Re: [Snac] [v6ops] "router cascade with DHCPv6-PD"

[Vasilenko Eduard <[e-mail]>]

Hi Ted,
I do not have enough experience on this topic to tell “what has a better chance to be accepted by the market”.
Maybe you are right. Maybe “DHCP Reconfirm Request” has a better chance.
Ed/
From: Ted Lemon [mailto:[e-mail]]
Sent: Thursday, February 2, 2023 11:05 PM
To: Vasilenko Eduard <[e-mail]>
Cc: Ole Troan <otroan=[e-mail]>; [e-mail]; V6 Ops List <[e-mail]>; Juliusz Chroboczek <[e-mail]>; Michael Richardson <[e-mail]>
Subject: Re: [v6ops] [Snac] "router cascade with DHCPv6-PD"

I think given that it's the standard and that it seems like it should work, the better question to ask is, "is there a reason to design a whole new protocol to replace it because we think it might not work?"

It's possible that trying to revive HNCP and (if necessary) fix it might produce a better outcome, but someone has to want to do that, and I don't see anyone working on it.

On Thu, Feb 2, 2023 at 10:22 AM Vasilenko Eduard <[e-mail]<mailto:[e-mail]>> wrote:
Is it really possible to assume that we could use it for PA blocks?
For a massive production.
Or for some reason “DHCP Reconfirm Request” would not happen?
Ed/
From: Ted Lemon [mailto:[e-mail]<mailto:[e-mail]>]
Sent: Thursday, February 2, 2023 7:01 PM
To: Vasilenko Eduard <[e-mail]<mailto:[e-mail]>>
Cc: Ole Troan <otroan=[e-mail]<mailto:[e-mail]>>; [e-mail]<mailto:[e-mail]>; V6 Ops List <[e-mail]<mailto:[e-mail]>>; Juliusz Chroboczek <[e-mail]<mailto:[e-mail]>>; Michael Richardson <[e-mail]<mailto:[e-mail]>>
Subject: Re: [v6ops] [Snac] "router cascade with DHCPv6-PD"

DHCP Reconfirm Request works for withdrawal.

On Thu, Feb 2, 2023 at 7:41 AM Vasilenko Eduard <[e-mail]<mailto:[e-mail]>> wrote:
It looks like we are talking about address appointment. ND is for this task by definition. DHCP too.

ND is capable of announcements and withdrawal.
DHCP is only for an announcement.
Big difference.

Ed/
From: v6ops [mailto:[e-mail]<mailto:[e-mail]>] On Behalf Of Ole Troan
Sent: Thursday, February 2, 2023 4:25 PM
To: Ted Lemon <[e-mail]<mailto:[e-mail]>>
Cc: [e-mail]<mailto:[e-mail]>; V6 Ops List <[e-mail]<mailto:[e-mail]>>; Juliusz Chroboczek <[e-mail]<mailto:[e-mail]>>; Michael Richardson <[e-mail]<mailto:[e-mail]>>
Subject: Re: [v6ops] [Snac] "router cascade with DHCPv6-PD"

Ted,

Exactly!

I don't know how you do stable prefix assignment to subnets with NDP/ICMP. I'm sure you could, but how much different than PD would it be at that point?

The original prefix delegation proposal did exactly that:
https://datatracker.ietf.org/doc/html/draft-haberman-ipngwg-auto-prefix-02

And the observation was then “this looks just like DHCP, why not use that?”

O.



On Thu, Feb 2, 2023 at 12:22 AM Klaus Frank <[e-mail]<mailto:[e-mail]>> wrote:
I agree, with that. And tbh, I personally think introducing DHCPv6-PD at
all was a conceptional mistake. It's duties should have been carried out
by ICMPv6 and NDP...

Also there is no real alternative for environments with dynamically
assigned prefixes or where clients are roaming between networks and need
a delegated prefix locally for e.g. Docker or VMs... Except for "use
DHCPv6-PD to get the prefix, run a DIY script to update configuration
files and restart radvd and/or bird..."

On 22.01.2023 20:56, Gert Doering wrote:
> DHCPv6-PD seems to be a half-finished routing protocol already, and trying to
> wiggle around admitting it.
>
> Gert Doering
>          -- NetMaster
>
> _______________________________________________
> v6ops mailing list
> [e-mail]<mailto:[e-mail]>
> https://www.ietf.org/mailman/listinfo/v6ops
--
Snac mailing list
[e-mail]<mailto:[e-mail]>
https://www.ietf.org/mailman/listinfo/snac