Re: [Idr] draft-haas-idr-bgp-diffract-00 - 2 week WG adoption call (10/12 to 10/26)

[Kaliraj Vairavakkalai <[e-mail]>]

I support adoption of this draft.

Since two solutions CT and CAR have been adopted by IDR as “experimental standards”,
it makes sense to have some simple means to interoperate between the two if they meet
in a deployment.

This draft makes a good first attempt proposal at this problem.

My comments/questions on the draft are:


  *   It can list InterAS option-B, option-A as simple alternatives on how to interop as-well.

(I have action item to give text for this)



  *   When interoperating in option-C mode (CT CAR translation), is the problem-scope confined

to interoperating at Border-Nodes ASBR/ABR (nexthop-self) while keeping the data plane constant (MPLS/SR/SRv6)?

IOW does the draft intend to solve interop between the heterogenous data-planes as-well?
   e.g. ‘CT with MPLS’ in one domain and ‘CAR with SRv6’ in second, and ‘CT with SR-MPLS’ in a third one.


  *   I think to keep things simple, translation at RR (nexthop unchanged cases) should be kept out of scope.

As RR is supposed to be as transparent as possible.  Translation between SAFIs at RR will need new

rules. It will no longer be plain RR functionality. Confeds also be out of scope?


Also, as “experimental standards”, IDR WG is waiting for operators to share their experience experimenting
with both proposals. Did IDR-chairs get any feedback from operators on the same? I think that input should
help guiding this interop considerations discussion as-well. In keeping interop-procedures as light as possible.

2cs,

Thanks
Kaliraj

From: Idr <[e-mail]> on behalf of Susan Hares <[e-mail]>
Date: Wednesday, October 12, 2022 at 10:00 AM
To: [e-mail] <[e-mail]>
Subject: [Idr] draft-haas-idr-bgp-diffract-00 - 2 week WG adoption call (10/12 to 10/26)
[External Email. Be cautious of content]

This begins a 2 week WG Adoption and IPR for draft-haas-idr-bgp-diffract-00.txt.
https://datatracker.ietf.org/doc/draft-haas-idr-bgp-diffract/<https://urldefense.com/v3/__https:/datatracker.ietf.org/doc/draft-haas-idr-bgp-diffract/__;!!NEt6yMaO-gk!BMlikoZVXnVbXtWpb2UFKvgwtQedwqYNFE4QBJY7d7CU4uGQl1LA_ducQ7HAo-lb0EN9_PtvIMYHSEg$>

(10/12/2022 to 10/26/2022).   The chairs will provide a summary of adoption issues
On 10/24 so that an revision of this draft may be created before the IETF draft deadline.

This draft was discussed at the 8/29 IDR interim and will be discussed at
the 10/13 IDR interim.  This draft provides details on how
CAR and CT could have interoperable solutions.

Jeff provides an excellent introduction to this draft with:

"BGP Routes with Color" have become a topic of interest in the IDR
   Working Group.  The authors of the various proposals have many things
   in mind to solve with them, how a color might be used, and the
   operational model for their solution.  In general, the solutions
   share some core properties: Routes for an IP endpoint, that have a
   domain-wide semantic element to differentiate forwarding (often
   called a color), and appropriate state to enable that differentiated
   forwarding.


   Examples of the technologies enabling differentiated forwarding

   include MPLS and Segment Routing.



   This document examines two of the current proposals - BGP Color-Aware

   Routing, and BGP Classful Transport - and proposes mechanisms to

   permit them to interoperate.



Cheers, Sue


Juniper Business Use Only