IETF 90 ISIS WG - Thursday Afternoon Session Chairs: Hannes Gredler Chris Hopps Scribe: Acee Lindem Intro, Adminastriva Glenn Parsons - Liasons sent to ISIS WG from IEEE. Were they received? Chris: Not received. Hannes: Sent to mailing list? Glenn: Sent to chairs E-mail. One had PDF attached. Chris: What was the purpose? Glenn: New projects and TLVs request. Chris: Technical liasons should posted to mailing list. Could be in form of a draft or other. Glenn: Documents can be sent to list. Donald Eastlake: Document is not yet in IETF liasons database. Document Status - Chris Hopps (See slides) Segment Routing Extensions - Stefano Previdi (See slides) Uma Chunduri: There is text on how to use prefix SID? Need to add text on how to use binding TLV. Stefano: Agrees that the Binding TLV needs more clarity. Has many sub-TLVs and the context must be defined for sub-TLV usage. Uma: Have usecases for Binding TLV. Stefano: Use cases are not needed here. Uma: Needed to reference use cases in Hannes' draft. Hannes: Perhaps what we could do is add some practical examples for RSVP, BGP, and LDP LSPs binding. Not formal use cases. Stefano: Would rather not go into applications in this ISIS draft. Peter Psenak: Should go into a separate document that could be referenced from both ISIS and OSPF. Alia Atlas: There is a SPRING WG for such a document. IS-IS Protocol Extension For Building Distribution Trees - Donald Eastlake (See slides) draft-yong-isis-ext-4-distribution-tree-02 Chris Hopps: Borrowed from TRILL? Donald: Differences from TRILL as these are separate trees per multicast group. Chris: I feel as though this draft has floated by perhaps without enough review, e.g., a router is in one area only but the text speaks about a router in more than one area. Donald: Border router is in two areas. Chris: In fact in IS-IS an L1L2 router is still only 1 area, we can discuss this offline. Hannes: Process has been missing. Has this been review by multicast experts. Donald: Ok. Hannes: Is there enough information for interoperable implementation? Donald: Believe there is enough for WG adoption. Alia: You need to talk about in PIM WG. Andrew Qu: Industry trend for network as a fabric. TRILL is plug and play. Chris: Not trying to shoot it down. It is a multicast feature and should be reviewed in that WG. Andrew: Does not replace PIM. Uses built-in capabilities. Mikael Abrahamsson: ISIS could be used in HOMENET and support multicast. Lucy Yong: This was presented in PIM. Chris: What was the result? Lucy: No objection from PIM. Xiaohu Xu: Applicable and implemented solution. Paul Unbehagen: Implementation is similar to what already exists in SPB. Kireeti Kompella: Move to multicast WG? Alia: Architecture for multicast trees need to be done in PIM. ISIS protocol extensions need to be done here. SBFD Discriminator - Les Ginsberg - (See slides) draft-ginsberg-isis-sbfd-discriminator Uma: Could use node administrator tags. Les: No these are not admin tags - they can only be used to BFD. Uma: Admin Tags are independent of application. Les: These are BFD discriminators only. Chris: These are different names spaces - adminsitrative vs BFD protocol. Rob Shakir: Do not want to mix BFD discriminators with admin tags. Tony Przygienda: Are going to need service access points as well these? Les: How does that apply? Tony: These are addresses? Les: Not IP addresses. Chris: Is Seamless-BFD accepted in BFD WG? Nobo Akiya: Usecase and base Seamless BFD documents have been accepted as BFD WG. Stephane Litkowski: Why not use gen-info TLV? Les: Set of information being advertised is quite small and not really applicable. However, that could be another option. Chris: Does anyone think this should not be a WG document? Chris: Who believes this should be a WG document? Stephane: This is useful. Chris: Appears to be support as WG document but will be confirmed on list. [ good amount of support ] IS-IS Route Preference for Extended IP and IPv6 Reachability - Les Ginsberg - draft-ginsberg-isis-route-preference-00.txt (See slides) Uma: Clarification on slides. Chris: I will also spin a BIS version of RFC 5308. Tony: What happens with old and new routers in same domain? Les: Potentially other ways to address this but this would require protocol changes and would result in more severe interoperability problems. YANG Data Model for ISIS protocol - Stephane Litkowski (See slides) draft-litkowski-isis-yang-isis-cfg-01.txt Kireeti Kopella: OSPF YANG model lead to a lot of fixes to base NETMOD core routing model. Does this require more fixes? Alia: Delighted with this work. There will be a cross-routing WG list for YANG models. Mikael: Great coordination. DT will join draft. Acee: Problems with NETMOD core routing module. Are you talking to them as well? Stephane: Yes - we are talking to Ladislav Lhotka. Chris: Anyone object to accepting? Chris: And people who support? [ overwhelming support ]