Mailing-List: contact dev-help@apr.apache.org; run by ezmlm
Precedence: bulk
Received-SPF: pass (nike.apache.org: domain of jorton@redhat.com designates
 66.187.233.31 as permitted sender)
Date: Mon, 10 Dec 2007 21:58:41 +0000
From: Joe Orton <jorton@redhat.com>
To: dev@apr.apache.org
Subject: Re: svn commit: r602176 - /apr/apr/trunk/network_io/unix/sockaddr.c
Message-ID: <20071210215841.GA29536@redhat.com>
Mail-Followup-To: dev@apr.apache.org
References: <20071207184730.046351A9832@eris.apache.org>
 <20071210090718.GA6065@redhat.com> <475D8C0E.7020800@rowe-clan.net>
 <20071210195547.GA23734@infiltrator.gizzard.com>
 <20071210203411.GA28175@redhat.com> <475DAD92.4020204@rowe-clan.net>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
In-Reply-To: <475DAD92.4020204@rowe-clan.net>
User-Agent: Mutt/1.5.17 (2007-11-01)
Organization: Red Hat UK Ltd, Amberley Place, 107-111 Peascod Street, Windsor,
 Berkshire, SL4 1TE, United Kingdom.
  Registered in UK and Wales under Company Registration No. 03798903
  Directors: Michael Cunningham (USA), Brendan Lane (Ireland),
 Matt Parson (USA), Charlie Peters (USA)

On Mon, Dec 10, 2007 at 03:20:18PM -0600, William Rowe wrote:
> Joe Orton wrote:
>> Either way, trying to work around that with an APR resolver hack seems 
>> completely wrong, -1, it will propagate v4-mapped IPv6 addresses when none 
>> are necessary, that may very well break/confuse other callers.
>
> Howso?  The user explicitly passes an IPv4 address and requests an IPv6
> flavor sa.

No, you've changed the behaviour for *every* family==AF_INET6 lookup 
whether that's for "www.google.com" or an IPv4 dotted quad.  Previously 
such lookups would fail where the hostname had no v6 address associated; 
now, where there is a v4 address associated, they will always succeed 
and give out the v4-mapped IPv6 address.

The previous behaviour makes far more sense, and it would not be 
unreasonable for applications to rely on it.  In fact it looks like the 
APR_IPV6_ADDR_OK flag is exactly a case which relies on that behaviour, 
and is now presumably broken.

joe