[LAC-TF] Sesion en IETF 95 (ERA: Re: implicaciones de declarar IPv4 historico)

[Fernando Gont]

Lee,

On 04/06/2016 07:10 PM, Howard, Lee wrote:
> Cómo no es prime-time?
> 
> What is missing? This draft document would only have affected IETF
> process, but if there are gaps the IETF needs to cover first, what still
> needs to be done?

With respect to deprecating IPv4, I'd say that IPv6 and IPv4 are really
two different protocols -- not really versions that could e.g. be
interoperable (i.e., v6 being backwards compatible). So in that sense
I'm not sure to what extent it can be argued that IPv6 is a new version
of IPv4 (yes, both are called IP, but that's it).

Nowadays, you cannot really prefer one protocol over the other, for the
simple reason that IPv6 is not as widespread as one would like it to be.
There are lots of deployed devices that simply do not support it.
In that respect, declaring IPv6 historic is in a sense, saying that the
Internet is running on an historic protocol, or pretending that you can
switch to that protocol, when in many cases, you simply can't

Regarding IPv6 being prime-time, there are issues when it comes to
implementations. Both in terms of host/uter implementations, and other
devices such as firewalls e.g., see:
<http://www.ipv6hackers.org/meetings/ipv6-hackers-1/zack-ipv6hackers1-firewall-security-assessment-and-benchmarking.pdf>.
(videos and stuff available at: <http://www.ipv6hackers.org>). (think
about NIDS etc support for IPv6 etc., too).

The above does not mean that you cannot deploy today. But rather that we
really still rely on IPv4, there are devices and boxes that will
probably continue to do ipv4-only for the rest of their lifetime, and
there scenarios for which if you really wanted to do 6, you'd have a
hard time.

P.S.: Thanks for joining this list. That's appreciated!
-- 
Fernando Gont
SI6 Networks
e-mail: fgont at si6networks.com
PGP Fingerprint: 6666 31C6 D484 63B2 8FB1 E3C4 AE25 0D55 1D4E 7492