Welcome to Etherpad! This pad text is synchronized as you type, so that everyone viewing this page sees the same text. This allows you to collaborate seamlessly on documents! Get involved with Etherpad at http://etherpad.org I-D 1. Presenter Agenda bashing, WG status reports Chairs Deborah: announcement George will be stepping down from MPLS WG chair after 25 yrs of contributions! He will remain the Technical Advisor for the MPLS WG. 2. draft-nslag-mpls-deprecate-md5 Stewart Bryant Stewart: Thanks for all the collaborations George over the years! Stefan: Operational concerns on deprecating MD-5 authenthentication Tony Li: MD-5 is almost considered broken/insecure. We should be pro-active is security Keyur: AO has vendor implementations and should be considered Jeff: inter-op issues with existing implementations of AO/bonica-draft, so there's no usable AO implementation John: X Acee: there's an RFC for keychain with overlapping lifetime. The use of such rolling keys is standardized and should be possible. 3. draft-farrel-mpls-sfc Stewart Bryant Xiaohu: this draft and ours are leverage MPLS - should the 2 draft be combined? Stewart: it's up to the chairs of both WGs. You're welcome to work with us on this, but up to chairs and ADs Geoerge: resistant for waiting for things to rechart.. Rather progress the work and then discuss the possiblity of combine Loa: no need for MPLS to recharter to do this work.. Other WG may need. Himanshu: there's an overlap, it does not matter where docs progress, but this should be discussed/resolved here 4. draft-xu-mpls-sr-over-ip Xiaohu The slides end with the authors asking for working group adoption, I asked if the were correct and Xiahu confirmed 5. draft-wijnands-mpls-mldp-multi-topology-00 Ice Wijnands Loa: like the fact it is ITA standard 6. draft-ietf-mpls-ldp-yang Kamran Raza YANG model of LSP Ping (draft-zheng-mpls-lsp-ping-yang-cfg) Kamran: requesting WGLC Loa: current abstract is very short.. it would be good to expand or move some of the contents into the abstract. George: you get the chance/time to handle that during the WGLC. Loa: WGLC process may include ask to respin if needed. Loa: no need to wait for next IETF meeting to ask WGLC - do it over email Acee: did a early YANG Dr. review on this document.. did not recieve any responses on my early review comment. Kamran: will check 7. Greg Mirsky YANG model of MPLS-TP OAM (draft-zhang-mpls-tp-yang-oam) Greg Mirky Italo: G.8152 for MPLS-TP OAM covers both IETF/ITU- should YANG be structured based on functions or tools Greg: it was not obvious they covered the LSP ping. Italo: they call it connectivity-check - you need a tool.. attributes are same.. whether use BFD or CCM Yuji: G.8152 describes the OAM architeture modelig like ME, MEG, MEP... Please considser G.8152 on this point. Latest draft G.8152 is LSed from ITU-T to IETF. YANG model of LSP Ping (draft-zheng-mpls-lsp-ping-yang-cfg) Greg Mirsky: no comments. From "Loa: no need to wait for next IETF meeting to ask WGLC" is about mLDP, make that clear and divide into two paragraphs, otherwise that part is ok 8. mplswg-related multi-vendor interoperability testing Carsten Rossenhoevel Sam Aldrin: can you share the testplan Carsten: the detailed testplan is proprietry Sam: how much of the label stack depth was tested? Have you tested Adj-SIDs? Carsten: label depth was not tested. Lot of testing was with basic SRv6.. Rudiger: --missed-- Carsten: try to fox on latest drafts/developments on SR. 9. draft-xie-mpls-ldp-bier-extension draft-xie-mpls-rsvp-bier-extension Jingrong Xie/Zhenbin Li Tarek: any consideration for make-before-break for RSVP-TE P2MP LSPs? or S2L subLSP graft/prune? Robin: the draft in its current state defines highlevel the objects, but later versions will consider it Loa: should this work be in TEAS or MPLS or BIER Pavan: at this time, no specific preference/detail enough Loa: need for coordination between the 3 WG chairs and WG. preference is to currently progress in MPLS WG 10. draft-ama-mpls-fm-rdi-00 Mahesh Jethanandani Greg: example is not correctly using the solution/AIS.. Mahesh: talking about associated non-corouted Greg: need to use it symmetrically - need to use both ends Himanshu: this is an inband fault detection scheme Greg: RDI - you're changing the content.. Greg: what happens if LER-A Italo: it is not clear the scenario (sent email to the list).. low-power (you're doing 100 pkts/sec) Mahesh: 500MHz CPU.. limited CPU - not many sessions Italo: bit more complex state machine Mahesh: you dont need another protocol.. you can reuse the same protocol. Not for every router Himanshu: this is needed for hierarchy.. By doing this, you can scale. Italo: why use FM instead of PSC? Greg: is the intended for the proposal standard or experimental? Greg: if multiple protocols exist, then it's a race condition George: if both are running in 2 layers. then implementers will be upset. The get 2 pulses and it not going to be acceptable Himanshu: may request a new flag Greg: there's a certain behavior associated with RDI More discussions needed.. No conclusions on this.