appeared the the
mod_perl process was not being immediately freed by the proxy. Normally
there will be 3 or 4 mod_perl procs in mode "W" Sending Reply, but after
around 20 - 30 seconds on 1.3.26 as the proxy all 30 (that is my
MaxClients for mod_perl) were in "W". This seems like a problem with
either the IOBufferSize that was recently added or possibly something to
do with the proxy not releasing the mod_perl process once it has gotten
the last byte...
--eric
David Dyer-Bennet wrote:
> E Kolve <ek...@corp.classmates.com> writes:
>
>
>>Has anyone noticed any performance problems using 1.3.26 as the front
>>end proxy to a backend mod_perl server?
>>
>>I upgraded a box running apache 1.3.22 as the frontend proxy to
>>1.3.26. Prior to upgrading the load was ~2.0 - 3.0. After upgrading,
>>the load went up to around 21 - 25. I then downgraded and the load
>>went back to normal. Could this have anything to do with the changes
>>to mod_proxy between 1.3.22 and 1.3.26? Any ideas?
>
>
> You don't actually talk about any performance problem. Were the
> proxied requests in fact running slowly? It seems to me a difference
> in the process structure or what how they wait could have horrible
> effects on the load average numbers while actually improving
> performance. Or not.
Re: apache 1.3.26 reverse proxy
Posted by David Dyer-Bennet <dd...@dd-b.net>.
E Kolve <ek...@corp.classmates.com> writes:
> Has anyone noticed any performance problems using 1.3.26 as the front
> end proxy to a backend mod_perl server?
>
> I upgraded a box running apache 1.3.22 as the frontend proxy to
> 1.3.26. Prior to upgrading the load was ~2.0 - 3.0. After upgrading,
> the load went up to around 21 - 25. I then downgraded and the load
> went back to normal. Could this have anything to do with the changes
> to mod_proxy between 1.3.22 and 1.3.26? Any ideas?
You don't actually talk about any performance problem. Were the
proxied requests in fact running slowly? It seems to me a difference
in the process structure or what how they wait could have horrible
effects on the load average numbers while actually improving
performance. Or not.
--
David Dyer-Bennet, dd-b@dd-b.net / New TMDA anti-spam in test
John Dyer-Bennet 1915-2002 Memorial Site http://john.dyer-bennet.net
Book log: http://www.dd-b.net/dd-b/Ouroboros/booknotes/
New Dragaera mailing lists, see http://dragaera.info