httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Alexei Kosut <>
Subject Re: patch? (Patch to fix the fatal missing VHost dir)
Date Sun, 02 Jun 1996 20:21:57 GMT
On Sun, 2 Jun 1996, Brian Behlendorf wrote:

> >   I never really understood why there were multiple config 
> >   files in the first place.
> I can see perhaps that different configuration directives are at diferent 
> administrative levels, for example you might want to allow a 
> semi-administrator the ability to add new content types or directory 
> indexing options without giving them the ability to change the User 
> setting.  

Right... examining the NCSA config files (same ones we have,
access.conf, httpd.conf and srm.conf) it all seems to make sense. In
fact, it might even make sense from a programming point of view:

httpd.conf - this contains directives that apply to the server as a
whole. This can be read in when the server starts, set up as per the
directives, and ignored.

access.conf - this contains directives dealing with access. This gets
checked whenever a request comes in, to see if we should allow the
request or not.

srm.conf - this contains directives dealing with server resource
management (I think that's what srm stands for). This is like one big
htaccess file for the whole server. After you've obtained access to
the request, you look at the directives in srm.conf and apply them to
figure out *how* to serve the request.

So we could break the config files up more succinctly: httpd.conf is
who (in a non-person related sense) the server should be, access.conf
is when the server should be (i.e. when it should allow access) and
srm.conf is how the server should be. The file accessed is what the
server should be, I guess, and I have no clue why. Maybe Rob McCool
liked the number three, so decided to make three config files.

Alexei Kosut <>      The Apache HTTP Server
      "War does not determine who is right, only who is left."

View raw message