Hi Bob,
Thank you for your review and insightful comments, we really apprieciate your
giving suggestions to improve this document, would you please see my feedback
inline?
From: Bob Hinden
Date: 2025-02-21 01:48
To: Chongfeng Xie
CC: int-area
Subject: [Int-area] Re: New Version Notification for
draft-xls-intarea-evn6-02.txt
Chongfeng,
I have a few comments on your draft, ranging from substantive to editorial.
Pref6: The draft defines this as a global prefix assigned to site. This is
a new term not used anywhere else that I can tell. It would be better to use
the terminology from RFC4291 than creating something new. It’s also not
defined in the Terminology section of the draft.
[Chongfeng]: OK, we will make the terminology consistent with RFC4291 and try
to avoid using new terminology.
I don’t like the idea of encoding the Ethernet Virtual Network information in
the IID. I think this violates RFC8064 "Recommendation on Stable IPv6
Interface Identifiers”. Using a new Destination Option would be a better
approach. It wouldn’t be creating stable IIDs and wouldn’t be limited to 64
bits.
[Chongfeng]: I have checked the RFC8064, it proposes not to use stable IPv6
address which is generated by embedding a stable link-layer address, e.g. MAC
address. The current approach of EVN6 does not violate this, since there is
no stable MAC address embedded in IPv6 address. Host ID in IID part is
dynamically allocated to each host by a EVN6 instance,It has no strong binding
relationship with the MAC address either, it can be used for traffic scheduling
in IPv6 network. However, I will seriously consider your proposal of using a
new Destination Option, it can be considered as an option of EVN6.
Section 4 Operation: Should the “IPv6 network” really be “Internet”. Or
what does "IPv6 Network” mean?
[Chongfeng]: The term of "IPv6 network" refers to IPv6 Internet currently.
In the second address diagram in figure 4, , s/Sit/Site/
[Chongfeng]: We will correct it.
Section 5.2 Broadcast: IPv6 doesn’t use Broadcast, it uses Multicast. The
text about ND using broadcast is incorrect.
[Chongfeng]: This is about the Broadcast and Multicast of Link layer, instead
of IPv6. I will check the text of ND using broadcast you mentioned.
The draft doesn’t discuss how the information in the mapping table is
provisioned.
[Chongfeng]: There has been a draft about the control layer of EVN6 in BESS
WG: draft-xie-bess-evpn-extension-evn6-01, it illustrate the extension of EVPN,
it can provide support to the provisioning of mapping table.
Hope this is helpful.
Bob
Can my feeback above address your concerns? Welcome to provide more suggestions.
Best regards
Chongfeng
> On Feb 19, 2025, at 8:06 PM, Chongfeng Xie <chongfeng....@foxmail.com> wrote:
>
>
> Hi all,
>
> We have submitted a new version of draft-xls-intarea-evn6, which is about
> "EVN6: Mapping of Ethernet Virtual Network to IPv6 Underlay for
> Transmission".
>
>
> Name: draft-xls-intarea-evn6
> Revision: 02
> Title: EVN6: Mapping of Ethernet Virtual Network to IPv6 Underlay for
> Transmission
> Date: 2025-02-20
> Group: Individual Submission
> Pages: 17
> URL: https://www.ietf.org/archive/id/draft-xls-intarea-evn6-02.txt
> Status: https://datatracker.ietf.org/doc/draft-xls-intarea-evn6/
> HTMLized: https://datatracker.ietf.org/doc/html/draft-xls-intarea-evn6
> Diff: https://author-tools.ietf.org/iddiff?url2=draft-xls-intarea-evn6-02
>
>
> Based on Zafar's comments in IETF 121, we have made changes to the format of
> the generated outer IPv6 addresses,
>
> 1) Add a new field 'Group Policy ID' in Interface ID, it is 24-bits
> identifier that indicates the source TSI Group membership being encapsulated
> by EVN6.
>
> 2) Instead of placing the MAC address directly in the IID, but using the
> 20-bits Host ID generated based the MAC address and then placing it in the
> Interface ID, this can mask MAC address information at the network layer and
> also distinguish hosts.
>
> Several editorial changes have been made as well.
>
> We are looking forward to receiving more comments and suggestsion from you.
>
> Best regards
> Chongfeng
> On behalf of all co-authors
>
>
>
>
> _______________________________________________
> Int-area mailing list -- int-area@ietf.org
> To unsubscribe send an email to int-area-le...@ietf.org
_______________________________________________
Int-area mailing list -- int-area@ietf.org
To unsubscribe send an email to int-area-le...@ietf.org
_______________________________________________
Int-area mailing list -- int-area@ietf.org
To unsubscribe send an email to int-area-le...@ietf.org
