W3C home > Mailing lists > Public > ietf-http-wg@w3.org > April to June 2011

Re: I-D draft-petersson-forwarded-for-00.txt

From: Willy Tarreau <w@1wt.eu>
Date: Thu, 7 Apr 2011 13:59:39 +0200
To: Poul-Henning Kamp <phk@phk.freebsd.dk>
Cc: Andreas Petersson <andreas@sbin.se>, ietf-http-wg@w3.org
Message-ID: <20110407115939.GA6428@1wt.eu>
Hi Poul-Henning,

On Thu, Apr 07, 2011 at 11:20:42AM +0000, Poul-Henning Kamp wrote:
> In message <20110407131343.69e5a3bd@hetzer>, Andreas Petersson writes:
> >On Thu, 07 Apr 2011 09:43:33 +0000
> >"Poul-Henning Kamp" <phk@phk.freebsd.dk> wrote:
> >
> >> One: Can we please settle on a single IPv6 representation, and make the 
> >> square brackets mandatory in all cases, not just when there is
> >> a port number present ?
> >
> >Yes, that is probably wise, the objection one might have is that it is
> >not the way IPv6-addresses is written in X-Forwarded-For.
> 
> I would be surprised of nobody does...

Well, I'd say I have never observed the square brackets in an x-f-f header
yet. I know for sure haproxy does not set them, I don't know how squid,
nginx nor apache proceed. I deduce from your question that varnish does
use them.

I don't see any reason to make them mandatory in place where no ports are
indicated, because there is no ambiguity. However, supporting both formats
at least for compatibility with existing products seems wise.

Regards,
Willy
Received on Thursday, 7 April 2011 12:00:10 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 27 April 2012 06:51:39 GMT