Date: Tue, 8 Feb 2000 19:30:57 +0000 (GMT)
From: James Sutherland <jas88@cam.ac.uk>
To: new-httpd@apache.org
Subject: Re: mod_proxy: proposal for v2.0
In-Reply-To: <38A05F16.7C75E452@sharp.fm>
Message-ID: <Pine.LNX.4.10.10002081918210.19416-100000@dax.joh.cam.ac.uk>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: new-httpd-owner@apache.org
Precedence: bulk
Reply-To: new-httpd@apache.org
Status: O

On Tue, 8 Feb 2000, Graham Leggett wrote:

> James Sutherland wrote:
> 
> > I must admit, it's not a solution I would consider ideal...
> > 
> > What about using Squid as a frontend? It could (with a rewriter) handle
> > both the load-balancing and URL-based proxying easily enough, while Apache
> > would handle all the backend work. Squid can also produce full logs.
> > 
> > That's probably the way I would do it (depending, of course, on the exact
> > circumstances: how many servers? What bandwidth? etc.)
> 
> When I last checked squid couldn't write
> split logfiles,
Split by vhost (one entry per file, in a different file depending on URL)
or by content (different entries in multiple files)? In either case, I'd
use a cron script to split up log files as needed, rather than imposing
the extra overhead directly.

Or use the backend server logs, with a suitable script to merge entries as
needed.

> handle URL redirects,
It can pass different URLs to different servers in any way you care to
define (all URLs containing cgi-bin to one server, all .shtml files to
another...) easily enough using a rewriter, as I suggested.

> password protect arbitrary URLs
This is done by the backend server as normal.

> or support SSL (direct SSL, not SSL tunneling),
True; depending on requirements, I would either use one [group of] servers
in a round-robin serving secure content directly, or have SSL front end
boxes running Apache+SSL.

> these are all webserver functions. Squid is a forward proxy first,
> with an http accelerator bolted on. Apache is a webserver first with a
> reverse_proxy bolted in. This to us makes a world of difference.
Depending on what "us" requires, I would be more inclined toward the Squid
solution; I would have thought the overhead of multiple extra Apache
layers handling every request would more than outweigh the benefits of
having the log files divided up realtime, rather than only every few
minutes...

James.