Mobile Ad hoc Networks (MANET) Agenda Meeting : IETF 74 Thursday March 26th 2009 Time : 1300-1500 Location : Continental 1&2 Chairs : Ian Chakeres ian.chakeres@gmail.com Joseph Macker joseph.macker@nrl.navy.mil Minutes : Justin Dean Jabber : manet@jabber.ietf.org Scribe : Alexandru Petrescu Audiocast: http://videolab.uoregon.edu/events/ietf/ URLs : http://www.ietf.org/html.charters/manet-charter.html http://www.ianchak.com/manet http://tools.ietf.org/wg/manet/ https://datatracker.ietf.org/public/meeting_materials.cgi?meeting_num=74 ========================================================= AGENDA o Administriva - Mailing list: manet@ietf.org - Scribe(s) - Blue Sheets - State your name at the microphone - IPR - RFC3979 o Agenda bashing o WG Status (Ian Chakeres) o PacketBB, TimeTLV, NHDP, OLSRv2 Discussion (Thomas Clausen) - RFC5444 - RFC5497 - draft-ietf-manet-nhdp-08.txt - draft-ietf-manet-olsrv2-08.txt o SMF Discussion (Joe Macker) - draft-ietf-manet-smf-08.txt o DYMO Discussion (Ian Chakeres) - draft-ietf-manet-dymo-17.txt o MIBs Discussion (Robert Cole) - DYMO MIB - NHDP MIB - SMF MIB - OLSR MIB o Open Microphone - Discussion, Related Work & Announcements - Protocol extensions ===Minutes=== Agenda Bashing Ian Chakeres (IC) presents Agenda for the meeting IC presenting wg progress IANA TIMETLV and packetbb have been published Would like to wg last call NHDP soon WGLC comming soon for SMF as well Should MIB documents become wg documents? ~15-20 yes 0 no Looking forward: extensions, starting as private drafts Thomas Clausen (TC) presenting packetbb, timetlv, nhdp olsrv2 update TC: the authors would like to send the document off to the IESG Teco boot: What are we doing about an official port numbers and udp ip assignments IC: we will bring this up on the list CD: if we make a choice we need to make the same choice for all the different documents TC: Status OLSRv2 status, will add metrics and no further changes added. TC: Interop info will be posted to the list Joe Macker presenting SMF slides: JM: We have tested working code using multiple gateways using pim and using multiple interfaces and differing algorithms JM: Any questions regarding SMF? no questions JM: Working on making experimental last call soon. Ian Chakeres presenting slides on dymo update: IC: hasn't been updated with final packetbb and timetlv documents IC: would like to see extensions for DYMO which up to this point were put off IC: I would like to encourage people to submit personal draft extensions TC: are we as a working group aiming to look for our next items? What are we trying to do with this all for documents? JM: we put a few papers off in the past which now would be a good time to bring forward TC: Shouldn't we wait until these documents are finished before we recharter? JM: I don't think its a black and white issue where we need to recharter TC: I will take this as a yes that we are opening the door for more discussion JM: I agree we are certainly opening the door Bob Cole presenting MIB document progress BC: now would be a good time to look through the dymo mib BC: The capabilities group was added to smf mib to allow a management device to query a for usable algorithms and allow selection IC: if you could point out to the list what those RMON are going to required. BC: SMF has the machinery taken from RMON right now, the reports table identifies the kind of report table which is available for reporting BC: RMON reports are typically tied to mibs and it might be cleaner to have it separate and just pointing to counters in various mibs. Open Mic: Charles Perkins: when are we opening up the floor for more discussions, specifically multicast. JM: We are already open to extensions to SMF. ===Jabber=== [12:31:38] *** ALDE (a.petrescu@jabber.org/ALDE) has joined the room as a participant [12:32:46] *** ALDE (a.petrescu@jabber.org/ALDE) has left the room [12:33:06] *** Alex Petrescu (a.petrescu@jabber.org/ALDE) has joined the room as a participant [12:39:09] absolutely [12:39:12] good thanks [12:39:17] yes I hear great [12:40:13] test, anyone hear? [12:40:20] yep [12:40:23] ok great [12:43:42] *** Alex Petrescu (a.petrescu@jabber.org/ALDE) has left the room [12:44:09] *** Alex Petrescu (a.petrescu@jabber.org/ALDE) has joined the room as a participant [12:52:36] *** Ian Chakeres (ian.chakeres@gmail.com/AdiumC0DA403D) has joined the room as a participant [13:08:12] MANET WG meeting SF 26 march 2009, Chairs are Ian Chakeres is IC and Joe Macker is JM. [13:08:42] IC presents slides at http://www.ietf.org/proceedings/09mar/slides/manet-0.pdf [13:09:21] IC: any comments on agenda? [13:10:04] IC: presentations are online at https://datatracker.ietf.org/meeting/74/materials.html search 'manet' [13:10:57] *** Thomas.Heide.Clausen (thomas.heide.clausen@gmail.com/MacBook Pro 17") has joined the room as a participant [13:12:15] *** shamus (shamus.sakai@jabber.org/Exodus) has joined the room as a participant [13:12:53] IC: question to group, should poll, raise hands, do people think these docs should become wg docs? [13:12:57] IC: opposed? [13:13:04] IC: 15-20 affirmative, no opposed [13:14:15] (someone in the physical room would be great if s/he typed 'slide title is...' whenever slide changes in the physical display room) [13:14:28] IC: next in agenda is Thomas Clausen is TC [13:14:49] TC presents slides at http://www.ietf.org/proceedings/09mar/slides/manet-4.pdf I believe [13:15:08] yes those are the slides [13:15:27] is slide displayed "Document Progress Overview"? [13:16:07] Progress: slide is now up [13:16:17] draft-ietf-manet-nhdp [13:18:13] Teco Boot is at the mic [13:18:17] Teco Boot is TB [13:18:38] TB: nhdp... now we have an official port number... maybe we must make decisions here... [13:18:49] TB: we can't say: maybe use UDP, or other..., we need to decide [13:19:07] TC: an element of answer: any change of kind necessitates a line or two... with respect to nhdp it's not an issue. [13:19:20] TC: maybe we do want to say now we shall run this over udp, or ip, whatever... [13:19:34] IC: I'll bring this to the list, have a discussion, select a choice, ok teco? [13:19:42] IC: we need to have a discussion of it [13:19:45] Chris Dearlove at mike [13:19:47] Chris Dearlove is CD [13:19:59] CD: if we need to decide, then we need to make all same choice [13:20:19] TC: for olsrv2 we essentially did same stuff as for ... [13:21:58] Slide: Status draft-ietf-manet-olsrv2 [13:22:17] Slide Interop/workshop 09 [13:23:58] Is this Justin Dean is JD? [13:24:04] Joe Macker [13:24:24] JM is presenting slides at http://www.ietf.org/proceedings/09mar/slides/manet-3.pdf I believe [13:25:10] yes [13:25:27] Slide: SMF Update [13:26:18] JM: questions? [13:26:29] JM: we'll get a final call on this, EXPERIMENTAL initially. [13:26:38] JM: no questions, on to next item [13:27:05] IC presents slides at http://www.ietf.org/proceedings/09mar/slides/manet-1.pdf I believe [13:27:27] http://www.ietf.org/proceedings/09mar/slides/manet-2.pdf [13:28:20] *** hiroki (keitam_nanashisan@jabber.org/Home) has joined the room as a participant [13:28:33] *** hiroki (keitam_nanashisan@jabber.org/Home) has left the room [13:28:39] Slide: Extensions [13:30:09] Joe Macker: informational documents are also welcome [13:30:12] JM: in addition to extensions, people need informational documents [13:31:11] TC: extensions discussed now? Or what do we try this? [13:31:26] JM: not disturb progress., but as chairs, people submitted things we've put off not ready at that time [13:31:34] TC: re-charter we want to determine? [13:31:37] JM: depends what means [13:32:12] JM: let's get all the course stuff right before we do that, makes more sense [13:32:24] JM: not a bw issue, we need to work with ADs not cross lines anyways [13:32:32] TC: I'd like to take that as yes, open the door [13:32:41] JM: certainly door is open now for more discussion [13:32:57] Bob Cole? [13:33:35] BC presents slides at http://www.ietf.org/proceedings/09mar/slides/manet-1.pdf I believe [13:33:58] slide: updates [13:37:30] slide: next work items [13:38:47] slide: questions [13:41:48] BC: how to build all that machinery in one MIB? [13:41:57] commenter commented silently [13:42:23] IC: if you could point out in email on list about what needs to be added about RMONs? Personal draft should do... paragraph you need to put in a set of objects? [13:42:26] IC: or? [13:42:34] BC: the machinery is in the smf mib now. [13:42:41] BC: there's a reports group, has two tables... [13:42:54] BC: specifies amount of reports, amount of storage, link types, how long it should keep [13:43:06] BC: report tables id's kind of performance it will collect in the report [13:43:17] BC: what I didn't like about it... it's not quite right in the SMF [13:43:27] BC: I put a couple of counters, there's a difference [13:43:36] BC: RMON typically reports what's the average, variance. [13:43:54] BC: then you have to go in and build an explicit metrics objects for each metric you'd want for each different MIB [13:44:15] BC: cleaner would be to... and just define a way for each counter want to have each value on a period of time, or the rate, or variations, o [13:44:21] BC: and then up to management stations to . [13:44:28] BC: if so then separate mib, [13:44:32] IC: points to different [13:44:37] BC: variance report... [13:44:41] BC: not sure answer? [13:44:47] IC: was informative how reports are done [13:45:01] BC: difference RMON reports are tied to MIBS, propose I separate, not sure doable [13:45:14] BC: I like the idea/notion of JM of do offline reporting, autonomous mode. [13:45:48] IC: is there a method more ... you'd ,... client, mib on device would hold on various mib item values over time, and then give you all values at once [13:45:59] BC: right now, list of counters, you hit them once on timers... [13:46:17] BC: could do that with walk through stuff, multiple objects at time... if you want to... every 10 seconds [13:46:35] BC: RMON reports allow to ... over the hour, or over the distribution over 5 minutes increments... [13:46:38] BC: thanks [13:46:40] IC: thanks [13:46:52] BC: I'll write up something describing what's mentioned [13:46:58] IC: flesh out on the list [13:47:13] IC: open microphone, any topics? [13:47:32] IC: otherwise, my words to you- please review the nhdp doc, the smf, and the dymo/olsrv2 [13:47:39] IC: focus on ... and smf would be first [13:47:51] IC: anyone signed the bluesheets , you have the bluesheets? [13:47:58] JM: early [13:48:09] JM: bluesheets in the back [13:48:12] Charles Perkins is CP [13:48:32] CP: when do we really start discussing the potential for adding more items... I thought multicast [13:48:34] JM: open up the floor for that [13:48:57] JM: review constantly nhdp and smf and all that, but we need other areas we need to look at, consistent with what we do [13:49:12] CP: how would you recommend we maintain consistency if we were to tackle multicast? [13:49:18] IC: example: adding groups to SMF protocol [13:49:24] IC: flooding based on groups [13:49:33] JM: how would you do ring search expanding... [13:49:39] JM: did in the past but no documents now [13:49:46] JM: we held off some docs in the past [13:49:55] CP: gateways operation [13:50:12] CP: worthwhile think about it now, because AUTOCONF there needs to be some coordination at least [13:50:19] CP: or a nice short dividing line MANET_AUTOCONF [13:50:34] JM: agree totally, certainly I have people doing PIMSMF operational networks [13:50:43] JM: best practice there is what, manage networks at the border [13:50:55] JM: there may be new mechanisms, or build things service discovery issues. [13:51:05] JM: whether here or autoconf ... discussion, here we do routing [13:51:21] CP: emphasize history bit here, in a different time universe, there was discussion here about gateway [13:51:40] CP: determination was different ... address assignment.. but algorithm didn't run smoothly, autoconf didn't run ok... [13:51:49] CP: very important area of gateway operation of manets suffered. [13:51:53] JM: not disagreeing [13:51:59] Justin Dean is JD [13:52:24] JD: please read nhdp document that was put up today, we go on last call... we need possible shape before sending off [13:52:27] TC: agree with JD [13:52:34] TC: what do you mean CP gway operation? [13:52:55] CP: many ways, I think work Ryuji and I did long ago, a default router to the Internet, ways it advertises its presence [13:53:03] CP:... [13:53:12] TC: I think we are in that case a reactive protocol specific consideration. [13:53:27] TC: because in OLSRv2 we do advertise default gateway, prefix which has length 0 [13:53:32] TC: existing protocol... [13:53:46] TC: I do understand though for reactive all nodes might not want to discuss something specific... [13:53:54] CP: agree, more urgent for reactive protocols is. [13:54:06] JM: this issue is none in the wg [13:54:19] JM: understood was ok, but another issue: multiple gateways... [13:54:29] JM: that's possibly a work between AUTOCONF and here [13:54:39] chris dearlove is at the mic [13:54:40] Chris Dearlove speaking [13:54:41] CP: multiple gways, proactive makes it easy but still issues [13:54:48] CD: I will ... [13:54:54] CD: metrics will help. [13:55:16] TB: multigateway then multihoming, there are many groups in multihoming [13:55:33] TB: not think metrics... maybe this ingress filtering problem. OR surprise me please. [13:55:47] CD: there's certainly an issue with mgateways the other way around, manets in the Internet. [13:56:17] JD: extensions or possible extension I'd be interested in: Security. We don't have extensions regarding security. We have some thoughts and sec cons [13:56:28] JD: we had some solutions before, it would be good now to consider [13:56:34] Ulrich Herberg is UH [13:56:47] UH: I work on TLV for signatures, extensions for and olsrv2 [13:56:54] UH: maybe I publish before next ietf [13:57:13] UH: BAE systems definitely interested. But there's no one single solution. There'd be hooks [13:57:25] (that last UH is CD) [13:57:39] CD: even one solution would be better [13:57:48] JM: radio tech different reqs, different threats. [13:57:55] JM: building blocks would be a good idea [13:58:17] JM: one IETF in time there was about security... maybe now people more savvy... on security, superior approach. [13:58:23] Jock Peaner of Drexel (JP?) [13:58:41] JP: manet applications may be interested in mcast broadcast but considering group object consistency maintenance [13:58:47] *** Thomas.Heide.Clausen (thomas.heide.clausen@gmail.com/MacBook Pro 17") has left the room [13:58:57] JP: foundation block or app block directly, similar to SMF but different semantics [13:59:06] JP: protocols out there could be easily put on... [13:59:09] JM: ok [13:59:21] JM: maybe some of these threads should be continued on the mailing list. [13:59:27] JM: there seems to be energy [13:59:36] JM: encourage go to mailing list, put thoughts. [13:59:48] JM: how mature some work is, should be identified. [13:59:55] Please review NHDP & SMF - as they will be WGLC soon. Thank you. [14:00:01] *** Ian Chakeres (ian.chakeres@gmail.com/AdiumC0DA403D) has left the room [14:00:08] JM: if no more then we wrap it up, see people in Stockholm... or around hallways. [14:00:26] (seems to be adjourned) [14:00:43] yes meeting is adjourned great job on jabber!