IETF
homenet@jabber.ietf.org
Wednesday, 7 November 2012< ^ >
Room Configuration

GMT+0
[13:44:25] joel jaeggli joins the room
[13:52:43] danyork joins the room
[13:56:05] bashi_rururu joins the room
[13:56:19] aacosta joins the room
[13:58:19] aacosta is now known as Alejandro Acosta
[13:59:41] xiaohong.deng joins the room
[13:59:45] Alejandro Acosta is now known as Alejandro Acosta (monitor)
[14:00:24] <Alejandro Acosta (monitor)> Almost starting. Agenda bash in the screen
[14:00:44] cabo joins the room
[14:02:17] <Alejandro Acosta (monitor)> Starting. Slide 2: Note Well
[14:02:24] <danyork> Great... we can hear the audio stream now
[14:02:40] Brian Carpenter joins the room
[14:02:47] Andrew Sullivan joins the room
[14:03:14] Markus Stenberg joins the room
[14:03:17] pusateri joins the room
[14:03:22] <joel jaeggli> I can do backup
[14:03:29] <joel jaeggli> meeting has started
[14:03:37] <Alejandro Acosta (monitor)> Slide 3: Administrivia
[14:03:42] Olafur joins the room
[14:03:44] dthaler joins the room
[14:04:01] <Alejandro Acosta (monitor)> Slide 4: Agend Bash
[14:04:01] 83a51488c4e1e550 joins the room
[14:04:04] pselkirk joins the room
[14:04:07] <Alejandro Acosta (monitor)> Thanks Joel!
[14:04:10] Michael Richardson joins the room
[14:04:11] Eliot Lear joins the room
[14:04:16] William Lupton joins the room
[14:04:34] 83a51488c4e1e550 leaves the room
[14:04:56] Alex Petrescu joins the room
[14:05:16] <Alejandro Acosta (monitor)> Lee making a comment
[14:05:21] pebersman joins the room
[14:05:24] Eliot Lear leaves the room
[14:05:30] <Alex Petrescu> he requested larger fonts
[14:05:33] fdupont joins the room
[14:06:00] Jason Weil joins the room
[14:06:08] <Alex Petrescu> Teco Boot asked to present(?) and explain BRDP in written
[14:06:43] <Alejandro Acosta (monitor)> Yes, maybe there will a re-order of the agenda
[14:06:44] <Alex Petrescu> Tim Chown to present
[14:07:01] Lee Howard joins the room
[14:07:07] teco.boot joins the room
[14:07:16] <Lee Howard> I guess the back of the room counts as remote participation
[14:07:20] <Alejandro Acosta (monitor)> Starting: Homenet Architecture
[14:08:06] cgrundemann joins the room
[14:08:12] <Alejandro Acosta (monitor)> Skipped some slides
[14:09:02] <Alejandro Acosta (monitor)> going back Slide 2: This slot…
[14:09:23] <Alejandro Acosta (monitor)> Slide 3: Changes since IETF84
[14:11:33] cgrundemann leaves the room
[14:11:37] suz joins the room
[14:11:56] <Alejandro Acosta (monitor)> Slide 4: Naming and service discovery
[14:12:13] Hassan joins the room
[14:12:32] <Alejandro Acosta (monitor)> Slide 5: Naming/SD text
[14:13:21] <dthaler> non-clarifiying questions/comments deferred to later in agenda?
[14:14:24] <dthaler> (yes)
[14:14:27] <Alex Petrescu> Why don't you try ask?
[14:14:36] <Alejandro Acosta (monitor)> @dthaler, yes
[14:14:42] <dthaler> he already said.
[14:14:44] <Alejandro Acosta (monitor)> Slide 6: Other open questions?
[14:15:31] <Alex Petrescu> Mark Townsley is co-chair is MT
[14:15:37] <Alex Petrescu> Lorenzo Collitti is LC
[14:15:47] <Alex Petrescu> Dave Thaler is DT
[14:15:51] <Alejandro Acosta (monitor)> Lorenzo in the mic
[14:16:00] <Alex Petrescu> LC - what i ssaying out of scope meaning in archi doc
[14:16:04] <Alex Petrescu> answer: not relevanc
[14:16:15] <Alex Petrescu> eMT: out of scope, inside, ouf oside hoùmenet
[14:16:22] <Alex Petrescu> LC: dpeennding on phases of moon?
[14:16:27] <Alex Petrescu> MT: not depend upon it
[14:16:35] <Lee Howard> out of scope = out of charter (?)
[14:16:37] <Alex Petrescu> MT: out and avoid
[14:16:49] <Alex Petrescu> DT: it should say - shoult not ref or doc npt 66
[14:16:57] <Alex Petrescu> DT: it should say you should be a router
[14:17:05] <Alex Petrescu> DT: if not mention npt66 then actual 1
[14:17:06] <Alex Petrescu> claps
[14:17:16] <Alex Petrescu> DT: say other than this which doesnt beg the q
[14:17:21] <Alex Petrescu> TC: ISP...
[14:17:22] <Alex Petrescu> DT: yes
[14:17:33] <Alex Petrescu> LC: sympathise with thatbut have to have alternative
[14:17:37] <Alex Petrescu> LC
[14:17:44] <Alex Petrescu> DT: alternative, but why not doing
[14:17:51] <Alex Petrescu> LC: what operational guidance should we give?
[14:18:19] <Alex Petrescu> LC: should understand what npt66 means to design, major reactions, mine, ought toundertdnstand, if npt66 could be used, then implications to archi
[14:18:20] <danyork> THANK YOU, Alex, for the detailed Jabber scribing
[14:18:30] <Alex Petrescu> LC: design space constraine dby, down the road
[14:18:34] <Alex Petrescu> Dave Oran Cisco is DO
[14:18:48] <Alex Petrescu> DO: homenet routers are, address translation is not considdered
[14:18:57] <Alex Petrescu> DO: not believe any address translation needed. period.
[14:19:00] <Alex Petrescu> Teco Boot is TB
[14:19:14] <Alex Petrescu> TB: as long good dolsution no, then, better translate osme times
[14:19:19] <Alex Petrescu> LC, room: no!
[14:19:25] <Alex Petrescu> TB: I hate that
[14:19:31] <Alex Petrescu> LC: we have better solutions
[14:19:43] <Alex Petrescu> LC: suggest to hunting this q until present solutions
[14:19:54] <Alex Petrescu> LC: say so,ething in this doc, transparent e2e, original IPv6
[14:20:11] <Alex Petrescu> LC: npt66 all of them aexplicitely notes limitations
[14:20:15] <Alex Petrescu> LC: some problems
[14:20:21] <Alex Petrescu> co-chair - good thing to say
[14:20:26] <Alex Petrescu> co-char
[14:20:32] <Alex Petrescu> Mike Richardson is MR
[14:20:53] <Alex Petrescu> MR: early 20s, NAT are routers, confused are they, but the audience... use the words of LC, I like those words
[14:20:58] <Alex Petrescu> co-chair - early 20s?
[14:21:17] <Alex Petrescu> co-chair - qs about DNS? reverse DNS branch you said?
[14:21:19] <Alex Petrescu> presenter - no
[14:21:25] <Alex Petrescu> co-chair - should there be smth?
[14:21:34] <Alex Petrescu> presenter - not strong feeling for or against
[14:21:38] <danyork> Sadly, I do have to agree with MR that many people *assume* that a "router" just does NAT. :-( (no need to relay, just a comment)
[14:21:45] each joins the room
[14:21:46] <Alex Petrescu> Alain Durand is AD
[14:21:52] <Alex Petrescu> AD: shoud provide some guidance
[14:21:59] <Alex Petrescu> cochair - provide text
[14:22:04] <Alex Petrescu> AD - two lines - its optional
[14:22:04] <each> I missed alain's question, can someone repeat it?
[14:22:11] <Alejandro Acosta (monitor)> Slide 7: Are we done?
[14:22:36] <Lee Howard> Alain said, "Should we make a recommendation on reverse DNS." Presumably about delegation or updating. He wants us to say, "It's optional."
[14:23:01] <each> thanks
[14:23:14] <Alex Petrescu> Erik Kline is EK to present
[14:23:20] <Alejandro Acosta (monitor)> Starting: perimeter-ident-01
[14:23:30] <Alejandro Acosta (monitor)> Slide 2: Scope and Terminology
[14:23:52] <Alejandro Acosta (monitor)> Slide 3: -00
[14:24:02] <danyork> This is the presentation, correct? http://www.ietf.org/proceedings/85/slides/slides-85-homenet-2.pdf
[14:24:17] <Alejandro Acosta (monitor)> @danyork, right
[14:24:25] <danyork> thanks
[14:24:29] <Alejandro Acosta (monitor)> Slide 4: -01
[14:25:26] <Alex Petrescu> (each homenet slideset is at https://datatracker.ietf.org/meeting/85/materials.html#wg-homenet)
[14:25:46] <Alejandro Acosta (monitor)> Slide 5: Learning algorithm
[14:26:32] cgriffiths joins the room
[14:27:17] cgrundemann joins the room
[14:27:43] kk.chittimaneni joins the room
[14:28:29] <Alejandro Acosta (monitor)> Slide 6: Filter policies: a use case
[14:28:46] Ole Troan joins the room
[14:28:51] <cabo> Can we see the whole slide, not with the last line covered by the UI elements? http://wiki.tools.ietf.org/group/wgchairs/wiki/MeetingAV
[14:28:52] <Alejandro Acosta (monitor)> Slide 7: Filter policy: interior anti-spoofing
[14:29:35] <Alex Petrescu> cabo - whom should we ask?
[14:29:47] <Alejandro Acosta (monitor)> @cabo, ok, I will try to say it later
[14:29:56] <Alejandro Acosta (monitor)> Slide 8: Filter policy: stateful exterior
[14:30:05] <Alejandro Acosta (monitor)> Slide 9: Some open questions
[14:30:12] <cabo> Don't say it, just take the WG chair aside and explain how to do it :-)
[14:30:14] <Alex Petrescu> DT and MR approaching
[14:30:52] <Alejandro Acosta (monitor)> DT in the mic
[14:31:10] <Alex Petrescu> DT: external gws...
[14:31:18] <Alex Petrescu> DT: same defs legacy device, thats actually a border
[14:31:27] <Alex Petrescu> DT: not what we want, but effectively reported
[14:31:36] <Alejandro Acosta (monitor)> Chairs trying to relocate the UI element
[14:31:38] <Alex Petrescu> DT: what wantoto change to lake it a homenet router?
[14:31:45] <Alex Petrescu> DT: that algo does right thing
[14:31:52] <Alex Petrescu> DT: guidance about fixed fction typed iface
[14:32:07] <Alex Petrescu> EK: assume fixed iface situation - either product defined, or legacy device
[14:32:33] <Alex Petrescu> DT: I say is that people doing this not necesszarily know product defined, need guidance undersntand whether or not it should be producte defined
[14:32:45] <Alex Petrescu> EK/ example tech, I move it to example section, far from exhausted.
[14:32:51] <Alex Petrescu> exhaustive
[14:33:03] <Alex Petrescu> MR: think that I agree with DT strongly
[14:33:09] <Alex Petrescu> MR: run on a port called LAN
[14:33:22] <Alex Petrescu> MR: iusers dont plug things as we expect them to
[14:33:34] <Alex Petrescu> MR: second layer, but probably would plug on upstream
[14:33:43] <Alex Petrescu> MR: internal not external, despite labelling
[14:33:51] <Alex Petrescu> MR: fixed function ports would not get this?
[14:34:18] <Alex Petrescu> EK: in my mind, there is a distinciton between ports LAN and no discovery algo on it, and aport labelled WAN (Internet) that really actually means upstream
[14:34:30] <Alex Petrescu> EK: labelling is different from... its a ficed category iface
[14:34:54] <Alex Petrescu> MR: I want to confirm; the pppoe uns on an 'e' ? is separate than iface...
[14:35:12] <Alex Petrescu> MR: so I could have a LAN port plugged to a router, as a n internal link, also plugged on modem, pppoe
[14:35:18] <Alex Petrescu> EK: collection of virtual ifaces
[14:35:23] <Alex Petrescu> MR: thats going to be apopular
[14:35:31] <Alex Petrescu> EK: cant apply stateful filters to interior traffic
[14:35:48] <Alex Petrescu> EK: trusting things form that upstream, unless you can apply filtering, by sort of next hop/previous hop
[14:36:06] <Alex Petrescu> DT: open qs on screen, my personal oppinions 1 -yes, 2- yes, 3-punt, 4-dont understand q
[14:36:14] Markus Stenberg leaves the room
[14:36:16] <Michael Richardson> +1 on dave's answer.
[14:36:23] <Alex Petrescu> EK: considering the ... iface, pppoe session, no exhaust outside?
[14:36:24] <Alex Petrescu> DT: ?
[14:36:26] <Alex Petrescu> EK: of course
[14:36:30] <Alex Petrescu> Ole Troan is OE
[14:36:37] <Alex Petrescu> OE/: 4 is a subset of 3
[14:36:43] <Alex Petrescu> DT...
[14:37:06] <Alex Petrescu> EK: try to discover routers, PD, ... but also discover nodes? should it rely on link upstream? back upstream?
[14:37:08] <Alex Petrescu> OT Cisco
[14:37:21] <Alex Petrescu> OT - plus one to Daves point, punt on 3 and yes on ...
[14:37:28] <Alex Petrescu> Markus Stenberg is MS
[14:37:48] <Alex Petrescu> MS: RA from CP, and from ISP? skip that and planning on punting, I dont think 4 is applicable
[14:38:08] Markus Stenberg joins the room
[14:38:15] <Alex Petrescu> DT: punt for 3, because, behaviour number of external side... point of id probleM? (this doc) ... interesting q ok, but this doc should focuson 1 and 2
[14:38:26] <Alex Petrescu> EK: punt == dont say anything ? or
[14:38:28] <Alex Petrescu> DT: ...
[14:38:31] <Michael Richardson> re: #4, it should not provide RAs, if it forms a routing adjacency. it should not emit RAs for awhile, while looking for that routing adjancy
[14:38:34] <Alex Petrescu> DT: what would you do with that.
[14:38:43] <Alex Petrescu> DT: how would you use id - other work.
[14:38:54] <Alex Petrescu> DT: scope this to how do I decide with internal/external
[14:39:05] <Alex Petrescu> OT: mixed mode iface, internal and external properties.
[14:39:12] <Alex Petrescu> OT: per neighbor popretiees?
[14:39:14] <Alex Petrescu> DT: yes
[14:39:18] <Alex Petrescu> DT: I agree
[14:39:25] <Alex Petrescu> Andre MLcGregor is AMG
[14:39:37] <Alex Petrescu> AMG: v4, if the extenral bit is on, dont do this, creaete chaos
[14:39:45] <Alex Petrescu> Andre McGregor
[14:39:45] each leaves the room: Replaced by new connection
[14:39:47] each joins the room
[14:39:50] <Alex Petrescu> Teco Boot i sTB to present
[14:40:04] <Alex Petrescu> (Andrew McGregor)
[14:40:24] <dthaler> punt = don't support mixed mode interfaces in this doc
[14:40:25] <Alejandro Acosta (monitor)> Starting: draft-boot-homenet-brdp
[14:40:42] <Alejandro Acosta (monitor)> Slide 2: Aim, origin and status of BRDP
[14:40:59] zcao joins the room
[14:41:27] <Alejandro Acosta (monitor)> Slide 3: Let’s bind exisFng stuff together
[14:41:40] <Alejandro Acosta (monitor)> exisFng=existing
[14:42:12] <Alejandro Acosta (monitor)> Slide 4: With BRDP, Border Router address and ISP
delegated prefix is disseminated in the edge
network
[14:43:22] <Alejandro Acosta (monitor)> Slide 5: Border Router acts as DHCP server when D-flag is set
[14:43:26] <Alex Petrescu> (problems with mic placing on body)
[14:43:36] <Alex Petrescu> co-chair - if people cant hear please le tus know
[14:44:01] <Alejandro Acosta (monitor)> Slide 6: Other fields: for BRIO forwarding and hints for costs to Internet DFZ
[14:45:24] <Alejandro Acosta (monitor)> Slide 7: BRDP based routing
[14:48:00] <Alejandro Acosta (monitor)> Slide 8: Next steps
[14:48:17] <Alejandro Acosta (monitor)> Slide 9: Basics
[14:48:27] <Alejandro Acosta (monitor)> (they will go back to Next steps later)
[14:48:32] <Alex Petrescu> TB: we still have some time I think
[14:48:43] <Alejandro Acosta (monitor)> Slide 10: 1) Customer Edge Router configures subnets on its interfaces
[14:48:49] xiaohong.deng leaves the room
[14:48:58] <Alejandro Acosta (monitor)> Slide 11: 2) Customer Edge Router starts sending BRIO
[14:49:24] <Alejandro Acosta (monitor)> Slide 12: 3) Customer Interior Router configures subnets on its interfaces
[14:49:37] each leaves the room
[14:49:41] <Alejandro Acosta (monitor)> Slide 13: 4) Customer Interior Router starts sending BRIO
[14:49:53] darshakthakore joins the room
[14:49:59] <Alejandro Acosta (monitor)> Slide 14: 5) Interior nodes select IP addresses as source address
[14:51:29] <Alejandro Acosta (monitor)> Slide 15: 6a) Customer Routers run a routig protocol
[14:51:41] bashi_rururu leaves the room
[14:51:47] <Alejandro Acosta (monitor)> Slide 16: 6b) Forwarding for destinations inside the Homenet
[14:51:51] <Alex Petrescu> TB talks to co-chair about slides
[14:52:06] <Alejandro Acosta (monitor)> Slide 17: 6c) Forwarding for destinations outside the Homenet
[14:52:19] Lee Howard leaves the room
[14:52:23] Lee Howard joins the room
[14:52:44] <Alejandro Acosta (monitor)> Slide 18: Multi-homing:
[14:52:57] <Alejandro Acosta (monitor)> Slide 19: 1) BRDP Basics are applied
[14:53:19] <danyork> Is someone available to relay a question to the mic?
[14:53:22] <Alejandro Acosta (monitor)> Slide 20: 2) BRDP based address configuration
[14:53:29] <Lee Howard> I can relay to mic
[14:53:34] <Alejandro Acosta (monitor)> Slide 21: 2) BRDP based address selecHon
[14:54:00] <Alejandro Acosta (monitor)> Slide 22: a) BRDP based rouHng, redirecting packets
[14:54:07] <danyork> MIC: has any security analysis been done on this approach? Is that planned? I see the Security Considerations section of the draft is "TBD" but on a brief read of the draft I worry an attacker could misuse it.
[14:54:09] <Alejandro Acosta (monitor)> Sorry.., 22 not yet
[14:54:12] <danyork> Thanks, Lee
[14:54:27] <Alejandro Acosta (monitor)> Slide 22: a) BRDP based routing, redirecting packets
[14:54:50] <Alejandro Acosta (monitor)> Slide 23: 3b) BRDP based routing, multipath
[14:54:58] <Alex Petrescu> Lee Howard is LH approaching mic
[14:55:12] <Alex Petrescu> Jim Gettys is JG approaching
[14:55:25] <Alex Petrescu> DT queues
[14:55:36] bashi_rururu joins the room
[14:55:45] <Alex Petrescu> co-chair: short we ar eon time
[14:55:54] <Alex Petrescu> LH: relaying from jabber - security ?
[14:56:06] <Alex Petrescu> JG speaking
[14:56:17] <Alex Petrescu> JG: config keep away from rt protocol, which evolve separately
[14:56:27] <Alex Petrescu> JG: can we use to distribute DNS and MNDPP info?
[14:56:33] <Alex Petrescu> TB: DNS not on border maybe
[14:56:40] <Alex Petrescu> JG: everybody downstream needs to know to find
[14:56:44] gridmerge joins the room
[14:56:44] <Alex Petrescu> TB: host needs also same
[14:56:49] <danyork> Thanks, Lee, although I didn't hear any answer.
[14:57:00] <Alex Petrescu> TB: DNS or name resolving flag, CPE devices as dns server, but not in my proposal
[14:57:05] <Alex Petrescu> G: is ther an implementation?
[14:57:08] <Alex Petrescu> TB: in old days opnet
[14:57:18] <Alex Petrescu> TB: not full blown not yet not today
[14:57:24] <Alex Petrescu> JG: different from ahcp?
[14:57:24] wmtownsley joins the room
[14:57:31] <Alex Petrescu> TB: v6 only on ND, lots of other details but
[14:57:37] <Alex Petrescu> JG: covering same terriotriy?
[14:57:44] <Alejandro Acosta (monitor)> (FYI, slide 8 on the screen: Next steps)
[14:57:47] <Alex Petrescu> TB: relation to routing, basic mechs for src address routing
[14:57:52] <Alex Petrescu> TB: wonder wbout ahcp
[14:58:00] <Alex Petrescu> co-chair: no time to detail, go
[14:58:04] <Alex Petrescu> DT: interesting thing shere
[14:58:11] <Alex Petrescu> DT: comment on step 0
[14:58:13] pebersman leaves the room
[14:58:14] <Alex Petrescu> DT: forwarding
[14:58:19] <Alex Petrescu> DT: step 0 is no defaul troute
[14:58:32] <Alex Petrescu> DT: works for if using for generate a packet, when generate address sel?
[14:58:35] <Alex Petrescu> DT: forwarding
[14:58:47] <Alex Petrescu> DT: achieve same, if augmenet def routes with src info
[14:58:53] <Alex Petrescu> DT: think abou tthat
[14:59:02] <Alex Petrescu> TB: problem is in detail, have to fixt ianywya
[14:59:18] <Alex Petrescu> DT: homenet router outgoing connection, important for router have def route to do addr sel
[14:59:25] <Alex Petrescu> DT: ...
[14:59:40] <Alejandro Acosta (monitor)> Starting: Bootstrapping Trust in a Homenet
[15:00:01] <Alex Petrescu> Michale Behringer is MB to present
[15:00:04] <dthaler> e.g. homenet router might need to establish TCP connection over the Internet to pull down software./fireware update or somethign
[15:00:10] <Alex Petrescu> Michael Behringer
[15:00:21] gridmerge leaves the room
[15:00:23] <Alex Petrescu> fixing cables and radios
[15:00:28] <Alex Petrescu> room - something sucks
[15:00:39] Robert Cragie joins the room
[15:00:39] <Alejandro Acosta (monitor)> Link for the presentation (just uploaded): http://www.ietf.org/proceedings/85/slides/slides-85-homenet-8.pdf
[15:01:03] <Alejandro Acosta (monitor)> Slide 2: Problem Statement
[15:01:29] <Alejandro Acosta (monitor)> Slide 3: Approach:
[15:01:59] darshakthakore leaves the room
[15:03:41] <Alejandro Acosta (monitor)> Slide 4: Approach: (Neighbor Discovery)
[15:03:54] darshakthakore joins the room
[15:04:05] <Alejandro Acosta (monitor)> Slide 5: Approach: 3) Domain Join
[15:05:02] <Alejandro Acosta (monitor)> Slide 6: Result
[15:05:43] <Alex Petrescu> MB talks to EK - they agree about the level of classifi cation/grain
[15:05:52] <Alejandro Acosta (monitor)> Slide 7: More Thoughts…
[15:05:58] <Alex Petrescu> MR queuing
[15:06:09] <Alex Petrescu> MS on mic
[15:06:24] <Alex Petrescu> MS: default mode without ... is same as using.. shared secret done
[15:06:31] <Alex Petrescu> MB agree
[15:06:45] <Alex Petrescu> MR: not sure its truee, did you consider doing how this may benefit from CGA addresses?
[15:06:52] <Alex Petrescu> Basal Thubert is PT
[15:06:54] <Alex Petrescu> PT-SEND
[15:06:59] <Alex Petrescu> PAscal Thubert
[15:07:11] <Alex Petrescu> MB - not entirely clear, I refer to Max P...
[15:07:19] <Alex Petrescu> Max is MP
[15:07:25] <Alex Petrescu> MP: not enough space to secure enogh
[15:07:32] <Alex Petrescu> MR: secure enough for what?
[15:07:37] <Alex Petrescu> MR: what is threat model?
[15:07:46] <Alex Petrescu> MP: default operatio, if we do, do right, otherwise ,o
[15:07:51] <Alex Petrescu> MP: need decision on that
[15:08:01] <Alex Petrescu> MR: pkx format use you? some idea of lifetime?
[15:08:12] dave.taht joins the room
[15:08:14] <Alex Petrescu> MR: kind of priority? my son takes router to his friends house, bring home
[15:08:16] <Alex Petrescu> PT: shoul dblow
[15:08:23] <Alex Petrescu> MB: renew operation
[15:08:24] <dave.taht> yea. jabber.
[15:08:38] <Alex Petrescu> MR: make short, when going to friend, go hme back; no have to factoryu reset each time?
[15:08:43] <Alex Petrescu> MB: currently do factory reset
[15:08:52] <Alex Petrescu> MB: interesting only if interest in security
[15:09:11] <Alex Petrescu> MR: bootstrap relationship? when two nodes next to each other, that , could be used when the devices are not in thomenet
[15:09:21] <Alex Petrescu> MR: intersting, this is, and secure enough for what?
[15:09:42] <Alex Petrescu> MR: even if automatic, smartphones one in others' house, ok print to my printer
[15:09:51] <Alex Petrescu> MB: design options these are to look into
[15:10:09] <Alex Petrescu> MB: if you enroll into a domain, it sticks, so you have to factory reset. Does this lake sense.
[15:10:28] <Alex Petrescu> EK: mos tinteresting to me is wha tyou lamekae optional
[15:10:51] <Alex Petrescu> EK: that is how to make layer 7 info, some appli on phone, this router, this mac address, this product name, shoud I form... this is cool.
[15:10:56] <Alex Petrescu> MB: this is just part of it
[15:11:04] <Alex Petrescu> DT: slide 5 said optional
[15:11:17] <Alejandro Acosta (monitor)> Almost Starting: OSPFv3-Based Home Networking - Report (*** not sure the URL of the presentation, if anybody knows please point it out ***)
[15:11:20] <Alex Petrescu> EK: admin domaoin knowledhe into sometinghin... cool seems to me.
[15:11:24] <Alejandro Acosta (monitor)> (*** not sure the URL of the presentation, if anybody knows please point it out ***)
[15:11:25] <Alex Petrescu> Jari Arkko is JA to present
[15:11:44] <Alex Petrescu> co-chair about slides receive at a particular time (late? early?)
[15:11:45] <Lee Howard> Jari just emailed slides to the list
[15:12:09] <Alejandro Acosta (monitor)> Yeah... it's at: http://arkko.com/ietf/homenet/ietf85_homenet_ospf.pdf
[15:12:13] <Alex Petrescu> This is Markus stands up
[15:12:35] <Alejandro Acosta (monitor)> Slide 2: Outline
[15:12:50] <dave.taht> have they tried ospfv3 over mixed wired and wireless links yet?
[15:12:59] <Alejandro Acosta (monitor)> Slide 3: Draft Updates
[15:13:07] <Alex Petrescu> (dont do this kind sof routers and ethernet at airport - security)
[15:13:16] <Robert Cragie> Re the domain certificate - if it is clearly bound to the address in the trust domain then this could be useful for the portability case as it could retain its address in a foreign trust domain and still potentially use the domain certificate to communicate securely through a foreign domain back to its "home" trust domain
[15:14:29] <Alex Petrescu> TB queueing
[15:14:59] <Alex Petrescu> TB: use all hosts, all hosts are receivers for ospf updates? they join?
[15:15:06] <Alex Petrescu> JA: this is per homenet routers, not per host
[15:15:15] <Alex Petrescu> TB: how can host resolve dns? local router?
[15:15:53] <Alex Petrescu> JA: maybe a picture, a host needs some info externally given to id; today from dhcp or from RA; idea here is implementations 'susten' and then they advertsise in dhcp...
[15:16:01] <Alex Petrescu> JA: hosts attach to routers and they get to know?
[15:16:11] <Alex Petrescu> TB: this is for distrib of info of al dhcp servers in the network?
[15:16:24] <Alex Petrescu> OT: perhaps adding dns infor is feature creep in the prefix assignment
[15:16:26] <Alex Petrescu> JA: agree
[15:16:37] <Alex Petrescu> OT: need more generic mech, theres quite a bot pof it
[15:16:45] <Alex Petrescu> JA: goo d point, some archi things I mention
[15:16:51] <Alex Petrescu> JA: I report what in my ipmpl
[15:17:15] <Alex Petrescu> JA: locally retrieve info from srces, and then locally distribute info, actulally go in my opspf impl, in my ipmlp
[15:17:31] <Alex Petrescu> JA: could sitrd some pieces of info in ospf, not done that because ... wise.
[15:17:57] <Alex Petrescu> MS: not sure not saying not distri to ... all ospf;, still a matter of debvate, we have arguments, archi different, homenet over bus, distr, ddns
[15:18:04] <Alex Petrescu> MS: different than ospf exists, discuss later
[15:18:12] <Alex Petrescu> JA: agree, not have to distri any info
[15:18:18] <Alex Petrescu> LC: also introduces limitations
[15:18:41] <Alex Petrescu> LC: think good idea to decopulep prefix assignmenet from dns, but also need one source of flooding info in homenet, and that is
[15:18:48] <Alex Petrescu> LC: it s a ino distr mech
[15:18:51] mbehring@cisco.com joins the room
[15:19:03] <Alex Petrescu> LC: suitable extensible routing protocol, if we have this, it has to run anayways
[15:19:11] <Alex Petrescu> LC: if routing not work then nothing is working
[15:19:23] <Alex Petrescu> LC: important to realiwe this only flooding whith this oother proposal
[15:19:39] <Alex Petrescu> LC: its always either the routers, or the hosts themselves they auth they way they like
[15:19:46] <Alex Petrescu> LC: itùs an auth througuh
[15:20:40] <Alex Petrescu> AMG: involvde I was in buidling uses a n modified ospf to distr info like dns and loads of other stuff
[15:20:41] <Alex Petrescu> AMG: works well I recommend
[15:20:44] <Alex Petrescu> David Lamparter is DL
[15:20:55] <Alex Petrescu> DL: on a router, quite a few issues
[15:21:02] <Alex Petrescu> AMG said that this routing works
[15:21:12] <Alex Petrescu> Barbara Stark is BS and at ATT
[15:21:20] <Alex Petrescu> room tells her where to stay with respect to mic
[15:21:29] <Alex Petrescu> BS: interesting this is
[15:21:40] <Alex Petrescu> BS: I was frustrated with DHCP pd not being built in CE routers
[15:21:43] <danyork> Audio is good for Barbara Stark now
[15:21:48] <Alex Petrescu> BS: they leep saying
[15:22:01] <Alex Petrescu> BS: I dont implement, but if easier, then a really good thing it would be
[15:22:14] <Alex Petrescu> AMG: I implemented exactly that it can be done, it works
[15:22:27] <Alex Petrescu> AMG: but you need another info distrib bus, and dhcp doesn work
[15:22:37] <Alex Petrescu> AMG: despite the recursive dhcp pd approach
[15:22:49] <Alex Petrescu> LC: two implementations - it means they can be impl
[15:22:58] <Alex Petrescu> LC: many impl DHCP PD exists as well, small
[15:23:07] <Alex Petrescu> LC: orthogonal to complexity of iplp
[15:23:13] <Alex Petrescu> LC: there are impl of dhcp pd
[15:23:25] <Alex Petrescu> LC: what is hard DHCP is you dont know how to hand tha tout
[15:23:41] <Alex Petrescu> LC: /56 to /62, if you chain it wouldnt work, its hard to decicde what to do
[15:23:52] <Alex Petrescu> LC: whereas this is a lot clearer and works in many other situqtions
[15:23:56] <Alex Petrescu> Ralph Droms is RD
[15:24:03] <Alex Petrescu> RD: strawman, not bother
[15:24:12] Markus Stenberg leaves the room
[15:24:14] <Alex Petrescu> RD: not argue, your but your argumen,tb is strong
[15:24:21] <Alex Petrescu> LC: which strong havent yet peresented.
[15:24:27] <Alex Petrescu> Hans Liu is HL
[15:24:47] <Alex Petrescu> HL: my promiss is that yeah, I ack LC's statement the q i show should i assign the prefix?
[15:24:55] <Alex Petrescu> HL: PD the rest of link?
[15:24:57] <Alex Petrescu> cochair cuts
[15:24:57] Markus Stenberg joins the room
[15:25:09] <Alex Petrescu> AMG what to hand out from recursive od is the tricky
[15:25:19] <Alex Petrescu> AMG we punted, it is a data har dproblem (iundeed)
[15:25:27] <Alex Petrescu> AMG this is a nice solution not to go in there
[15:25:31] <Alex Petrescu> JA: allocation, decent
[15:25:44] <Alejandro Acosta (monitor)> Resuming presentation Slide 4: Implementation Report
[15:25:49] <Alex Petrescu> LC: quick, if you only have a /60 with dhcp pd you run out of space really quick
[15:26:00] dave.taht leaves the room
[15:26:36] <Alex Petrescu> MS: I am happy to say that core part of is now git, I put jabber url, open source, openwrt stuff
[15:26:42] <Alex Petrescu> MS: code is there in github
[15:26:56] <Alex Petrescu> MS: hard to say, there is a README, somebody advetnure;,live dangerously
[15:26:59] <Alejandro Acosta (monitor)> Slide 5: Interop Report
[15:27:04] <Alex Petrescu> JA: great news
[15:27:15] <Markus Stenberg> here's the Cisco implementation of prefix assignment: https://github.com/fingon/hnet-core
[15:27:27] <Markus Stenberg> (some other stuff still pending, such as openwrt packaging feed etc)
[15:27:29] equinox joins the room
[15:28:01] <Alejandro Acosta (monitor)> Slide 6: Interop Observations
[15:28:06] cgriffiths leaves the room
[15:28:14] <Alejandro Acosta (monitor)> Slide 6 skipped
[15:28:21] <Alejandro Acosta (monitor)> Slide 7: Interop Observations 2
[15:29:39] <Alejandro Acosta (monitor)> @Markus..., thanks for the link
[15:29:50] <Alejandro Acosta (monitor)> Going back to Slide 6
[15:30:01] <Alejandro Acosta (monitor)> Interop Observations
[15:30:35] <Alejandro Acosta (monitor)> @Markus, thanks for the link
[15:30:51] <Alejandro Acosta (monitor)> Slide 8: Things to Consider – Draft Details
[15:31:05] <Alejandro Acosta (monitor)> Slide 9: Things to Consider – Brittle Timer Defaults
[15:31:16] <Alex Petrescu> TB queueing
[15:32:13] <Alex Petrescu> TB: two issues
[15:32:16] intvelt joins the room
[15:32:22] <Alex Petrescu> TB: sped of timers, you want to have connectivity
[15:32:32] <Alex Petrescu> TB: each router decides on timers others nodes dont aspect
[15:32:41] <Alex Petrescu> TB: this is a restriction on ospf must use same timers
[15:32:46] <Alex Petrescu> TB: we should change that in ospf
[15:32:52] <Alex Petrescu> JA: general ospf issue
[15:33:00] <Alex Petrescu> JA: our draft not try to address autoconf in ospf
[15:33:05] <Alex Petrescu> JA: theres more problems with that
[15:33:08] <equinox> is this the last slide?
[15:33:13] <Alex Petrescu> JA: needs tob ehandled somehow... and that it
[15:33:22] <Alejandro Acosta (monitor)> @equinox: yes
[15:33:29] <Alejandro Acosta (monitor)> this is the last slide
[15:33:32] <Alex Petrescu> oc-chair: we said early, open source is important because of consuers we target, very cool
[15:33:42] <Alex Petrescu> co-chair: comcast agreed to give space on bits and bytes table
[15:33:54] <Alex Petrescu> co-chair:...
[15:33:59] <Alex Petrescu> Mark Townsley is MT
[15:34:05] <Alex Petrescu> Tim Chown is TC
[15:34:12] <Alex Petrescu> TC: we have 2 froim Fin;, and 3rd from?
[15:34:15] <Alex Petrescu> room: France
[15:34:22] <Alex Petrescu> Acee Lindem is AL
[15:34:32] <Alex Petrescu> AL: for timers, for commercial routers, thats my place
[15:34:34] <Alex Petrescu> room laughs
[15:34:40] <Alex Petrescu> room aout mic position
[15:34:53] <Alex Petrescu> AL: not seen a variance in people not taking recommendation
[15:35:01] <Alex Petrescu> AL: they cant bury too much, adjacency
[15:35:12] <Alex Petrescu> AL: dead variables smaller intervals smaller than hello intervals
[15:35:25] <Alex Petrescu> JA: .. not let you do yours, I do mine, I see you do something I adjust mine
[15:35:32] <Alex Petrescu> AL: but whos definitive guidant?
[15:35:34] <Alex Petrescu> room in hello
[15:35:37] <Alex Petrescu> JA: my ipml
[15:35:44] <Alex Petrescu> David Laporter is DL
[15:36:02] <Alex Petrescu> DL: concerns abour, need fine tuning, slowly deprecating prefix, vs quickly deprectaed prefix
[15:36:09] <Alex Petrescu> DL: not sufficiently ripe in the draft
[15:36:12] <Alex Petrescu> JA: agree
[15:36:34] <Alejandro Acosta (monitor)> Starting: draft-haddad-homenetmultihomed
[15:36:45] <Alex Petrescu> Damien Saucez is presenting is DS
[15:36:55] cgriffiths joins the room
[15:37:09] <Alejandro Acosta (monitor)> Slide 2: Note well
[15:37:39] <Alejandro Acosta (monitor)> Slide 3: I-D.arch-06
[15:37:45] cgriffiths leaves the room
[15:38:05] ctg1701 joins the room
[15:39:02] <Alejandro Acosta (monitor)> Slide 4: Motivation
[15:40:09] ctg170187633 joins the room
[15:40:13] <Alejandro Acosta (monitor)> Slide 5: Requirements
[15:40:34] Ole Troan leaves the room
[15:40:49] <Alejandro Acosta (monitor)> Slide 6: Proposition
[15:41:08] ctg1701 leaves the room
[15:41:24] dave.taht joins the room
[15:42:10] cheshire joins the room
[15:42:12] <Alex Petrescu> MR queueing
[15:42:22] <Alex Petrescu> MR: what does this box goes?
[15:42:25] <Alejandro Acosta (monitor)> Slide 7: MSP in a nutshell (it's actually slide 6 in the presentation)
[15:42:28] <Alex Petrescu> MR: the middlebox thanks
[15:42:36] <Alex Petrescu> room lauhgs
[15:42:39] <Alex Petrescu> MR: cant get in all gws
[15:42:49] <Alex Petrescu> DS: ...
[15:42:53] <Alex Petrescu> MR: sorrows ring
[15:42:56] <Alex Petrescu> room laughs
[15:43:12] <Alex Petrescu> MR: such box connected to isp1 and 2? no need of gw1 and 2
[15:43:15] <Alex Petrescu> DS: ye
[15:43:22] <danyork> Sauron's Ring, i.e. One Ring To Rule Them All :-)
[15:43:26] <Alex Petrescu> DS: you still have to decide whether left or right
[15:43:45] <Alex Petrescu> MR: posit the existenc of that box then no problem?
[15:43:49] <Alex Petrescu> DT: only one prefix
[15:43:52] <wmtownsley> (as line forms, chair wonders whether he should have provided a flame-retardant suit for presenter)
[15:43:56] <Alex Petrescu> DS: if left, then ISP left
[15:44:05] <Alex Petrescu> MR: protocol will help laptop pick right address?
[15:44:17] <Alex Petrescu> DS: the address belonging to homen etzork ar enot sending to isp
[15:44:27] <Alex Petrescu> MR: going to encapsulate to this has provider?
[15:44:30] <Alex Petrescu> queue laughs
[15:44:37] <danyork> wmtownsley - :-)
[15:44:37] <Alex Petrescu> DT: mor elike mobile ip
[15:44:39] <Alex Petrescu> chair cuts
[15:44:56] <Alex Petrescu> OT: baiscally solvng a mhoming problem by single homing self to someone else
[15:44:58] <Alex Petrescu> DS: yes.
[15:45:06] dave.taht leaves the room
[15:45:08] <Alex Petrescu> DS;: failure of middleblink? or of route?
[15:45:18] <Alex Petrescu> AMG: paint twice your traffic, doesnt fly where I come from
[15:45:21] <Alex Petrescu> DS: not understand
[15:45:32] <Alex Petrescu> AMG: you pay for traffic crossing either isp, then dns providing...
[15:45:35] <Alex Petrescu> DT/ its a feature
[15:45:49] <dthaler> that was DO (Dave Oran) saying it's a feature not DT
[15:46:05] <Alex Petrescu> LH: it does implement what you tae a problem, extract to a higher levelr of abstraction, I propose an archi principle, invoking a od box leads to just worse
[15:46:19] <Alex Petrescu> LH: create new problems by creating single box
[15:46:22] <cgrundemann> "Invoking a god box leads to religious wars."
[15:46:27] <Alex Petrescu> SD: create new problems
[15:46:31] <Alex Petrescu> LH:...
[15:46:35] <Alex Petrescu> TC: down a rathole
[15:46:52] <Alex Petrescu> TC: we saw interesting things from JA, and mumtihimoing IPv- without NAT boxes, also interesting
[15:47:03] <Alex Petrescu> TC: careful about the wholes we go into
[15:47:04] <Michael Richardson> hosts will learn. It won't be immediately, but they will learn. punting from gw-1/gw-2 when the hosts picks the wrong place should work in the interum.
[15:47:08] <Alex Petrescu> DS: thanks
[15:47:21] <Alejandro Acosta (monitor)> Slide 8: A day in the life.......
[15:47:23] <Lee Howard> LH: Propose as an architectural principle: "Invoking a god box leads to religious wars."
[15:47:39] <Alex Petrescu> Bob Hinden is BH
[15:47:41] <danyork> Nice comment, Lee.
[15:47:42] <Alex Petrescu> BH: ginat problem
[15:47:53] <equinox> awesome comment :D
[15:48:09] <Alex Petrescu> BH: reason of mhoming is more or less multiple choices, but if single box then... loose other adv. I dont want single point of failure
[15:48:12] <Alex Petrescu> cochair cuts
[15:48:16] <Alex Petrescu> BS really?
[15:48:19] <Alex Petrescu> room go ahead
[15:48:31] <Alex Petrescu> its not BS itùs Cathy Aronson
[15:48:40] <Alex Petrescu> CA: dont really understand point of...
[15:48:57] <Alex Petrescu> DS: have to choose left right, if private, ugly translation, if you have... after you have DI,
[15:48:57] kk.chittimaneni leaves the room
[15:49:02] <Alex Petrescu> cochair or more prefix in the home
[15:49:06] <Alex Petrescu> CA: yeah...
[15:49:06] <Alejandro Acosta (monitor)> Resuming Slide 9: Why not Shim6 (MPTCP) directly on the hosts?
[15:49:14] <Alejandro Acosta (monitor)> Skipped slide 9
[15:49:36] <Alex Petrescu> LC presenting
[15:49:55] <Alejandro Acosta (monitor)> Starting: Source + Destionation-Based Routing
[15:50:01] <Alejandro Acosta (monitor)> Slide 1: The Scenario
[15:51:02] Hassan Zaheer joins the room
[15:51:25] <Alejandro Acosta (monitor)> Slide 2: Goals
[15:51:25] <Lee Howard> slide 2 Goals
[15:51:34] <Lee Howard> sorry, I had significant lag
[15:51:42] pusateri leaves the room
[15:51:52] <Alex Petrescu> (sometimes jabber tells me "bouncing")
[15:51:55] <Alejandro Acosta (monitor)> Slide 3: Some questions...
[15:52:33] Ole Troan joins the room
[15:52:37] <dthaler> what's the link to this presentation?
[15:52:39] bashi_rururu leaves the room
[15:52:49] <danyork> What slides are these?
[15:52:56] <danyork> Heh... what dthaler said?
[15:52:56] <Alejandro Acosta (monitor)> Slide 4: Solution approaches
[15:52:59] <wmtownsley> Jari posted the slides to the homenet list
[15:53:08] <danyork> Ahhh.
[15:53:10] <wmtownsley> It will be on the site soon.
[15:53:30] <wmtownsley> Sorry, we operate on the bleeding edge here. All information is as fresh as it can be :-)
[15:53:36] <Alejandro Acosta (monitor)> Slide 5: Source + destional-based routing
[15:54:13] <Alejandro Acosta (monitor)> URL of this presentation: http://arkko.com/ietf/homenet/ietf85_homenet_source_routing.pdf
[15:54:38] Hassan leaves the room
[15:55:03] <danyork> Found it: http://arkko.com/ietf/homenet/ietf85_homenet_source_routing.pdf
[15:55:54] <Alejandro Acosta (monitor)> Slide 6: Implementing source + destional - based routing
[15:56:22] dave.taht joins the room
[15:56:30] <Alejandro Acosta (monitor)> Slide 7: Open questions
[15:56:39] <danyork> (thank you for posting the link, Alejandro - I didn't see your link before posting it again.)
[15:57:01] Lee Howard75432 joins the room
[15:57:14] <Alex Petrescu> AL queueing
[15:57:34] <Alex Petrescu> AL: quickly, met before I should - what is the priority - first dst?
[15:57:36] <Alex Petrescu> room first src
[15:57:46] <Alex Petrescu> LC: I zrote in draft is
[15:57:51] <Alex Petrescu> AL: I think its wrong in slide
[15:58:03] bashi_rururu joins the room
[15:58:06] <Alex Petrescu> LC: maintain def routing table, and ... , an done rt table per upstream
[15:58:15] <Alex Petrescu> LC: explains routing, dst based forwarding
[15:58:37] <Alex Petrescu> LC: second, do best match of src of packet into the accpetable sources, that gives which rt table, and fwd based on that
[15:58:38] Lee Howard leaves the room
[15:58:42] <Alex Petrescu> LC: if no match, then def rt table
[15:58:51] <Alex Petrescu> AL: right... could be collapsed to two loookups
[15:58:55] <Alex Petrescu> LC: Internet
[15:58:58] <Alex Petrescu> Kerry Lynn is KL
[15:59:12] <Alex Petrescu> KL: deprecte adresses? propagate back to hosts?
[15:59:14] <Alex Petrescu> LC: yes have to
[15:59:24] <Alex Petrescu> KL: speed of doing so? simultaneouslyr?
[15:59:30] <Alex Petrescu> KL: why not redundant ISP
[15:59:47] <Alex Petrescu> LC: 1 speed of propa - assumes rt protocol running, tspeed of ISP is...
[15:59:55] <Alex Petrescu> LC: homenet router assigned prefix on that
[16:00:01] <Alex Petrescu> LC: hjost will immediately know
[16:00:09] <Alex Petrescu> LC: backup failorver - cant do that because.
[16:00:20] <Alex Petrescu> KL: you just said ISP down... cant use that src address
[16:00:41] <Alex Petrescu> LC: explains - we tell host by deprecating - look odnt use, yse I take your point we could have ones that always depr?
[16:00:44] <Alex Petrescu> LC: noted
[16:00:52] <Alex Petrescu> AMG I reaaly llike this approach
[16:01:23] <Alex Petrescu> AMG: explanatation wtiiht rt tables mutliple? not necessary. htere are millions other ways, at least three other ways in linux which dont require multi rt tables
[16:01:32] <Alex Petrescu> LC: you say write down reqs? and one example?
[16:01:43] <Alex Petrescu> AMG: yes, right; certainly if mumltiple rt tables work... then..
[16:01:55] <Alex Petrescu> AMG: autocon fprocess - if you want to say, maybe in enterprise section
[16:02:07] <Alex Petrescu> LC: worry?
[16:02:10] <Alex Petrescu> AMG: enterprise
[16:02:12] <Alex Petrescu> sorry
[16:02:19] <Alex Petrescu> cochair cuts
[16:02:33] <Alex Petrescu> MR: first method - would you still do that for 2 router adjacent?
[16:02:37] <Alex Petrescu> LC: src+dst?
[16:02:41] <Alex Petrescu> MR: no, the...
[16:02:47] <Alex Petrescu> MR: that s what happens if ...
[16:02:50] <Alex Petrescu> LC: you have to
[16:02:55] <Alex Petrescu> MR: but if you look on wire
[16:03:02] <Alex Petrescu> MR: gw2 they are both
[16:03:06] <Alex Petrescu> room laughs claps
[16:03:27] <Alex Petrescu> MR: in this case, router with an R woul dhave to, cloud in middle, if pick wrong, routers have to fix for them
[16:03:35] <Alex Petrescu> LC: if on elink, and oone host , and two routers
[16:03:41] <Alex Petrescu> LC: forwards to adjacent
[16:03:57] <Alex Petrescu> LC: could do that wioth only dst based routing, if theyre adjacent, otherwise have to this
[16:04:04] <Alex Petrescu> MR: yes, aggree
[16:04:08] ribiere@cisco.com joins the room
[16:04:23] <Alex Petrescu> BC: req about ISP... dorp ISP drops then you deprecate prefix? problematic with RFC62
[16:04:41] <Alex Petrescu> BC: no router at that time? if this legacy routers in network? this other router is new?
[16:04:49] <Alex Petrescu> BC: router would say not able to route that prefix
[16:04:53] <Alex Petrescu> BC: solution should be resilient
[16:05:04] <dthaler> that's Barbara Stark
[16:05:09] <Alex Petrescu> LC: gateway whole;, have not yet figure out how to make ok with..
[16:05:15] <Alex Petrescu> (sorry, thats BS right)
[16:05:22] ribiere@cisco.com leaves the room
[16:05:30] <Alex Petrescu> LC: maybe right, maybe need to hack around it
[16:05:42] <Alex Petrescu> John Brozowsky of Broadcast? is JB
[16:06:06] <Alex Petrescu> JB: failorver, operator community, whats the rationaliwation of multiple link? or just we can solve it?
[16:06:07] <Alex Petrescu> cochair cuts
[16:06:12] <dthaler> John Brzowsky (unsure of spelling) of Comcast
[16:06:19] <Lee Howard75432> Brzozowski
[16:06:26] <dthaler> yeah that :)
[16:06:32] <Alex Petrescu> LC: idea maybe one thing istwo ISPs could gather, get rid of mye problem home
[16:06:43] <Alex Petrescu> DC
[16:06:45] <Alex Petrescu> DC: explains
[16:06:56] <Alex Petrescu> JB: not my question was
[16:07:20] <Alex Petrescu> LC: my Internet connection, two yuear ago, my Comcast my LTE modem I want to route packets, yes in that case I need metrics
[16:07:25] <Alex Petrescu> JB: not my question was
[16:07:41] <Michael Richardson> is there a draft about this? I want to know to reference this work on the llist.
[16:07:46] <Alex Petrescu> JB: youve seen it somewhere? where did you see this simultaneous active connections?
[16:07:52] <Alex Petrescu> LC: I have this my home, my garden
[16:08:08] <Alex Petrescu> LC: dont solve this problem. small percentage with people having this will do hacks
[16:08:19] <Alex Petrescu> LC: correct, rare usecase today, more common as time goes on
[16:08:22] <Alex Petrescu> LC: not have to
[16:08:32] <Alex Petrescu> LC: because you have dhcp, permanent essentially is about...
[16:08:39] <Michael Richardson> this is rare today because it doesn't work.
lots of people with small offices/home offices have two links.
[16:08:39] <Alex Petrescu> LC: tomorrow I expect thsi commong
[16:08:58] <Alex Petrescu> DC: I dont want two simultaneous active links an dI want both used no break connections
[16:09:10] <Alex Petrescu> TC: similar, useful archi text, warning lights in archi text
[16:09:21] <Alex Petrescu> TC: deprecated prefixes? you must have ULAs
[16:09:25] <Alex Petrescu> LC: yes, need ULAs
[16:09:28] <Alex Petrescu> TC: just check
[16:09:35] <Alex Petrescu> LC: onc eyou have connection, yes.
[16:09:48] <Alex Petrescu> Dino Farinacci Cisco is DF
[16:09:55] <Alex Petrescu> DF: questions back to example
[16:10:14] <Alex Petrescu> DC: at my house I have, I want mutihome, but I dont need muti address
[16:10:21] <Alex Petrescu> DC: I could split this way
[16:10:48] <Alex Petrescu> DC: immportant;, my host has a static addres - I move out my home into car, survivability maybe not, but stikll, set of VIDs
[16:10:58] <Alex Petrescu> DC: why not use LISP?
[16:11:05] <Alex Petrescu> Dsorry its DF
[16:11:14] <Alejandro Acosta (monitor)> Alex, in case I forget later or we ran out of time thanks a lot for the great transcription you have been doing
[16:11:22] <Alex Petrescu> DF: with LISP no need to do loockups.
[16:11:29] <Alex Petrescu> LC: solution space - no outside help
[16:11:42] <Alex Petrescu> DF: you believe LISP needs outside help?
[16:11:50] <Alex Petrescu> DF: LISTP just as any other
[16:11:54] <Alex Petrescu> Dave Oran is DO
[16:12:03] <Alex Petrescu> DO: src based routing is out there since time
[16:12:30] <Alex Petrescu> DO: something , try to keep things separately - dont assume mutiple rt tables required - you end up with inconvenients
[16:12:47] <Alex Petrescu> DO: nervous about coupling prefix withdraw with routing co,muptation, because.
[16:12:50] <Alex Petrescu> LC: but they work.
[16:13:18] <Alex Petrescu> DO: if two prefixes, one in middle of ocean... try to couple this with withdrawing - its dangerous
[16:13:22] <Alex Petrescu> LC: ...
[16:13:28] <Alex Petrescu> DO: you do this when links flap?
[16:13:33] <Alex Petrescu> LC: yes
[16:13:34] <Alex Petrescu> cochair do offline
[16:13:36] <Alejandro Acosta (monitor)> Almost starting: homenet and mdnsext
[16:13:37] <Alejandro Acosta (monitor)> -------------------
[16:13:42] <Alex Petrescu> TC presenting
[16:14:00] darshakthakore leaves the room
[16:14:09] <Alejandro Acosta (monitor)> Slide 2: Naming and service discovery
[16:14:15] danyork raises his hand for being in mdnsext :-)
[16:15:50] <Alejandro Acosta (monitor)> slide 3: mdnsext – draH problem statement
[16:15:58] Ole Troan leaves the room
[16:16:38] dthaler leaves the room
[16:16:39] dthaler joins the room
[16:17:05] darshakthakore joins the room
[16:17:13] <Alejandro Acosta (monitor)> Slide 4: mdnsext mail list
[16:17:39] <Alejandro Acosta (monitor)> Slide 5: Mdnsext BoF
[16:17:42] SM joins the room
[16:17:59] <Alex Petrescu> RD : Area Director
[16:18:08] <Alex Petrescu> RD: bof went very well
[16:18:22] <Alex Petrescu> AD: appreciate number of people who put hands in air will contribute to thiat
[16:18:31] <Alex Petrescu> RD: Tims work appreciated for bof happen
[16:18:37] <Alex Petrescu> RD: honestly better than expect
[16:18:43] bashi_rururu leaves the room
[16:18:45] <Alex Petrescu> RD: please participate to mailing list
[16:19:08] <Alex Petrescu> JA queueing
[16:19:59] <Alex Petrescu> JA: comment
[16:20:13] <Alex Petrescu> JA: this WG is already chartered to look at naming, maybe less so than dnsext
[16:20:20] <Alex Petrescu> JA: here focus naming down reqs
[16:20:32] <Alex Petrescu> JA: figure one or multiple scenarios, rathern than uncertainty
[16:20:37] <Alex Petrescu> JA: expand now?
[16:20:46] <Alex Petrescu> JA: work on reqs, not wait for 4 months
[16:20:52] <Alex Petrescu> cochairs position is push reqs
[16:21:00] <Alex Petrescu> cochairs will not address solutions
[16:21:07] <Alex Petrescu> TC: more slide about this
[16:21:14] <Alex Petrescu> DT: next slide go ahead
[16:21:18] <Alejandro Acosta (monitor)> Slide 6: Draft mdnsext goals
[16:21:24] <Alex Petrescu> RD queues
[16:21:30] dave.taht leaves the room
[16:22:16] <Alejandro Acosta (monitor)> Slide 7: Presentations
[16:22:17] dave.taht joins the room
[16:22:44] <Alex Petrescu> cochairs no time very brief
[16:23:12] <Alex Petrescu> DT: bottom stuff in red, mbms - consensus on both service disc and name group? great doc discussion on service disc reqs, approx to homenet
[16:23:17] darshakthakore leaves the room
[16:23:19] <Alex Petrescu> DT: no good doc on naming reqs
[16:23:40] <Alex Petrescu> DT: even if service disc disc, we still need that disc here, and eqs for name resolution
[16:23:58] <Alejandro Acosta (monitor)> Slide 8: Skipped
[16:24:12] <Alex Petrescu> RD: fviolent agreement with JA, no need of work in mdnsext until anothe rbof, do that in parallel ( reqs)
[16:24:23] <Alejandro Acosta (monitor)> Starting: IPv6 Home Network Front End Naming Delegation
[16:24:23] <Alex Petrescu> RD: easy decision, but not wait for it, work on it now.
[16:24:25] <Alejandro Acosta (monitor)> -----------------
[16:24:33] <Alex Petrescu> cochair cuts someones presentation from agenda
[16:24:42] <Alex Petrescu> Daniel Migault is DM presents
[16:24:49] <Lee Howard75432> Ole Troan, last presentation on revised agenda, was ut
[16:24:51] <Lee Howard75432> cut
[16:24:51] cgrundemann leaves the room
[16:24:58] <Alejandro Acosta (monitor)> Slide 2: Goal of the Document
[16:25:30] equinox leaves the room
[16:25:37] <Alejandro Acosta (monitor)> Slide 3: Problem Statement
[16:25:40] <Alex Petrescu> cochair we listened to what you gusy said, thank you, its a WG
[16:26:33] yuaoki joins the room
[16:27:17] <Alejandro Acosta (monitor)> Slide 4: Problem Statement
[16:29:01] <Alejandro Acosta (monitor)> Slide 6: Arquitecture Desc
[16:29:05] <Alejandro Acosta (monitor)> slide 5 skipped
[16:29:19] yuaoki leaves the room
[16:29:54] fdupont leaves the room: Computer went to sleep
[16:30:12] <Alejandro Acosta (monitor)> Slide 7: Architecture Description
[16:30:23] <Alex Petrescu> cochairs ask presenter about time
[16:30:29] <Alex Petrescu> presenter says almost done
[16:30:35] <Alejandro Acosta (monitor)> Slide 8: CPE Configuration
[16:30:53] equinox joins the room
[16:31:15] <Alejandro Acosta (monitor)> Slide 9: CPE Configuration
[16:31:29] <Alejandro Acosta (monitor)> Slide 10: Position toward Homenet Architecture
[16:31:40] <Alex Petrescu> DT queueing
[16:31:43] <Alejandro Acosta (monitor)> Slide 11: Next Steps
[16:31:52] <Alex Petrescu> Francis Dupont is FD queueing
[16:31:54] <Alex Petrescu> cochairs cut
[16:32:01] Andrew Sullivan leaves the room
[16:32:06] <Alex Petrescu> FD: suggest send email to dns mail list
[16:32:11] <Alex Petrescu> cochairs no time
[16:32:21] Jason Weil leaves the room
[16:32:24] <Alex Petrescu> cochairs sign bluesheets
[16:32:30] mbehring@cisco.com leaves the room
[16:32:33] <Alex Petrescu> cochairs humming?
[16:32:35] pselkirk leaves the room
[16:32:40] <Alex Petrescu> cochairs see another round for LC of
[16:32:44] SM leaves the room
[16:32:47] <Alejandro Acosta (monitor)> @Alex, thanks a lot for your transcription
[16:32:48] <Alex Petrescu> cochairs another round? hum!
[16:32:52] Markus Stenberg leaves the room
[16:32:56] <Alex Petrescu> not another round? hum!
[16:32:58] <Alex Petrescu> it sanged
[16:32:58] Olafur leaves the room
[16:33:06] wmtownsley leaves the room
[16:33:07] ctg170187633 leaves the room
[16:33:11] zcao leaves the room
[16:33:19] teco.boot leaves the room
[16:34:14] joel jaeggli leaves the room
[16:34:51] suz leaves the room
[16:35:39] Brian Carpenter leaves the room
[16:35:39] Robert Cragie leaves the room
[16:36:34] Hassan Zaheer leaves the room
[16:37:07] <Alex Petrescu> adjourned
[16:37:10] Alex Petrescu leaves the room
[16:39:07] joel jaeggli joins the room
[16:39:18] dthaler leaves the room
[16:39:22] cheshire leaves the room
[16:39:38] cabo leaves the room
[16:40:09] dave.taht leaves the room
[16:40:21] Michael Richardson leaves the room
[16:40:38] Michael Richardson joins the room
[16:41:08] Michael Richardson leaves the room
[16:41:31] Michael Richardson joins the room
[16:44:52] Alejandro Acosta (monitor) leaves the room: Machine going to sleep
[16:45:06] Jason Weil joins the room
[16:45:07] cabo joins the room
[16:46:21] Lee Howard75432 leaves the room
[16:46:59] intvelt leaves the room
[16:52:30] Jason Weil leaves the room
[17:01:41] ctg1701 joins the room
[17:04:46] Brian Carpenter joins the room
[17:07:19] Markus Stenberg joins the room
[17:08:35] Markus Stenberg leaves the room
[17:11:36] Brian Carpenter leaves the room: offline
[17:24:59] ogudmund joins the room
[17:27:49] equinox leaves the room
[17:33:08] ogudmund leaves the room
[17:48:13] Markus Stenberg joins the room
[17:49:16] ctg1701 leaves the room
[17:53:51] Markus Stenberg leaves the room
[17:54:20] woolf joins the room
[17:56:24] cabo leaves the room
[18:03:02] Ole Troan joins the room
[18:03:25] Ole Troan leaves the room
[18:03:55] Ole Troan joins the room
[18:04:21] Ole Troan leaves the room
[18:04:25] Ole Troan joins the room
[18:04:29] Michael Richardson leaves the room
[18:04:34] Robert Cragie joins the room
[18:04:38] joel jaeggli leaves the room
[18:05:02] Robert Cragie leaves the room
[18:05:06] cabo joins the room
[18:05:20] woolf leaves the room
[18:05:22] Michael Richardson joins the room
[18:05:54] joel jaeggli joins the room
[18:11:15] pselkirk joins the room
[18:12:04] joel jaeggli leaves the room
[18:12:31] joel jaeggli joins the room
[18:25:27] ctg1701 joins the room
[18:33:55] danyork leaves the room
[18:34:16] danyork joins the room
[19:12:19] Michael Richardson leaves the room
[19:12:27] Michael Richardson joins the room
[19:14:03] William Lupton leaves the room
[19:20:27] Michael Richardson leaves the room
[19:20:33] Michael Richardson joins the room
[19:21:12] Michael Richardson leaves the room
[19:27:13] Ole Troan leaves the room
[19:31:09] Michael Richardson joins the room
[19:31:25] cabo leaves the room
[19:33:09] Michael Richardson leaves the room
[19:33:26] pselkirk leaves the room
[19:33:50] ctg1701 leaves the room
[19:33:52] Michael Richardson joins the room
[19:34:15] Michael Richardson leaves the room
[19:35:03] Michael Richardson joins the room
[19:35:24] joel jaeggli leaves the room
[19:37:19] Michael Richardson leaves the room
[19:39:30] joel jaeggli joins the room
[19:39:35] William Lupton joins the room
[19:41:06] cabo joins the room
[19:49:18] Michael Richardson joins the room
[19:51:21] Michael Richardson leaves the room
[19:51:21] William Lupton leaves the room
[19:51:39] Michael Richardson joins the room
[19:52:28] Michael Richardson leaves the room
[19:54:03] Michael Richardson joins the room
[19:57:21] ctg1701 joins the room
[20:00:25] ctg1701 leaves the room
[20:00:25] ctg1701 joins the room
[20:21:56] cabo leaves the room
[20:22:09] cabo joins the room
[20:23:55] cabo leaves the room
[20:42:45] joel jaeggli leaves the room
[20:43:45] joel jaeggli joins the room
[20:48:24] William Lupton joins the room
[20:53:27] joel jaeggli leaves the room
[20:56:50] joel jaeggli joins the room
[21:03:13] pselkirk joins the room
[21:06:36] ctg1701 leaves the room
[21:08:30] Michael Richardson leaves the room
[21:09:17] Michael Richardson joins the room
[21:09:36] cabo joins the room
[21:13:24] ctg1701 joins the room
[21:14:41] pselkirk leaves the room
[21:14:45] pselkirk joins the room
[21:19:26] ctg1701 leaves the room
[21:19:26] ctg1701 joins the room
[21:21:00] Michael Richardson leaves the room
[21:21:11] Michael Richardson joins the room
[21:26:33] ctg170167718 joins the room
[21:27:26] ctg1701 leaves the room
[21:27:26] ctg170167718 leaves the room
[21:28:26] ctg1701 joins the room
[22:05:11] William Lupton leaves the room
[22:09:01] joel jaeggli leaves the room
[22:09:40] ctg1701 leaves the room
[22:10:50] Michael Richardson leaves the room
[22:11:15] Michael Richardson joins the room
[22:11:32] Michael Richardson leaves the room
[22:11:49] Michael Richardson joins the room
[22:12:10] Michael Richardson leaves the room
[22:12:18] Michael Richardson joins the room
[22:16:26] cabo leaves the room
[22:18:50] William Lupton joins the room
[22:33:45] pselkirk leaves the room
[22:36:57] danyork leaves the room
[22:40:08] cabo joins the room
[22:41:37] joel jaeggli joins the room
[22:42:30] Michael Richardson leaves the room
[22:42:39] Michael Richardson joins the room
[22:50:36] ctg1701 joins the room
[22:54:12] wmtownsley joins the room
[22:54:53] wmtownsley leaves the room
[23:29:37] joel jaeggli leaves the room
[23:42:50] joel jaeggli joins the room
Powered by ejabberd Powered by Erlang Valid XHTML 1.0 Transitional Valid CSS!