Re: [v6ops] New Version Notification for draft-ietf-v6ops-dhcp-pd-per-device-05.txt

[Jen Linkova <[e-mail]>]

On Sun, Nov 5, 2023 at 11:13 PM Gert Doering <[e-mail]> wrote:
> As long as the draft still insists on "it needs to be a prefix
> suitable for SLAAC" while it's fairly clear that nobody really has
> the energy to get a non-/64-SLAAC document through 6man, this sort
> of word smithing will not affect my objection to the draft.

I'd like to draw your attention to the 6man thread about interface ID,
btw. So I'd not say "nobody".
I do not think it's a question of energy. If it would mean it would
stop that endless and fruitless discussion about "/64 is too much", it
would be time and energy well spent.
The question is "is it a good idea?". If the community thinks so, we
shall get it done. Once and for all.
But until it's done we have to support it .

> > 3. Rewriting examples in the prefix length consideration section to
> > make it clearer that the proposal does not require an excessive amount
> > of IPv6 space.
>
> How can it do anything else but require excessive amounts of space
> when it insists on SLAAC-suitable prefixes?

It allows the devices to extend the network and removes the
scalability (and security) issues the network has.
As an operator, I'm willing to pay that price. Your experience may vary.

> Let me repeat that no wifi or wired network that I will ever have an
> influence on will ever delegate /64s to end devices.

That's perfectly fine. You are not going to do it on your network -
it's your deployment choice.
 A lot of networks (like small enterprises or home networks) would
never do that - the proposed deployment model doesn't provide any
benefits in that scenario.
Some of those networks do not have scalability issues and ND proxy
works just fine. Some of those networks are going to be dual-stack
forever so partially broken IPv6 is not a concern for them.

>You can have /80s
> or smaller all you want.  So your implementations will need to take that
> into account - and if they can't deal with it, will have to use NAT66
> then.

So it will be the status quo. You don't do it now, you are not doing
it in future.  Devices would continue to work on your network as well
as they are doing it now.
Other networks might look at pros and cons and make a different decision.

--
Cheers, Jen Linkova