You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@httpd.apache.org by Scott Horowitz <sc...@theskyiscrape.com> on 2004/11/01 18:18:25 UTC
Re: [users@httpd] httpd processes and ram - Success!
Sorry, I was indeed on Apache 1.3.33 (I got it mixed up with php 4.3.9..
I've been updating everything I can think of trying to get this to work).
And "apachectl restart" was NOT freeing up the memory for whatever reason,
my only option was a physical reboot.
However, that said, I wanted to officially announce success! I finally went
ahead and upgraded to Apache 2.0.52 and I'm 99.9% sure that it has fixed the
problem. The httpd processes have been staying steady at 5mb a piece. So I
don't know what the problem was, but it must have been a memory leak in
Apache 1.3.33. The server is how humming away.
Well I've learned my lesson, always use the latest stable versions ;) Thanks
again for all of your help Joshua, obtaining more ram was also definitely
needed.
Scott
On Sun, 31 Oct 2004 19:56:24 -0700, Scott Horowitz
<sc...@theskyiscrape.com> wrote:
> Okay, I've been slowly pinpointing the problem on my web server (Apache
> 1.3.9, RedHat 7.3, 1GB of ram).
If that is really the version of apache you are using, then you have
more serious problems than RAM. That version is way out of date and
has security as well as other bugs that have been fixed in more recent
versions. It is a waste of time to try to debug problems in a version
that old.
> However, as time passes, the httpd processes begin to use more and more
> ram. After only a few hours, if that, many of the httpd processes show
> up as 12+mb each. Is there any reason that they slowly eat away at the
> ram?
A memory leak? Most likely in one of the perl/php/whatever scripts
that you are using. It is highly unlikely that the core apache has a
memory leak that big, even in your old version.
> Is it normal for httpd processes to use more ram over time? Even if I do
> a "apachectl restart", the memory doesn't decrease very much (it only
> goes back down to 700+mb). My only solution is to do a complete server
> reboot.
An apache restart will free the memory used by apache's child processes.
You can also look at the MaxRequestsPerChild directive which can be
used to force apache to recycle processes more frequently. But you
should really start by tackling your two major problems: a terminally
old version of apache (and Redhat, but I don't know how serious that
is) and some scripts that are leaking tons of memory.
Joshua.
---------------------------------------------------------------------
The official User-To-User support forum of the Apache HTTP Server Project.
See <URL:http://httpd.apache.org/userslist.html> for more info.
To unsubscribe, e-mail: users-unsubscribe@httpd.apache.org
" from the digest: users-digest-unsubscribe@httpd.apache.org
For additional commands, e-mail: users-help@httpd.apache.org
---------------------------------------------------------------------
The official User-To-User support forum of the Apache HTTP Server Project.
See <URL:http://httpd.apache.org/userslist.html> for more info.
To unsubscribe, e-mail: users-unsubscribe@httpd.apache.org
" from the digest: users-digest-unsubscribe@httpd.apache.org
For additional commands, e-mail: users-help@httpd.apache.org