[Idr] Re: WG LC on draft-ietf-idr-ts-flowspec-srv6-policy-08 (02/10/2026 - 02/24/2026)

[[e-mail]]

Hi,

Thanks for your feedback.

 

IHMO, the draft-ietf-idr-ts-flowspec-srv6-policy looks like more a hack than a solution while draft-li-idr-flowspec-sr-policy is probably a real solution.

I’m definitely against standardizing the encoding of draft-ietf-idr-ts-flowspec-srv6-policy for the reason I have given in my email.

 

> Regarding multiple solutions for one issue, it is not uncommon in IETF. For example, we have OSPF and ISIS for IGP and we have PCEP and BGP for SR Policy distribution.

 

This is not the same, here you propose two different encodings for the same thing in the same protocol (we are not talking about solutions involving different protocols). If IETF ships both, this will create a mess in the industry in term of interoperability, vendors will never be kind to support both (extra effort for 0 value).

 

Let’s try to stick with one good solution.

 

 

Brgds,

 

Stephane

 

 

From: 李振强 <[e-mail]> 
Sent: Wednesday, February 25, 2026 4:25 AM
To: slitkows.ietf <[e-mail]>; idr <[e-mail]>; keyur <[e-mail]>
Cc: idr-chairs <[e-mail]>; draft-ietf-idr-ts-flowspec-srv6-policy <[e-mail]>; 刘松 <[e-mail]>
Subject: Re:[Idr] Re: WG LC on draft-ietf-idr-ts-flowspec-srv6-policy-08 (02/10/2026 - 02/24/2026)

 

Hello Stephane,

 

Thank you for your comments, suggestion and support.

 

Wenying Jiang and Yisong Liu, authors of draft-ietf-idr-ts-flowspec-srv6-policy, are my colleagues. We have regular internal discussions.

 

Yes, draft-li-idr-flowspec-sr-policy defines new standard actions for Flowspec version 2 (FSv2), whereas draft-ietf-idr-ts-flowspec-srv6-policy is mainly for Flowspec version 1 (FSv1). It uses existing methods to quickly solve the problem encountered in field network deployment scenarios. With the community's consensus, I am very glad to define new standard actions for FSv1 to avoid ambiguity. Maybe a separate document is needed for this purpose.

 

Regarding multiple solutions for one issue, it is not uncommon in IETF. For example, we have OSPF and ISIS for IGP and we have PCEP and BGP for SR Policy distribution.

 

For clarifications related to interactions with other actions type, I will try to update draft-li-idr-flowspec-sr-policy before the submission deadline of the coming IETF meeting.

 

Thanks again.

 

Best Regards,

Zhenqiang Li

China Mobile





----邮件原文----
发件人："slitkows.ietf" <[e-mail] <mailto:[e-mail]> >
收件人："李振强" <[e-mail] <mailto:[e-mail]> >,'idr' <[e-mail] <mailto:[e-mail]> >,'keyur' <[e-mail] <mailto:[e-mail]> >
抄　送: idr-chairs <[e-mail] <mailto:[e-mail]> >,draft-ietf-idr-ts-flowspec-srv6-policy <[e-mail] <mailto:[e-mail]> >,"刘松" <[e-mail] <mailto:[e-mail]> >
发送时间：2026-02-24 23:02:10
主题：[Idr] Re: WG LC on draft-ietf-idr-ts-flowspec-srv6-policy-08 (02/10/2026 - 02/24/2026)

Hi Zhenqiang,

 

 

Regarding https://datatracker.ietf.org/doc/draft-li-idr-flowspec-sr-policy/, have you already discussed with authors of draft-ietf-idr-ts-flowspec-srv6-policy ?
IMO, we should not come up with multiple proposals/solutions.

I tend to agree that draft-li-idr-flowspec-sr-policy has a clearer encoding with less ambiguity compared draft-ietf-idr-ts-flowspec-srv6-policy (see my response to this WGLC).

 

In both drafts, clarifications related to interactions with other actions type (especially redirect to VRF) will have to be clarified.

 

 

Brgds,

 

Stephane

 

 

From: 李振强 <[e-mail] <mailto:[e-mail]> > 
Sent: Tuesday, February 24, 2026 11:25 AM
To: idr <[e-mail] <mailto:[e-mail]> >; keyur <[e-mail] <mailto:[e-mail]> >
Cc: idr-chairs <[e-mail] <mailto:[e-mail]> >; draft-ietf-idr-ts-flowspec-srv6-policy <[e-mail] <mailto:[e-mail]> >; 刘松 <[e-mail] <mailto:[e-mail]> >
Subject: [Idr] Re: WG LC on draft-ietf-idr-ts-flowspec-srv6-policy-08 (02/10/2026 - 02/24/2026)

 

Hello everyone,

 

The method proposed in this document is feasible and has already been used in China Mobile's field network. I believe this document can pass the WGLC with further polish and resolution of some technical concerns.

1. A consistent acronym format should be adopted for Flow-spec/Flowspec throughout the document. 

2. Clarification is needed on whether Service_id_x carried in PrefixSID is a mandatory requirement across all SRv6 application scenarios.

3. This document uses the PrefixSID defined in RFC8669 to carry SRv6 SID. It is unclear whether this will result in an update to RFC8669, and this point needs confirmation.

4. For the category of this document, I am not sure standard track is correct. Given that no IANA action is required, BCP (Best Current Practice) or Experimental may be more appropriate for the intended status.

 

To address the same problem, draft-li-idr-flowspec-sr-policy (https://datatracker.ietf.org/doc/draft-li-idr-flowspec-sr-policy/) proposes approaches for FSv2. We welcome comments and feedback on this document, too.

 

Best Regards,

Zhenqiang Li

China Mobile

 

----邮件原文----
发件人：Keyur Patel  <keyur=[e-mail] <mailto:keyur=[e-mail]> >
收件人：idr  <[e-mail] <mailto:[e-mail]> >
抄　送: (无)
发送时间：2026-02-11 05:00:41
主题：[Idr] WG LC on draft-ietf-idr-ts-flowspec-srv6-policy-08 (02/10/2026 - 02/24/2026)

Hi Folks:
 
The working group last call has been issued for draft-ietf-idr-ts-flowspec-srv6-policy-08.txt, “Traffic Steering using BGP FlowSpec with SR Policy”. Please reply to the list with your comments. This adoption call will conclude on February 24th, 2026.
 
The draft can be found at: https://datatracker.ietf.org/doc/html/draft-ietf-idr-ts-flowspec-srv6-policy-08. The implementation report for the draft can be found in Section 4 of the draft. Please note that the version 5 of the draft has changed the status of the draft to “Standards Track” after the discussion on the mailing list.
 
Authors, please reply indicating whether you're aware of any relevant IPR that hasn't been disclosed.
 
Best Regards,
Keyur