Re: [Idr] Review Updates to draft-ietf-idr-bgp-prefix-sid

["Acee Lindem (acee)" <[e-mail]>]

Hi  Alvaro,

There are implementations of SRv6 today for various use cases (e.g. BGP VPNs, TE, NFV, Spray)
on multiple operating systems and multiple hardware platforms.

There have been demos and interop exercises between multiple vendors for SRv6 including
trials with operators.

The document change is related to one specific TLV which was not needed.

The SRv6 base specification describes the different control protocol usage and signalling in
https://tools.ietf.org/html/draft-filsfils-spring-srv6-network-programming-03#section-8.4
via IGPs, BGP and BGP-LS.

- The IGP specifications for SRv6 are covered by draft-bashandy-isis-srv6-extensions and
  draft-li-ospf-ospfv3-srv6-extensions.
- The BGP VPN specifications is covered by draft-dawra-idr-srv6-vpn.
- The BGP-LS specification for SRv6 are covered by draft-dawra-idr-bgpls-srv6-ext.

With the above documents the SRv6 control plane is complete.

We propose to deprecate the IPv6 SID TLV defined in draft-ietf-idr-bgp-prefix-sid.
It is not required for any use-case.

Thanks Much,
Acee


From: Alvaro Retana <[e-mail]>
Date: Tuesday, February 20, 2018 at 1:52 PM
To: Acee Lindem <[e-mail]>, IDR List <[e-mail]>
Cc: "[e-mail]" <[e-mail]>, "[e-mail]" <[e-mail]>, Susan Hares <[e-mail]>
Subject: Re: Review Updates to draft-ietf-idr-bgp-prefix-sid

On February 17, 2018 at 3:41:02 PM, Acee Lindem (acee) ([e-mail]<mailto:[e-mail]>) wrote:

Acee:

Hi!

There is one more change we’d like to make to the draft that requires WG approval. One revelation during the review process for this draft is that the SRv6 use case as currently documented isn’t really that compelling and there are some details that are unspecified. Other than validation, the IPv6 SID TLV hasn’t been implemented. Consequently, we’d like to remove the IPv6 SID TLV and SRv6 dataplane specification from the draft. Are there any objections (other than the obvious one that it is very late to be making this change)?

Here are a list of implementations and none have implemented the IPv6 SID:
https://trac.ietf.org/trac/idr/wiki/draft-ietf-idr-bgp-prefix-sid%20implementations

Note that the code point for the IPv6 SID would be deprecated since validation has been implemented in several implementations.

If there are no objections, I could make this change next week and we could allow for an additional week for review.

This is a significant change!  Thanks for specifically pointing out the lack support in the implementations for the IPv6 SID.  Please consider adding an rfc7942-style section to the document.

While I don’t have a specific objection to you simplifying the document, it is a decision that shouldn’t be made *after* the IESG has approved the publication of the document.  OTOH, better late than never!

idr-chairs:  I am returning the document to the WG for discussion and update.  If the WG reaches consensus on the proposed changes, I want to make sure that it is done with the understanding that the related IPv6 work will be done later.  IOW, the result should not be that the IPv6 work is abandoned — I am sure that is not the intent.  Note that we don’t need to discuss or adopt a specific alternative before processing this document.

Thanks!

Alvaro.