Re: Non routable IPv6 registry proposal

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 





On Wed, Jan 20, 2021 at 6:28 PM Fernando Gont <fgont@xxxxxxxxxxxxxxx> wrote:
Philip,

On 20/1/21 17:06, Phillip Hallam-Baker wrote:

> I have IPv6 service from Verizon but I obviously can't use it on my
> internal network because my IPv6 address changes every few months. This
> is certain to be the case for virtually every residential Internet drop
> and the vast majority of business customers.

Depends on what you're implying. You could certainly use them, and the
network would eventually gracefully renumber..  -- whether this would be
practical, is a different question.

Nah, thats not going to work.

People complain about NAT being complex. But it is ubiquitous and it just works because it is something used every day.

Any code path that runs only once every three years... its just not going to work reliably enough not to be a PITA.

 
You don't need this for IPv6. i.e., even if you translated (NPT) , you
don't need to multiplex all hosts into the same address -- this is/was
done in IPv4 because IPv4 addresses are scarce. BUt that's not the case
with IPv6.

Precisely the point I was making when identifying the precise features of IPv6 NAT we are going to need on an ongoing basis.

 
> This is probably sufficient. But a registry model would make for more
> efficient allocation of the space and allow the allocation to be bound
> to a public key whose private part is held by the registrant.

That comes at the price of running the registry. And I'm curios: if
you're going to pay, why not get a routable prefix, and simply not
announce it via BGP?

Ah that is a very important point: Because these are not going to be aggregated addresses. The idea here is that people use these address blocks on a fine grain basis. So a Fios drop with 256 customers might have over a thousand of these private address spaces being used inside it but only one AS number for that whole set. 

The point of using non routable addresses is precisely so we can break the rules that are required to make routing work. We are not going to have routing tables with 2^50 AS numbers in them. That wouldn't work.

[Index of Archives]     [IETF Annoucements]     [IETF]     [IP Storage]     [Yosemite News]     [Linux SCTP]     [Linux Newbies]     [Mhonarc]     [Fedora Users]

  Powered by Linux