Re: [Int-dir] Intdir telechat review of draft-ietf-v6ops-dhcp-pd-per-device-07

[Tim Chown <[e-mail]>]

Hi,

> On 4 Apr 2024, at 10:34, Jen Linkova <[e-mail]> wrote:
> 
> On Thu, Apr 4, 2024 at 7:06 PM Tim Chown <[e-mail]> wrote:
>>> The thing is that the host behaviour is explicitly (and intentionally)
>>> out of scope. For me, as a network administrator, it doesn't matter
>>> how exactly the client configures that address: my network
>>> design/topology would be the same. The client can be a RFC7084-type
>>> router (that's what happens when someone plugs a CPE to an access
>>> port), or use smth like rfc7278 - or smth else. Up to the client, the
>>> network routes thet whole prefix to that device, do whatever you want
>>> with it.
>> 
>> OK, so maybe make it clearer that it is out of scope, as I was expecting to find some comment about it, and I could not.
> 
> The last paragraph of the Introduction states that:
> "This document focuses on the behaviour of the network. Host behaviour
> is not defined in this document."
> Do you think we need to make it more clear?

Sorry, I missed that, but the question of how more precisely this model would be deployed in, for example, a campus WiFi network, would be worth documenting. How addresses on the wireless side are configured, how the routing is done, whether a mixed environment is practical, interactions with eduroam (802.1x), that sort of thing. Perhaps that’s a second informational document. 

>> we do frequently have campus admins asking how they get host entries into the DNS with SLAAC, so it’s a FAQ.
> 
> Yes, I agree it's a problem to solve, but I think it might be more in
> scope for my 6MOPS draft...What do you think?

Sounds fine :)

There’s also probably something to be said somewhere about address accountability, it’s the flip side of the privacy considerations but useful in a campus environment.  I don’t think the draft mentions this (or at least has one unrelated instance of the word ‘account’).

Best wishes,
Tim