Re: [v6ops] Chair decision on WGLC for draft-ietf-v6ops-dhcp-pd-per-device-04

[Nick Buraglio <[e-mail]>]

On Sun, Nov 5, 2023 at 6:39 PM Martin Huněk <[e-mail]> wrote:

> Hi,
>
> Dne pondělí 6. listopadu 2023 0:59:42 CET, David Farmer napsal(a):
> > Are you saying you have more than 500,000,000 hosts and therefore need
> > ~8000 /48s to do /64 per host?
>
> This was actually bit of sarcasm. The reaction to Jordi as he suggested
> that we could give 1 /48 to every human on the planet and be good X years
> to come. :-)
> >
> > I have approximately 50,000 to 75,000 hosts on my WiFi networks at
> > any moment on my largest campus. My WiFi networks are split into 4 WiFi
> > roaming zones, and I would assign a /48 per roaming zone. We currently
> > assign /16s of RFC 1918 IPv4 space to each WiFi roaming zone.
> >
> > That is ~260,000 /64s or 4 /48s for 50,000 to 75,000 hosts. Yes, we are
> an
> > LIR, but within ARIN policy, I have assigned /40s for each of my 5
> campuses
> > statewide. My largest campus would qualify for a /36 within ARIN policy.
>
> Lucky you. We are not an LIR so we are limited to a single /48. Our
> upstream does have /29, but it is allowed to assign only /48 without RIPE's
> approval.


I agree with David’s assertion that this is a policy issue. However,  I
room a quick read and that a case for that approval reads to be very easy
to make.

>
> >
> > You have an IPv6 address policy problem, not an IPv6 architectural
> problem.
> > The idea that all organizations should fit in a single /48 is nonsense.
>
> No dispute there. I could use bigger address space, but without this draft
> I would manage even with /48. With it, I'm not quite sure about that.


This is the question that should be asked before designing an addressing
architecture: does a network need the scaling that this provides?
If no, then it is a non-issue. Don’t use it. Use a /64 or whatever is
mechanism is currently in play.
I certainly won’t use prefix per host in my local office or at my home
where I have a /48 and a /56 respectively. That being said, I have places
where this is very useful and should be considered.
In this circumstance, because of their size, these locations have the
appropriate address space to do it.
It’s only an addressing concern where it’s a mismatch for the architecture.

>
>
> Regards,
> Martin
> >
> > On Sun, Nov 5, 2023 at 17:27 Martin Huněk <[e-mail]> wrote:
> >
> > > Jordi,
> > >
> > > Please tell me where I can get my ~8000 /48s for the non-LIR University
> > > I'm working for. So far, I have 1 /48. And because we started our
> address
> > > plan in around 2007, it does not take into consideration every single
> host
> > > having /64.
> > >
> > > Also, as we are surrounded by other networks, I cannot expect to get my
> > > assignment extended, not by a single bit, even when the RIPE NCC would
> > > allow it (it currently does not). That would mean renumbering the whole
> > > network, which is not something I'm keen to do.
> > >
> > > Theoretically, It is possible by current RIPE policy to extend one's
> > > assignment/allocation. However, this is pure fiction in practice. I've
> > > tried that in the case of my second employer (LIR ISP) to get my /29
> > > extended to /28, but no luck, even when meeting the criteria.
> > >
> > > Saying that we can give every human /48 could work if we would route
> IPs
> > > to people in global routing tables. Fortunately, we want to aggregate
> ...
> > >
> > > Regards,
> > > Martin
> > >
> > > Dne neděle 5. listopadu 2023 19:57:25 CET jste napsal(a):
> > > > In a /3 there are 35.184.372.088.832 /48s
> > > >
> > > > if we underestimate the utilization to the 50%, that get’s us to
> half:
> > > 17.592.186.044.416 /48s
> > > >
> > > > If the population gets to 34.359.738.368 people (not sure if that
> will
> > > fit in the earth), and we estimate that every human will live 100
> years,
> > > and after that we don’t recover every /48, then with a single IPv6 /3,
> we
> > > will have a total IPv6 life-span of 51.200 years.
> > > >
> > > > If we use the rest of the /3s, then we get up to 409.600 years.
> > > >
> > > > And to be honest, I don’t think it makes sense to bury each /48 when
> a
> > > human pass away, same I don’t think that we should calculate it per
> human
> > > instead of householder, but I prefer to be pessimistic in terms of
> lifetime.
> > > >
> > > > Regards,
> > > > Jordi
> > > >
> > > > @jordipalet
> > > >
> > > >
> > > > > El 5 nov 2023, a las 19:46, Gert Doering <[e-mail]>
> escribió:
> > > > >
> > > > > Hi
> > > > >
> > > > > On Sun, Nov 05, 2023 at 07:42:03PM +0100,
> [e-mail]
> > > wrote:
> > > > >> It is pure maths.
> > > > >
> > > > > Geoff's maths disagree with yours.  I prefer to err on the side of
> > > caution.
> > > > >
> > > > > Gert Doering
> > > > >        -- NetMaster
> > > >
> > > >
> > > > **********************************************
> > > > IPv4 is over
> > > > Are you ready for the new Internet ?
> > > > http://www.theipv6company.com
> > > > The IPv6 Company
> > > >
> > > > This electronic message contains information which may be privileged
> or
> > > confidential. The information is intended to be for the exclusive use
> of
> > > the individual(s) named above and further non-explicilty authorized
> > > disclosure, copying, distribution or use of the contents of this
> > > information, even if partially, including attached files, is strictly
> > > prohibited and will be considered a criminal offense. If you are not
> the
> > > intended recipient be aware that any disclosure, copying, distribution
> or
> > > use of the contents of this information, even if partially, including
> > > attached files, is strictly prohibited, will be considered a criminal
> > > offense, so you must reply to the original sender to inform about this
> > > communication and delete it.
> > > >
> > > >
> > > >
> > > >
> > >
> > > _______________________________________________
> > > v6ops mailing list
> > > [e-mail]
> > > https://www.ietf.org/mailman/listinfo/v6ops
> > >
> >
>
> _______________________________________________
> v6ops mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/v6ops
>