Re: [v6ops] Continuing WGLC: ietf-v6ops-dhcp-pd-per-device-03

[Jen Linkova <[e-mail]>]

Hi Chongfeng

On Sat, Oct 7, 2023 at 9:04 PM Chongfeng Xie <[e-mail]> wrote:
> I support the progress of this draft and hope that it will eventually pass the last call. Nevertheless，I have the following suggestions:
> 1) This draft is about using DHCPv6 PD to Allocate Unique IPv6 Prefix to Client, the approache is described mainly in text. In order to help readers, >especially those outside the v6ops community, understand it more quickly, it is recommended to add an diagram to show the overall architecture and the >inter-relationships between the entities in it.

Yes, we'll add a diagram to -04 version of the draft.

> 2) Regarding to the scenario of virtual host, in reality, the implementation of virtual host can be divided into multiple methods, some methods do not >require the address obtained by the underlying physical host to assign an address to the upper layer virtual host. For example, in a cloud system, a >dedicated cloud management platform is generally used to configure addresses for tenants' DHCP servers, rather than being assigned by physical hosts. >Therefore, I hope to provide an explanation of the application scope of this solution in virtual hosts.
> 3) In addition, this draft mentions considerations or requirements for components, such as First-Hop Router and DHCPv6-PD Server. Should it also mention requirements or considerations for clients?

I'd like to answer both of your comments here. We intentionally focus
on the network side of the design.
After an extensive discussion on the list, we'd like this draft to
focus on the network size of things. This is a guidance for an
operator (which might or might not control the endpoints). As I can
see, Ole just submitted a draft discussing some aspects of the client
behaviour.
Obviously some endpoints might not need a prefix - because they do not
run virtual systems at all, or their systems are addresses via some
management platform (btw the latter wouldn't solve the scalability
issue caused by a large number of addresses scalability). But virtual
systems are just an example, one of possible scenarios when a client
might need a prefix. We provide examples but we'd like to avoid
enumerating all of them, as the such list is doomed to be incomplete.

> 4）The following Nits shoule be fixed
>    section 7： Thsi-->This
>    Section 9：conenctivity--->connectivity

Sure, thanks for pointing this out, -04 will have those typos fixed.

> From: 【外部账号】Jen Linkova
> Date: 2023-10-06 23:48
> To: V6 Ops List
> CC: Vasilenko Eduard
> Subject: [v6ops] Continuing WGLC: ietf-v6ops-dhcp-pd-per-device-03
> Dear v6ops WG,
>
> We have just submitted the -03 version for
> ietf-v6ops-dhcp-pd-per-device - thank you very much to everyone who
> provided feedback!
> Please review and let us know if the draft is ready to progress.
>
> The main changes:
> - Introduction is rewritten (much shorter and, hopefully, more clear)
> - some text about multiple addresses use cases moved to Appendix.
> - The Applicability and Limitation section is moved up, so it's in the
> beginning - hopefully it would help the readers to understand what
> scenarios are in scope and what are not.
> - Client Mobility section added to discuss what happens when a client
> moves between network attachment points.
> - the draft is now explicit that it focuses on "a pool per link"
> scenario, which also drastically simplifies the routing and relay
> requirements (the routing section is updated as well).
> - the draft now suggests Active Leasequery as one of the mechanisms to
> keep the relay state in sync with the server.
> - The Prefix Consideration section has been updated and states that
> the endpoints are not required to use SLAAC - it's just the network
> allows them to do that, if they need to.
> - various changes to address other comments received during the WGLC so far.
>
> I'd like to ask everyone who has expressed concerns during this WGLC
> (you are in Cc: - hopefully I didn't miss anyone) to review the new
> version and let the author know if you still have concerns.
>
> Thank you!
>
> On Fri, Oct 6, 2023 at 8:35 AM <[e-mail]> wrote:
> >
> > Internet-Draft draft-ietf-v6ops-dhcp-pd-per-device-03.txt is now available. It
> > is a work item of the IPv6 Operations (V6OPS) WG of the IETF.
> >
> >    Title:   Using DHCPv6-PD to Allocate Unique IPv6 Prefix per Client in Large Broadcast Networks
> >    Authors: Lorenzo Colitti
> >             Jen Linkova
> >             Xiao Ma
> >    Name:    draft-ietf-v6ops-dhcp-pd-per-device-03.txt
> >    Pages:   18
> >    Dates:   2023-10-06
> >
> > Abstract:
> >
> >    This document discusses an IPv6 deployment scenario when individual
> >    clients connected to large broadcast networks (such as enterprise
> >    networks or public Wi-Fi networks) are allocated unique prefixes via
> >    DHCPv6 Prefix Delegation (DHCPv6-PD).
> >
> > The IETF datatracker status page for this Internet-Draft is:
> > https://datatracker.ietf.org/doc/draft-ietf-v6ops-dhcp-pd-per-device/
> >
> > There is also an HTML version available at:
> > https://www.ietf.org/archive/id/draft-ietf-v6ops-dhcp-pd-per-device-03.html
> >
> > A diff from the previous version is available at:
> > https://author-tools.ietf.org/iddiff?url2=draft-ietf-v6ops-dhcp-pd-per-device-03
> >
> > Internet-Drafts are also available by rsync at:
> > rsync.ietf.org::internet-drafts
> >
> >
> > _______________________________________________
> > v6ops mailing list
> > [e-mail]
> > https://www.ietf.org/mailman/listinfo/v6ops
>
>
>
> --
> SY, Jen Linkova aka Furry
>
> _______________________________________________
> v6ops mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/v6ops



-- 
SY, Jen Linkova aka Furry