Message-Id: <3.0.3.32.19970811211215.0087ad30@localhost>
Date: Mon, 11 Aug 1997 21:12:15 -0700
To: new-httpd@apache.org, new-httpd@apache.org
From: Brian Behlendorf <brian@organic.com>
Subject: Re: Dean broke things again :)
In-Reply-To: <Pine.BSF.3.95q.970811213705.708A-100000@valis.worldgate.co
 m>
References: <3.0.3.32.19970811202138.0080b7c0@hyperreal.org>
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

At 09:41 PM 8/11/97 -0600, Marc Slemko wrote:
>On Mon, 11 Aug 1997, Brian Behlendorf wrote:
>
>> 
>> It looks like Dean's directory_walk optimizations, committed on August 8th,
>> broke content negotiation on "DirectoryIndex index" configurations.  Hey,
>> that's getting to be a pretty good litmus test :)  Currently, such a
>> configuration will fail to recognize "index.html" as a variant, and instead
>> return a directory index.  Curiously, that same patch seems to also fix the
>> PATH_INFO CGI problem we had earlier, the one which the patch right before
>> it (also from Dean) apparently did *not* fix, even though Dean thought it
>> would.  I.e.
>
>Works fine for me...
>
>index.html file exists, DirectoryIndex set for index, multiviews
>enabled, http://path/ gives me the contents of index.html

We really need that regression suite;

Probing further I see this only works if no vhosts (IP-based or host-based,
doesn't matter) are defined, but if they are, then it only works for the
*last* vhost definition in the config file.

	Brian


--=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=--
"Why not?" - TL           brian@organic.com - hyperreal.org - apache.org