IS-IS WG Meeting Minutes Scribe: Acee Lindem (acee@cisco.com) - Intro, Adminastriva, Document Status - (Christian Hopps, Hannes Gredler) - IS-IS Auto-Configuration - Les Ginsberg (See slides) https://datatracker.ietf.org/doc/draft-ietf-isis-auto-conf Les: Feels next revision is ready for last call. Hannes: Implementations? Les: Not that I know of. Bing Liu: Believes there is an implementation in progress but it doesn't reflect the latest version. - Advertising TE protocols in IS-IS - Chris Bowers (See slides) https://datatracker.ietf.org/doc/draft-hegde-isis-advertising-te-protocols Les: Propose that the problem is the same in both IGPs and use cases are the same. Chris Bowers: I agree. There are cases where attributes will clearly be the same for all application (e.g., SRLG). Acee: Disagrees that there are not use cases for different SRLGs per application. Adrian Farrel: Please comment on a solution with separate instances of the IGP for separate applications. Stephane Litkowski: Could use multiple topologies. Chris Bowers: Multiple instance brings more complexity. Shraddha: Doesn't like multi-instance in this case. Les: Duplication of topologies for attribute advertisement also introduces complexity - NVO Control Plane Protocol Using IS-IS - Xiaohu Xu (See slides) https://datatracker.ietf.org/doc/draft-xu-isis-nvo-cp Paul Unbehagen:IEEE experience with IS-IS was seamless. Where is this implemented? Xiaohu: Can be implemented on virtual or real switches Paul: Trying to make this work on virtual switch was difficult. Implemented IS-IS on leaf node. May want to look at what we did. - IS-IS Routing with Reverse Metric - Naiming Shen (See slides) https://datatracker.ietf.org/doc/draft-ietf-isis-reverse-metric Naiming: WG document for quite awhile. Hannes: Implementations? Naiming: Not that I know of. Uma Chunduri: Feels this draft is very useful. - IS-IS Routing for Spine-Leaf Topology - Naiming Shen (See slides) https://datatracker.ietf.org/doc/draft-shen-isis-spine-leaf-ext Naiming: Requests WG adoptions Chris Hopps: Did extensive review and comments will be posted to the IS-IS list. David Lamparter: Similar to HOMENET approach with just OL bit. Believes it is a sound approach. Hannes: Could use mesh groups as well. Naiming: Yes - but mesh groups only solves problem of flooding reduction. This proposal reduces requirements on leaf node implementations. - Carrying Geo Coordinates Information In IS-IS - Naiming Shen (See slides) https://datatracker.ietf.org/doc/draft-shen-isis-geo-coordinates Naiming: Request WG adoption Acee: Only different for OSPF is that a link-scoped RI LSA is used for signaling on a link as opposed to IS-IS using IS-IS IIH. - IS-IS Extensions for Segment Routing - Stefano Previdi (See slides) https://datatracker.ietf.org/doc/draft-ietf-isis-segment-routing-extensions/