Minutes of the Netext (Network-Based Mobility Extensions) working group meeting at IETF87 ----------------------------------------------------------------- TUESDAY, July 30, 2013, Afternoon Session I (1300-1500) Room: Tiergarten 1/2 Credits for the minutes: 1. Charles Perkins 2. Sandra Cespedes The chairs would also like to thank Suresh Krishnan for chairing the meeting at short notice because of travel issues faced by the current co-chair, Basavaraj Patil who was able to make it to the meeting for just the last 15 minutes. Rajeev Koodli (co-chair) participated remotely via jabber/meetecho. - Update on working group documents presented by Suresh. Discussion of the WG documents follows: - Revision of documents that will be on Last Call. The process will be on the mailing list. Juan Carlos-> There is the WG draft pertaining to the logical interface from 2007. It has received comments but it has been stalled for at least a year. That's something that should be addressed. Sri-> Julien said all the comments have been resolved. So the document should be ready for the last call. Suresh --> Is there something you have to present right now? If not, then write it to the mailing list. Sri--> The only issues that were brought up have been already resolved. Changes will be highlighted at the mailing list. Comment from Rajeev: Chairs think that we need some discussion on on LIF --------------------------- 1. Proxy Mobile IPv6 Extensions to Support Flow Mobility I-D: draft-ietf-netext-pmipv6-flowmob CJ Bernardos Charlie --> How can you tell if the policies are consistent Carlos --> You can use many mechanisms. Depend on the deployment but the draft doesn't defining a specific mechanism. Suresh --> It is a difficult issue. That is way the decision is to leave that outside of the scope of the document Juan Carlos --> The idea here and the reason for concern is that the document needs to clarify the need for the policies to be consistent. Behcet -> (Related to his comments on the draft) I'm tired of talking about this. There is a fundamental misunderstanding of what is this issue. I have posted it to the list and didn't get a reply after 3 months. Suresh --> Look for the information on the tracker. Behcet --> I can't find it now. Maybe next week. Carlos: I copied the description from the tracker. It is about using the BID when you have different bindings for different connections. We are using the binding updates to use them with different flows. Suresh: Please take 5 minutes to read the comment of Behcet and indicate if you think there is a problem. Sri: I don't see issues about using BID. State is created on the LMA. I don´t understand why are you arguing on this issue for many months. I think the draft is fine as is. Juan Carlos: I'm trying to read through. I don't see any issue on this one. I don't think there is any problem. Ruji: I didn't follow this problem. I don't see the problem here. Marco: I don't see the issue. It's just about the identification. Charlie: You are using the BID to identify the flow? Carlos: No, it's for identifying the MAG, for the flow you have flow IDs. Charlie: You can use the BID for a flow ID. Carlos: You need to identify the different flows to identify the bindings. Marco: It is only used for the identification of the binding, so the LMA can identify it. John: I haven't follow in detail but I don't see any problem with this. Suresh: I personally think it's fine, but let the list decide and give Behcet time to think more about this. Suresh: What is the relation of this document with the logical interface's draft. Is that required for this draft? Carlos: There is no normative requirement. --------------------------- 2. Quality of Service Option for Proxy Mobile IPv6 I-D: draft-ietf-netext-pmip6-qos-03 Marco Liebsch Charlie: Bearers are not trivial to be explained. Marco: We received the comment to clarify the role of bearers in the document. Bearers and TFT are related to cellular environments, so here we should not talk about those specific things but instead about policies per-mobile node and per-flow. Suresh: I think it's good if you receive more comments on this draft. Volunteers: Charlie, Josh, Rajesh. Get back with comments in three weeks. --------------------------- 3. Separation of Control and User Plane for Proxy Mobile IPv6 I-D: draft-wakikawa-netext-pmip-cp-up-separation-00 R. Wakikawa Suresh: The goal is to gauge whether there is WG support to adopt this draft at the end of this session. Ryuji: separation of control plane and user plane also need to update for the alternate CoA option - UP and CP addresses of LMA can be co-located, not specified - WiFi scenario (WLC == wireless LAN controller) - has been proposed repeatedly proposed since 2008 - Charlie has agreed to work togethero on this document Rajeev (jabber): Is there a a need for the WLC and MAG to be collocated? Ryuji: No, we use it here just as an example. Marco: I think it's a useful work. How far should this extension go? Ryuji: We are thinking on defining a single mobility option. Just the signaling. Suresh : Calls for voting on adopting the document. The room agrees on adopting the draft. --------------------------- 4. Dynamic CoA Support for PMIPv6 I-D: TBA Sri Gundavelli Sri presenting: Indicates the draft is not ready yet. Bulk Revocation needs a MAG identifier, which is not specified MAG identifier proposal is close to the format in RFC 4382 No comments from the WG. Need a draft before further discussion. --------------------------- 5. Mapping PMIP Quality of Service in WiFi Network I-D: draft-kaippallimalil-netext-pmip-qos-wifi-02 J. Kaippallimalil John Kaippallimalil - has been presented in [netext] before - is complementary to the PMIP QoS draft - WiFi AP is modeled as a PEP, as well as WLC - QoS information from WLC to WiFI AP - PMIP 802.11e mapping - WiFi AP does not know how to control traffic according to WLC / PEP parameters - What about QoS from the client (i.e., upstream)? - Marco Liebsch: current QoS draft should be aligned -- John K. It is intended to be aligned - Marco: No mechanism yet on the MAG to retrieve the QCI - Brian: comment from Rajeev: does the draft assume a particular 802.11 implementation? - Sri, Marco: Looks like a reasonable effort - Charlie: What is the connection between AP <--> WLC versus PMIP? If IP-within-IP replaced PMIP, would anything change - Document review: Charlie, Ashutosh, Rajesh, Marco Meeting adjourned at 2:13pm ---------------------------