> CCAMP Agenda For IETF 91 > Version: Oct 27, 2014 > > First Session > THURSDAY, November 13, 2014 > 1640-1910 - Thursday Afternoon Session III > Room: Coral 5 > > Num Start Duration Information > 0 16:40 10 Title: Administrivia & WG Status > Draft: > Presenter: Chairs David Sinicrope: BBF Liaison on packet-optical integration. It is an implementation agreement (or a profile) about packet-optical integration/optimization. It defines 3 different scenarios: transponder integrated into the router, present as a shelf or integrated into the optical node. The document provides reference architecture and nodes requirements for each of the 3 scenarios. The BBF is almost done with scenario 1 and there is a number of MIBs they'd like to reference so they're asking for status update and a forecast on when CCAMP thinks that MIBs might be completed. Same liaison sent to ITU-T SG15/Q6 to ask for application codes. Paul Doolan: It is not just MIBs, the liaison is referring also to two other individual documents about LMP and RSVP-TE extensions. > 1 16:50 30 Title: WG reorganization discussion (CCAMP, TEAS & MPLS) > Draft: > Presenter: Chairs Loa Andersson: should the generic Yang models be in TEAS or LIME, what's the difference and the definition of generic? Lou Berger: LIME yang model is a generic OAM model, while TEAS is a generic TE model. Adrian Farrel: Generic means looking at a TED unaware of the underlying technology type, and there might be yang extensions to manage technology details. Loa Andersson: Please enhance in TEAS charter. Loa Andersson: What about MPLS-TP? It seems it will be done all in MPLS. Do PALS and BESS do any MPLS-TP work? Adrian Farrel: BESS is unlikely, PALS is possible (PWs are in the TP space). Andy Malis: Speaking as PALS co-chair, only the statically provisioned PWs portion of MPLS-TP have been done outside MPLS and will continue in PALS. Daniele Ceccarelli: BGP-LS extensions for non-packet belong to CCAMP or IDR? Adrian Farrel: IDR. Extensions to those protocols will be done in cooperation with WGs owning those protocols. Agreement among chairs decides in which WGs such extensions are made. I guess IDR chairs wants the BGP-LS extensions to be done in IDR. Please shout on the ML and we'll make the charter clearer. Amy Ye: Drafts on bandwidth availability can be possibly made generic? Comments received to extend to optical error. That's non packet and then should remain in CCAMP? Lou Berger: right now they are technology specific and belong to CCAMP, if you include multiple technologies they belong to TEAS scope. Dieter Beller: What happens to WG docs being moved? Start back from individual status? Lou Berger: No. WG status only means chairman approval to work on it. Adrian Farrel: keeping things general is important. I will ask TEAS chairs to keep an eye on what is going on in MPLS and CCAMP on technology specific work and see whether it needs to be moved to TEAS for generalization. Ruediger Kunze: Is there enough work to keep CCAMP open? Adrian Farrel: I expect there is enough work to justify a WG for a while. Lou Berger: New working might be coming for CCAMP: 6TiSCH and detnet (BoF), to me it looks like the work we did for MEF. Adrian Farrel on yang models: we had a lot of side meetings on yang models and we agreed with the OPSA ADs that a strong coordination is needed. At the time being whoever wants to push yang models, please write drafts, present/advertise them to all the WGs you think should be involved and post to the YANG coordination mailing list. Make sure everyone is aware of your work. Hopefully we'll have some structured work in the next months before Dallas. Lou Berger: Polling: [1] How do people feel about the discussed breakdown: majority in the room ok. [2] Objections to proposed breakdown? None. [3] Concerns that have not been discussed yet? no one. Adrian Farrel: Process will move on. > 2 17:20 20 Title: WG draft issues & actions > Draft: > Presenter: Editors/authors, Chairs - draft-ietf-ccamp-rsvp-te-domain-subobjects Dhruv Dhody: Peering document in PCE updated and discussed. How do the chairs prefer to proceed wrt sync of the two documents? The PCE document is waiting for IRO updates but the domain sequence is ok. This document does not include IRO, there is no dependency. Adrian Farrel: You can take the doc to the last call and then put it on hold waiting for the other document to be completed. - draft-ietf-ccamp-te-metric-recording-04 Matt Hartley: We were waiting for the SRLG documents completion. Once they're ready we'll update this one. Lou Berger: SRLG document stable, please proceed with update. - draft-ietf-ccamp-otn-signal-type-subregistry-00 Lou Berger: no need for a designated expert; docs should be cleaned up and LC'd fairly fast. This document changes the allocation policy of one of the registry. It changes from "standard track" only to "standard track + specification exist" and allows an informational document to allocate an IANA codepoint, which at present only standards-track docs may do at present. Do we want to make this change across all registries managed by CCAMP? George Swallow: Is there any difference between informational drafts from WG and coming from any other direction? Adrian Farrel: there are three ways for informational drafts to go ahead. WG, AD sponsorship and independent stream editor. Lou Berger/George Swallow: Some concern that we don't want to create a system whereby codepoints could be allocated without the WG's knowledge/approval. Lou Berger: Two actions to be taken. 1) this doc should be changed to be "standards-action + ietf-review", 2) and we should extend that to all CCAMP registries. Anyone in the room willing to do this document revision? This shouldn't hold up the doc and we should be able to get it last called fairly fast. Any volunteers? Matt Hartley: I'll do it. > 3 17:40 10 Title: RSVP-TE Extensions for Collecting SRLG Information > Draft: https://tools.ietf.org/html/draft-ietf-ccamp-rsvp-te-srlg-collect > Presenter: Matt Hartley Matt Hartley: would appreciate comments from the WG on what to do about the recent IPR disclosure relating to this Lou Berger: Does anyone have any concerns about it? (no hands) > 4 17:50 10 Title: RSVP-TE Path Diversity using Exclude Route > Draft: https://tools.ietf.org/html/draft-ietf-ccamp-lsp-diversity > Presenter: Dieter Beller Lou Berger: Does each node need to understand each diversity type? Only the edge nodes need to do that, please fix in the text. You should be clear of what processing is expected on a transit domain. Lou Berger: high-level comments: 1. please don't put suggested values in, unless you're defining an entirely new IANA-controlled field. 2. I'd be interested to see a reference for the RSVP-TE FEC. > 5 18:00 10 Title: Requirements for Very Fast Setup of GMPLS LSPs > Draft: https://tools.ietf.org/html/draft-malis-ccamp-fast-lsps-03 > Presenter: Andy Malis Deborah Brungard: Section 5 is still speaking about potential solution Andy Malis: Will take out. Deborah Brungard polling: [1] how many have read the draft: high number [2] how many think it's ready for WG adoption? Less (half) [3] how many have concerns? None. Paul Doolan: what depth of operator support do you have? Andy Malis: AT&T + Verizon, and DISA. > 6 18:10 10 Title: RSVP-TE Signaling Procedure for GMPLS Restoration and Resource Sharing-based LSP Setup and Teardown > Draft: https://tools.ietf.org/html/draft-zhang-ccamp-gmpls-resource-sharing-proc > Presenter: Pawel Brzozowski Lou Berger: Document in the middle of adoption. It is clear that the WG is interested in the topic. Suggestion to authors: figure out clearly what they want to achieve. There is a lot of text, sometimes redundant, some others wrong. Don't take it as lack of encouragement. The poll will last one more week. > 7 18:20 10 Title: Extensions to Resource Reservation Protocol For Fast Reroute of Traffic Engineering GMPLS LSPs > Draft: https://tools.ietf.org/html/draft-tsaad-ccamp-rsvpte-bidir-lsp-fastreroute > Presenter: Rakesh Gandhi Lou Berger: polling: [1] How many think this is an interest topic? reasonable number, not huge [2] How many read? slightly more [3] Anyone reservations? none. Good one to discuss on the list. Himanshu Sha: GMPLS has the segment protection. Are we willing to extend FRR to cover the corouted bidirectional LSPs? Lou Berger: If you have a look at the original work, it allows for the coexistence with FRR. At the time we did that document we said ourselves that sooner or later someone would come out with how FRR works with GMPLS. That's how I read this document. > 8 18:30 10 Title: A Yang Data Model for Abstract TE Topologies > Draft: https://tools.ietf.org/html/draft-liu-yang-abstract-te-topo-00 > Presenter: Xufeng Liu / Vishnu Pavan Beeram Dhruv Dhody: are we ready to take draft-clemm (network topology YANG model) as a base? I'm not very happy about that draft. There are discussions in netmod going on. Pavan Beeram: We just needed a place to define node, links, contamination point. Draft-clemm is already doing that. Lou Berger: Does it make sense to have a network topology separate from a TE topology? Adrian Farrel: All the 3 drafts are exceeding their mandate. I don't have an answer to the question. I can't see a proposal of which information is in which of the models. I don't fully understand the proposal. If I was doing the modelling from scratch I would see any difference between the TED topology and the abstract TE topology. Lou Berger: it is a cross WG work, we don't own all of that. Adrian Farrel: we should have an interim meeting (virtual or physical) to sort this out. Dieter Beller: Relationship with ACTN? They are willing to deal with abstracted topology. Lou Berger: We have a charter for TEAS, CCAMP and MPLS. Abstract TE topology modelling fits in TEAS charter. > 9 18:40 20 Title: G.698.2 SNMP MIB, LMP and Yang drafts > Draft: https://tools.ietf.org/html/draft-galikunze-ccamp-g-698-2-snmp-mib > https://tools.ietf.org/html/draft-dharinigert-ccamp-g-698-2-lmp > https://tools.ietf.org/html/draft-dharini-netmod-g-698-2-yang > Presenter: Gabriele Galimberti Dieter Beller: The only parameter you're adding to the application code is the optical power at TX and RX. What is the purpose of communication that piece of info? It should be documented in the ID. Gabriele Galimberti: The TX and RX optical power values can be very useful to understand if the optical trail is correctly working. Dieter Beller: The receiving is giving you indication. I still have concerns since it does not say anything about OSNR or optical impairments. Gabriele Galimberti: This is only one of the parameters, others are needed. Paul Doolan: We need to ask SG15/Q6 to specify the parameters needed to state a path feasibility. There is no optical power defined in Q6. Gabriele Galimberti: This set of parameters is the output of the joint meeting with Q6 in Orlando. Deborah Brungard: Same question always comes up. I suggest to take power out of the document and then we adopt the document. This has been the concern raised over and over with this document. Then work on a liaison to Q6 to understand if should include. Fatai Zhang (via Jabber):: I think it is better to make these draft more generic, not just for G.698.2 Deborah Brungard: do you want to include the power levels? Fatai Zhang (via Jabber):: anything relate to general WSON. for example, MIB work, why not define a WSON MIB draft? Lou Berger: these parameters are not in g874 and that's the concern Fatai Zhang (via Jabber):: contributions to ITU should be done to define them, if it is wanted. Otherwise the drafts defined in IETF cannot include. Lou Berger: Poll: [1] Assuming the power is removed, is there interest in working on the MIB? There is interest. [2]What about LMP? Still interest but just the authors? [3]Are there other objections (re LMP and SNMP) other than power? None. > 10 19:00 10 Title: Proprietary application codes considered dangerous > Draft: https://tools.ietf.org/html/draft-doolan-proprietary-ac-00 > Presenter: Paul Doolan > Adjourn 19:10