httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Dean Gaudet <dgau...@arctic.org>
Subject Re: about threads (yes it's an obsession :-)
Date Tue, 29 Jun 1999 19:05:39 GMT
On Tue, 29 Jun 1999, Laurent Sinitambirivoutin wrote:

> hello guys :-)
> 
> before bothering you again i've read the docs about linuxthreads
> well, under linux threads are "viewed" as processes by the kernel, i
> mean they enter in the 512 entries of a regular linux kernel. I still
> see this as a weak point because my real problem at eurosport is that I
> just can't apache go forking (unless i want my server to kneel down of
> course) I thought that threads could solve the problem cuz I believed
> they wouldn't be accounted as one of the 512 process allowed... sniff !
> So, my question is: will things like apr (for now) and soon mpm will
> allow me to sleep well regarding the process limit ??

no apr, or apache-apr/pthreads won't solve it (they use native threads). 

the MPM I'm working on can solve it, but you'll be waiting for longer (it
only requires as many threads as there are active connections at any one
instant). 

apache-nspr can solve it as well (nspr uses userland threads). 

As can using zeus, or thttpd... (neither use threads). 

Or using squid in front of an apache-1.3 (see the accelerator docs for
squid).

or using multiple machines, or a bigger machine (more RAM?). 

I'm honestly surprised you have enough bandwidth that one box isn't
satisfactory.  Either you have dynamic content chewing CPU (in which case
a threaded apache is going to be of little extra help); or you haven't
tuned your server; or ... I don't know.  But unless you've got enough
traffic to be filling many many T-1s I don't think the problem is Apache.

You should probably find some unix tuning documentation, there's some new
linux docs... I don't know where they are, I can't help you find them.

Start with "vmstat 1", maybe it'll provide you with some insight.

Dean



Mime
View raw message