Date: Sat, 25 Jan 1997 16:45:18 -0500 (EST)
From: "Jason A. Dour" <jad@bcc.louisville.edu>
To: new-httpd@hyperreal.com
Subject: Re: Countdown to 1.2b5 
In-Reply-To: <199701252142.PAA08715@sierra.zyzzyva.com>
Message-ID: <Pine.LNX.3.95.970125164326.29998A-100000@bcc.louisville.edu>
Comments: Getting paid to be a geek is cool...
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: new-httpd-owner@apache.org
Precedence: bulk
Reply-To: new-httpd@hyperreal.com

-----BEGIN PGP SIGNED MESSAGE-----

On Sat, 25 Jan 1997, Randy Terbush wrote:
> >   * suexec setuid() before directory checks for nfs mounted filesystems
> >        Status: no patch yet, Randy said he'd look at it
> 
> This presents some potential problems for logging the transaction
> once we have setuid(). I would prefer to postpone changes here until
> after 1.2b5.

	Agreed.  For 1.2b6dev...definitely.

> I also raised a question about the expected behavior of create_argv().
> Currently, the code will not pass foo=bar as argv[1]. My question
> is "should it?". As RobH points out, perhaps this would encourge use
> of the QUERY_STRING.  I need some ruling before ignoring or fixing this.

	I rule for fixing.  It *has* worked in the past, but for what
reasons I do not know.  I personally have scripts that use it, and I know
many others who do.  Passing a single argument is a valid part of the CGI
spec, isn't it?

Jason
# Jason A. Dour <jad@bcc.louisville.edu>                            1101
# Programmer Analyst II; Department of Radiation Oncology; Univ. of Lou.
# Finger for URLs, PGP public key, geek code, PJ Harvey info, et cetera.

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBMup+8Zo1JaC71RLxAQHNQgP/Sg2UlorZONP40PfSnVT+hLnc5qGdjXI9
k0ucwG+AHinWwDvHtKRqSkQhA1s1yULAnok6JHOcd+A18ZaD+yujkv/knCXx+UnY
33yWqoeejkuhqL47xW7GyZuJ2052gmgcoeUlzLaH8+W9L2bdSZFhg4OLmS56ZN80
GDZneRyz7LQ=
=1T4q
-----END PGP SIGNATURE-----