Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 2 Nov 1998 10:17:31 -0500 (EST)
From:      Robert Watson <robert@cyrus.watson.org>
To:        "Jordan K. Hubbard" <jkh@time.cdrom.com>
Cc:        dg@root.com, Jun-ichiro itojun Itoh <itojun@itojun.org>, obrien@NUXI.com, Garrett Wollman <wollman@khavrinen.lcs.mit.edu>, current@FreeBSD.ORG
Subject:   Re: IPv6 in -current 
Message-ID:  <Pine.BSF.3.96.981102100920.15227E-100000@fledge.watson.org>
In-Reply-To: <23723.910007450@time.cdrom.com>

next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, 2 Nov 1998, Jordan K. Hubbard wrote:

> Yeah, if I had my druthers (and what the hell are "druthers" anyway,
> and who here has ever had any that they knew of?  Why is English such
> a peculiar language?  And why...  Erm, excuse me, I guess that's not
> really important right now), I'd want to see the IPv6 bits integrated
> with the following provisos:

English is a truly bizarre language.  I'll dig up my OED this evening and
let you know :-P.

> 1. If you make the world with NOIPV6, the tools are built in the
>    traditional fashion without any support for IPv6 at all.  This
>    would let the solution-in-a-box folks continue to compile binaries
>    with the smallest possible footprint, assuming that some of them
>    will have no need for IPv6.

Actually, I think a far more exciting option at some point in the process
would be NOIPV4, as that would imply that our network code was pretty
modular. :)  The other cool thing would be to see NAT capable of IPv4/IPv6
translation so that we could market FreeBSD as an easy solution the the v6
enclave problem.  Yet another case where BSD networking is the best choice
(or something).

> 2. If a binary (like ping) has been compiled with both v4/v6 support
>    and you simply want to turn its IPv6 behavior off for some reason,
>    it should check a well-known environment variable from its main()
>    to switch the relevant code in and out.  Purists might even
>    argue that IPv6 should be turned off by default and only enabled
>    through such an environment variable (or compiler flag) rather than
>    the other way around.  I guess I don't care either way so long as
>    IPv6 eventually, at the suitable time, becomes a desirable out-of-box
>    default for FreeBSD.

I suppose a useful one might actually be an environmental variable that
specifies what the 'default' IP version is for interpretting hostnames
that have both IPv4 and IPv6 addresses.  That is, if my /etc/hosts (or
eventually DNS) returns both an IPv4 for my.friendly.host and an IPv6
address, which should get priority for use in programs (such as ping).
>From the point of view of IP addresses, ping should be able to distinguish
just fine which one you need.

> I also agree that I don't think we can fence-sit on this one too much
> longer, as much as I also *hate* the idea of alienating some other
> group of hard-working IPv6 people.  Not all decisions are either easy
> or avoidable.

Heh.  Good thing you core people are in the hot seat, not me. :)

  Robert N Watson 

Carnegie Mellon University            http://www.cmu.edu/
TIS Labs at Network Associates, Inc.  http://www.tis.com/
SafePort Network Services             http://www.safeport.com/
robert@fledge.watson.org              http://www.watson.org/~robert/


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-current" in the body of the message



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.3.96.981102100920.15227E-100000>