Re: Extending a /64

On Fri, 20 Nov 2020, 01:21 Tony Whyman, <tony.whyman@mccallumwhyman.com>
wrote:

>
> On 19/11/2020 11:30, Nick Hilliard wrote:
> > the key phrase is:
> >
> > Mark Smith wrote on 19/11/2020 05:22:
> >> [...] without any intention to ever
> >>     directly connect to other enterprises or the Internet itself.
> >
> > If the ICAO can guarantee that ipv6 assignments to aircraft will never
> > be used to directly connect to other third party networks, then ULA
> > will work fine. I doubt that this is a safe assumption.
> >
> > Nick
> >
> Correct, this is not a safe assumption. Also, a 40-bit global ID does
> not allow us enough remaining bits. We could set the L bit to 0 and then
> use our own format

- but happens if an internet (e.g. a military one)
> wants to do the same,

Both of you would be ab using both ULAs and ignoring IPv6 specifications.

What makes you think that once you have a high order prefix, you have the
freedom to arbitrarily define the format and semantics of any of the low
order bits?

If you do that, you'll be ignoring many IPv6 specifications including the
main IPv6 addressing architecture RFC, have to have custom implementations,
and may encounter problems that nobody else encounters.

Why would you want to shoot yourself in the foot like that?

or another RFC comes along and defines a new ULA
> format with L=0? All that we would really be doing is to squat on a
> portion of the global address space and this is not what we should be
> doing.
>
> --------------------------------------------------------------------
> IETF IPv6 working group mailing list
> ipv6@ietf.org
> Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6
> --------------------------------------------------------------------
>