Re: [Idr] WG adoption call for draft-scudder-idr-entropy-label-01 (9/6/2022 to 9/20/2022)

[[e-mail]]

Sue,




Orange Restricted
From: Idr <[e-mail]> On Behalf Of Susan Hares
Sent: Tuesday, September 6, 2022 10:32 PM
To: [e-mail]
Subject: [Idr] WG adoption call for draft-scudder-idr-entropy-label-01 (9/6/2022 to 9/20/2022)

This begins a 2 week WG Adoption and IPR call for:
https://datatracker.ietf.org/doc/draft-scudder-idr-entropy-label/

The co-authors should respond to this message with an IPR statement.

The specification revised the BGP attribute for Entropy Label Capability.
The abstract of the document states:

  "This specification defines the Entropy Label Capability Attribute
   version 3 (ELCv3), a BGP attribute that can be used to inform an LSP
   ingress router about an LSP egress router's ability to process
   entropy labels.  This version of the attribute corrects a
   specification error in the first version, and an improper code point
   reuse in the second.



1)    We already have a solution for this; an IDR WG document. https://datatracker.ietf.org/doc/html/draft-ietf-idr-next-hop-capability-08

"This document also defines a Next-Hop capability to advertise the

   ability to process the MPLS Entropy Label as an egress LSR for all

   NLRI advertised in the BGP UPDATE.  It updates RFC 6790<https://datatracker.ietf.org/doc/html/rfc6790> with regard

   to this BGP signaling."

I don't recall that the WG expressed an issue on this solution. Nor asked for the use of a non-transitive attribute.
Now if there is a compelling reason to use a transitive attribute, this can be discussed by the WG as part of regular WG work on draft-ietf-idr-next-hop-capability. If needed, both drafts may also be merged.


2) Initiating a technical comparison of transitive vs non-transitive attribute:

Non-transitive:
- require the software upgrade of PE, ASBR, RR

Transitive:
- require the software upgrade of PE, ASBR.
- require the re-implementation, test and debugging of the non-transitive-like filtering.

All in all, small gain in term of deployability at a small cost in term of implementation. Obviously, "small" is in the eye of the beholder.
Regarding software upgrade, the only difference is the BGP Route Reflector which: is a small number of nodes (typically at least two order of magnitude smaller), control plane only / no impact on customers, typically centralized in a very small number of locations, control plane only so very possibly we are talking about a VM or container. I would not call this a 'forklift upgrade'.

Again, this is a discussion that we can have on the WG document that we have (draft-ietf-idr-next-hop-capability)

3) lack of feature coverage
draft-ietf-idr-next-hop-capability is a generic tool allowing the advertisement of different kind of BGP Next-hop dependent capabilities. Entropy label is one, but the MPLS WG is currently defining new data plane features (In Stack Data, Post Stack Data) which will require to advertise different type de capabilities, possibly with related parameters. Therefore the general tool is needed. (and some people in the MPLS WG have even proposed to deprecate the Entropy Label for the New Thing, so defining a new BGP Capability just for Entropy Label may seem a bit late). There are other usages for other dataplane such as IOAM (another IDR WG using draft-ietf-idr-next-hop-capability)

4) In all cases, I don't think we need two solutions for this simple problem. (the operational differences are small, a few %)


In your comments consider:
1) Does this specification fixes errors in versions 1 and 2?

2) Are there any additional errors or weakness in this specification
of version 3?  For example, has this specification clearly described what
happens if version 1, 2 and 3 exist in a network?

Not really. As already expressed on the list, what happens if ELCv2 and ELCv1 exist in the network is BGP sessions reset with a major impact on the network.
So ELCv2 is not to be used. But this draft is more or less saying the opposite in Appendix A, therefore I object the content of Appendix A.
ELCv2 is a proprietary solution. Migrating away of ELCv2 may be handled without the IETF.

3) Will deployment of version 3 of Entropy Label Capability
BGP attribute aid in fixing problems in current networks?

4) Are there enough implementations that this draft should
Be moved quickly to WG LC?

There are no implementations of ELCv3.

Regards,
--Bruno

Cheerily, Sue


_________________________________________________________________________________________________________________________

Ce message et ses pieces jointes peuvent contenir des informations confidentielles ou privilegiees et ne doivent donc
pas etre diffuses, exploites ou copies sans autorisation. Si vous avez recu ce message par erreur, veuillez le signaler
a l'expediteur et le detruire ainsi que les pieces jointes. Les messages electroniques etant susceptibles d'alteration,
Orange decline toute responsabilite si ce message a ete altere, deforme ou falsifie. Merci.

This message and its attachments may contain confidential or privileged information that may be protected by law;
they should not be distributed, used or copied without authorisation.
If you have received this email in error, please notify the sender and delete this message and its attachments.
As emails may be altered, Orange is not liable for messages that have been modified, changed or falsified.
Thank you.