From: Jim Jagielski <jim@jaguNET.com>
Message-Id: <199606091505.LAA06340@shado.jaguNET.com>
Subject: Re: feature idea
To: new-httpd@hyperreal.com
Date: Sun, 9 Jun 1996 11:05:06 -0400 (EDT)
In-Reply-To: <199606091421.KAA16353@volterra.ai.mit.edu> from "Robert S. Thau"
 at Jun 9, 96 10:21:23 am
Content-Type: text
Sender: owner-new-httpd@apache.org
Precedence: bulk
Reply-To: new-httpd@hyperreal.com

Robert S. Thau wrote:
> 
>   How about a compromise; using syslogd only for error_log?
> 
> Actually, I'd been assuming throughout this whole discussion that
> use of syslogd for anything would be a configurable option, since some
> sites (like, ahem... mine) are not encountering descriptor-consumption
> problems at all, and are perfectly happy now with what they've got.
> 

Our systems are limited to 128fds per process, so we would be hit earlier
than almost anyone else ('cept those systems with 64 or even, Ack!, 20!).
I'm guessing that we would hit a problem with somewhere around 35-40
virtual hosts (assuming about 3 fd's per VH), but I figure at that
point it makes more sense to buy another server... In other words,
I'd add another server before we would realistically hit our limit.
So I'm happy too.

In any case, having some feel for the number of possible maximum hosts
per system (or # of fd's available/process) would be a good addition
to the docs.
-- 
Jim Jagielski  << jim@jaguNET.com >>   |      "That's a Smith & Wesson,
  **  jaguNET Access Services  **      |       and you've had your six" 
      Email: info@jaguNET.com          |             - James Bond
++    http://www.jaguNET.com/         +++      Voice/Fax: 410-931-3157       ++