v6ops - Thursday, 28 July 2011

v6ops

v6ops@jabber.ietf.org

Thursday, 28 July 2011< ^ >

Patrik Halfar has set the subject to: v6ops -- IETF80

Room Configuration

GMT+0
[01:29:56] liubingpang joins the room
[01:36:06] liubingpang leaves the room
[01:42:25] liubingpang joins the room
[01:59:44] liubingpang leaves the room
[05:26:53] Carlos Martinez leaves the room
[11:51:43] Joel Jaeggli leaves the room
[12:04:16] Joel Jaeggli joins the room
[12:04:17] danny leaves the room
[12:49:49] Joel Jaeggli leaves the room
[12:51:25] Joel Jaeggli joins the room
[12:54:22] Joel Jaeggli leaves the room
[12:55:32] Carlos Martinez joins the room
[12:58:43] Joel Jaeggli joins the room
[12:59:52] Joel Jaeggli leaves the room
[13:00:02] liubingpang joins the room
[13:00:18] Joel Jaeggli joins the room
[15:37:32] liubingpang leaves the room
[15:47:06] Joel Jaeggli leaves the room
[15:49:28] Carlos Martinez leaves the room
[15:51:28] Joel Jaeggli joins the room
[16:27:05] liubingpang joins the room
[16:52:24] liubingpang leaves the room
[16:59:02] Joel Jaeggli leaves the room
[17:03:25] Joel Jaeggli joins the room
[17:10:38] Joel Jaeggli leaves the room
[17:32:58] doug.otis joins the room
[17:40:09] doug.otis leaves the room
[17:53:52] Joel Jaeggli joins the room
[18:32:03] jpc joins the room
[18:58:21] Joel Jaeggli leaves the room
[19:02:25] jpc leaves the room
[19:04:20] Joel Jaeggli joins the room
[19:14:56] juampe.cerezo joins the room
[19:15:29] Bjoern A. Zeeb joins the room
[19:16:40] Joel Jaeggli leaves the room
[19:16:59] Arturo Servin Ü joins the room
[19:17:35] Joel Jaeggli joins the room
[19:18:37] <Joel Jaeggli> V6ops Begins in 2 minutes
[19:18:48] <Bjoern A. Zeeb> Can you update the topic?
[19:19:33] <Joel Jaeggli> not with this client
[19:19:44] <Joel Jaeggli> also I can't even see it
[19:19:58] miyano joins the room
[19:20:30] tsavo_work@jabber.org/Meebo joins the room
[19:20:57] Arturo Servin Ü leaves the room
[19:21:55] Akira Kato joins the room
[19:23:15] Wes George joins the room
[19:24:02] dudisaki joins the room
[19:24:02] <Joel Jaeggli> meeting commences
[19:24:09] <Joel Jaeggli> fred - agenda
[19:24:44] Ruri Hiromi joins the room
[19:24:48] danwing joins the room
[19:24:53] jinmei joins the room
[19:25:06] sven joins the room
[19:26:07] Atarashi Yoshifumi joins the room
[19:26:17] <Joel Jaeggli> if we moved two presentations to friday we could give you 18 minutes instead of 13
[19:27:19] lochii joins the room
[19:27:21] shinmiyakawa joins the room
[19:27:29] <Joel Jaeggli> wojec dec presenting for ole - stateless 4v6 on behalf of the other offers
[19:27:51] Barbara S joins the room
[19:28:51] <Joel Jaeggli> 4v6 ipv6 address - this is a regular ipv6 address except for the 4v6 index
[19:29:10] <Joel Jaeggli> described in the solutions documents
[19:29:28] Ole Troan joins the room
[19:29:33] <Joel Jaeggli> pretty much the only prefix that needs to be passed
[19:30:25] <Joel Jaeggli> 4v6 index provides enough information to comput and ipv4 address and port index.
[19:30:30] <Joel Jaeggli> mapped tunneling
[19:30:38] <Joel Jaeggli> examples
[19:31:33] <Joel Jaeggli> got a cpe, and it has inside it a port constrained nat44 function, along with a 4v6 function
[19:32:06] <Ole Troan> anyone on audio wanting slide numbers?
[19:32:11] <Joel Jaeggli> native ipv6 network - then 4v6 gateway
[19:34:02] <Joel Jaeggli> inside the cpe we nat the source ip to a public v4 address and port range provided by the v6 address encapsulated and sent to the 4v6 gateway where it is set on the internet
[19:34:16] jweil joins the room
[19:34:29] <Joel Jaeggli> on the way back the difference is more accentuated.
[19:34:31] weiyinxing joins the room
[19:37:31] <Joel Jaeggli> dave thayler - along with supports v4 options also supports icmpv4 error codes
[19:38:12] <Joel Jaeggli> exmple operation of 4v6 in a 3gpp system
[19:40:00] gigix73 joins the room
[19:40:57] <Joel Jaeggli> translation 4v6 versus mapped mode in 3gpp translation has signficantly less impact
[19:42:09] liubingpang joins the room
[19:42:46] <Joel Jaeggli> conclusion in the draft - operational impact of both appraoches should be documented
[19:43:08] arifumi joins the room
[19:43:23] <Joel Jaeggli> ole troan - I just wanted to state for the mintues that there isn't any difference between encapsulation and double translation
[19:43:29] Brian Haberman joins the room
[19:43:37] <tsavo_work@jabber.org/Meebo> I thought IETF did not want double translation.
[19:43:39] <Joel Jaeggli> do you see this going forward in softwires
[19:43:39] behcet.sarikaya joins the room
[19:44:20] <Joel Jaeggli> dec - we didn't get to have the dicussion in softwires. belive that v6ops is best at captuing the opperational issues.
[19:44:51] EuiJong Hwang joins the room
[19:45:17] <Joel Jaeggli> hue deng - would like less impact.
[19:45:45] <Joel Jaeggli> dec -dispell aguements from the mic
[19:46:21] <Joel Jaeggli> ? shima - going to be a work item for softwires.
[19:47:20] <Joel Jaeggli> mark handley - the vast majority of both operations are the same the question is do we compress the headers out or not. I support the minimal effort which is encapsulation.
[19:47:44] <lochii> s/handley/townsley
[19:48:13] Joonhyung Lim joins the room
[19:48:17] <Joel Jaeggli> in terms of where this work gets done, I think it's a faulure of our process
[19:48:47] <Joel Jaeggli> alain durand - this should be done in softwire, this should be done in an interim meeting in sept.
[19:49:03] <Joel Jaeggli> fred - so we're actually going to have the discussion
[19:49:42] <Joel Jaeggli> jari a - we should put them in the same place (softwire) rather than split them all over the place.
[19:50:37] <Joel Jaeggli> sing yue from cernet - sometimes we deploy single translations sometimes double, we haven't found a corner case where one works and the other doesn't
[19:51:00] <Joel Jaeggli> please move this faster
[19:52:11] <Joel Jaeggli> rajiv sari - you got a lot of use to understand the pros and cons of this solution. Help up us quantify the operational impact of each approach.
[19:52:11] resnick joins the room
[19:52:23] mcharlesr joins the room
[19:52:55] <Joel Jaeggli> fred - suggest that people here, read your draft and comment ot softwire where the work will be done
[19:53:24] <Joel Jaeggli> alain - trying to work on a document that deals with the different quadrants of the problem space.
[19:54:02] geir joins the room
[19:54:17] <Joel Jaeggli> dec - question of whether softwire can do translation work
[19:54:55] <Joel Jaeggli> dave t - the difference between tunneling and translation is very small, would be an odd division to put those in different places.
[19:55:22] <Joel Jaeggli> fred - think the matter is closed
[19:55:55] <Joel Jaeggli> hemant singh - cpe routers
[19:56:01] <Joel Jaeggli> presentation
[19:57:01] <Joel Jaeggli> a few changes, with the so called advanced version of this draft
[19:57:41] <Barbara S> Sure doesn't sound like Hemant. :) Sounds like Ole. Is there a presentation from Ole?
[19:57:58] PETRESCU-EXOPC joins the room
[19:57:59] <Joel Jaeggli> latest version of draft is not the one the screen
[19:58:07] <Joel Jaeggli> er on the website
[19:58:16] <Joel Jaeggli> sorry
[19:58:28] <Joel Jaeggli> s/draft/slides
[19:59:02] <Joel Jaeggli> updates to the 01 version
[19:59:08] <Joel Jaeggli> removed nd proxy
[19:59:37] <Joel Jaeggli> prefix delegation was agreed with mobile netowrks
[19:59:55] <Joel Jaeggli> dhcpv6 calrification
[19:59:56] vincent-afnic joins the room
[20:00:33] <Joel Jaeggli> request from homenet to move the lan side of the requirements, to the homenet wg
[20:00:56] <Joel Jaeggli> document will be split into wan requirements vs lan side which will go there.
[20:01:44] <Joel Jaeggli> should we do the split or go forward with the document we have
[20:01:48] <Joel Jaeggli> options
[20:01:54] <Joel Jaeggli> give up on document
[20:01:55] lochii leaves the room
[20:02:25] <Joel Jaeggli> move to softwires (dslit parts)
[20:02:56] <Joel Jaeggli> resumbit 6204 with dslite and 6rd requirements
[20:02:59] <Joel Jaeggli> continue
[20:03:26] lochii joins the room
[20:04:33] <Barbara S> I like 6204+ option or option 4
[20:04:53] <Joel Jaeggli> ? - include requiremnts for pcp?
[20:05:27] <Joel Jaeggli> alain - pcp has been through two wg last calls
[20:05:45] lochii leaves the room
[20:05:52] Nick Hilliard joins the room
[20:06:11] lochii joins the room
[20:06:17] <Joel Jaeggli> mark t - 6rd and dslite and and pcp and making them come up and behave well together, that work is already here.
[20:06:22] kfantonio joins the room
[20:08:42] victor.kuarsingh joins the room
[20:09:05] <Barbara S> inclusion of PCP will start discussions of do we also include UPnP
[20:09:48] <Joel Jaeggli> alain - on the cpe itself we have a bunch of dhcp options we fire see what happens and then cancel the others.
[20:10:43] <Joel Jaeggli> erik from cisco - we may end up with two documents which are not complatible if we split them
[20:11:08] <Joel Jaeggli> hemant - homenet can't do things on the lan side thta might have impact else where.
[20:11:08] geir leaves the room
[20:12:31] <Joel Jaeggli> lorenzo - what 6204 are doing and what homenet is doing are relatively partitionable problem.
[20:12:50] <Joel Jaeggli> I don't know that you need to make homenet have al lthe lanside requirements.
[20:14:09] <Joel Jaeggli> on the other questions I don't understand why 6204 which is about how they interact, we didn't put those pieces in their respective working groups rather we did here was show how to make them work together.
[20:14:42] Dave Thaler joins the room
[20:14:43] ahpj@jabber.org/Meebo joins the room
[20:15:07] geir joins the room
[20:15:25] <Joel Jaeggli> fred - question for barbara does that start here or on bbf?
[20:15:29] <Barbara S> start them wherever someone is suggesting referencing pcp
[20:15:30] <Joel Jaeggli> or softwire
[20:16:42] velt joins the room
[20:16:43] <Joel Jaeggli> alain − discussion starts with relay mode
[20:17:38] <Joel Jaeggli> joel - spliting it was my expression this morning.
[20:17:40] geir leaves the room
[20:18:02] <Joel Jaeggli> hemant - lan side requirements are where we have real urgency from operators
[20:18:45] dougm.work joins the room
[20:18:48] <Joel Jaeggli> fred - unless homenet feels badly then my expression would be to go for the thing with the nearest completion.
[20:20:03] <Joel Jaeggli> mark t - the wan side type I agree with you, the lan side, homenet should look a, we're now 9 days old
[20:20:49] <Joel Jaeggli> erik ckline we need to have cpe vendors to have built this aerly next year.
[20:21:26] <Joel Jaeggli> jari arko - we should publish what we have, if there's flaky stuff we should pull it.
[20:21:36] Tomas Podermanski joins the room
[20:22:18] <Joel Jaeggli> lorenzo - in a closer reading some of the lan side stuff, dhcp will definintly hurt homenet work
[20:22:40] <Barbara S> thx for relaying me
[20:22:46] <Joel Jaeggli> chen gun ipv6 practices in china mobile ip
[20:22:52] <Joel Jaeggli> - no speaker found
[20:23:23] <Joel Jaeggli> ipv6 router advertisement guard -artouro
[20:23:33] Carlos Martinez joins the room
[20:23:35] <Joel Jaeggli> presently on behalf of fernando gont
[20:23:38] rjs joins the room
[20:24:17] <Joel Jaeggli> rfc 6104 introduces problem statement
[20:24:33] <Joel Jaeggli> 6105 specifices ra-guard
[20:24:43] sven leaves the room
[20:25:49] keith.brazington joins the room
[20:26:31] <Joel Jaeggli> ra-guard-evasion describes evasion attack
[20:28:33] geir joins the room
[20:28:35] josoinin joins the room
[20:29:14] <Joel Jaeggli> mitigations
[20:29:14] fgont joins the room
[20:30:36] <Joel Jaeggli> fred - so you made this presentation in 6man - what was the upshot in 6man ?
[20:30:52] <fgont> They are two different things: in 6man its about an ipv6 spec check
[20:31:04] <Joel Jaeggli> autoro - he asking for acptanct becuase it updates 6105
[20:31:22] <fgont> in this case, it's documentation of the issue (update the ra-guard spec), and documentation of an operational solution
[20:31:59] <fgont> thanks joel
[20:32:07] <fgont> NOt really
[20:32:14] <fgont> ra-guard is an informational document
[20:33:02] <fgont> note that ra-guard was produced by v6ops, not 6man
[20:33:42] <fgont> ra-guard was produced in v6ops, not 6man
[20:33:48] <Joel Jaeggli> ron -we would be be steping on the charter of 6man (joel outside of the problem statemetnt)
[20:34:11] <Joel Jaeggli> fred- we'll talk with bob and brian but it seems like the work can be done in 6man
[20:34:31] <Joel Jaeggli> chen gun ipv6 practices in china mobile ip network
[20:35:03] victor.kuarsingh leaves the room
[20:36:28] <Joel Jaeggli> china mobile ipv6 progam has started 3 years ago
[20:37:26] <Joel Jaeggli> slide testing methodology in ip bearer networks
[20:41:06] fgont leaves the room
[20:41:27] Bill joins the room
[20:41:36] Bill leaves the room
[20:41:55] fgont joins the room
[20:42:29] rjs leaves the room: Replaced by new connection
[20:44:11] victor.kuarsingh joins the room
[20:44:20] <Joel Jaeggli> wes george - comments on presntation and draft when you talk about mtu the transport network needs to have a transport network sized so that you don't have to reduce mtu
[20:45:35] <Joel Jaeggli> we find that the vendor doesn't support this. you're probably at the point where name and shapeme is a good idea - the avalue of bringing it to a public forum is just that . are there things that the ietf should be doing with your experienced.
[20:46:04] <Joel Jaeggli> chen - gap anaylsis is necessary
[20:47:03] <Joel Jaeggli> chris palmer - empathis strong ly I think it's great but I have a question, are you driving load from your users against. this
[20:47:13] <Joel Jaeggli> we do have plan for that.
[20:47:43] <Joel Jaeggli> chris palmer would like a trial
[20:48:25] Dave Thaler leaves the room
[20:49:08] <Joel Jaeggli> hue deng - we do have a project with a national populatio that will be done this year
[20:49:33] <PETRESCU-EXOPC> Hui Deng talked about many users and new trials later this year
[20:49:50] <Joel Jaeggli> new presentation nat64 cpe operations
[20:50:25] Joonhyung Lim leaves the room: Replaced by new connection
[20:50:25] Joonhyung Lim joins the room
[20:50:28] <PETRESCU-EXOPC> (He also said that this, as others said, may be useful in determining useful things to do at ietf)
[20:50:30] resnick leaves the room
[20:51:18] <Joel Jaeggli> yup thansk
[20:52:04] <Joel Jaeggli> hue deng - I guess this kind of deployment has been widely used
[20:53:14] <Joel Jaeggli> fred - if the other draft is covering the same material at least they could be merged.
[20:54:42] <Joel Jaeggli> dave freedman - A Discard Prefix for IPv6 <http://tools.ietf.org/html/draft-hilliard-v6ops-ipv6-discard-prefix>
[20:55:03] <Joel Jaeggli> formally reserve a prefix for the list
[20:55:10] <Nick Hilliard> how many, roughly?
[20:55:16] <Joel Jaeggli> er for the discard prefix
[20:55:18] <Nick Hilliard> yep
[20:55:21] <Joel Jaeggli> half a dozen
[20:55:48] <Nick Hilliard> ok, curious about this from the point of view of taking comments on board
[20:55:50] <Joel Jaeggli> since most providers use this feature there needs to be a best practice for it
[20:56:03] <Joel Jaeggli> people will commnet on anything
[20:56:10] <Nick Hilliard> heh
[20:56:18] <Joel Jaeggli> use the documentation prefix?
[20:56:21] <Joel Jaeggli> yeah not
[20:56:36] <Joel Jaeggli> do we want one address or more than one?
[20:56:51] <Joel Jaeggli> I want to make a cute hex
[20:56:54] <Joel Jaeggli> not
[20:57:12] <Joel Jaeggli> does it need to be global unicast or not?
[20:57:29] <Joel Jaeggli> operational problem?
[20:57:56] <Joel Jaeggli> hard wired discard is a standards track item
[20:58:15] <Joel Jaeggli> is there anything else?
[20:58:28] <Joel Jaeggli> comments - james woodyat why not ual?
[20:58:55] <Joel Jaeggli> ula
[20:59:33] <Joel Jaeggli> wes goerge - one comment I make given that you don't want it to leak if it's a /64 it probably with hit most people's filters
[21:00:01] <Nick Hilliard> there are conceivable situation where you would want to leak the prefix
[21:00:02] <Joel Jaeggli> what about using bogons?
[21:00:04] <Nick Hilliard> e.g. to downstreams
[21:00:11] <Nick Hilliard> for DoS analysis, etc
[21:00:25] sm joins the room
[21:00:27] <Joel Jaeggli> lorenzo one address question is do we need more
[21:01:08] <Joel Jaeggli> warren having multpile discards would be useful like 5 or 6
[21:01:21] <Joel Jaeggli> rob - question for the wg don't have a bcp
[21:01:26] <Joel Jaeggli> on implmetnation
[21:03:25] Bill joins the room
[21:03:29] <Joel Jaeggli> warren - I'd rather everyone uses the same
[21:04:21] <Joel Jaeggli> a /64 is plenty big but you do need more than one.
[21:04:59] <Joel Jaeggli> fred - hearing a fair umount of support - (hum)
[21:05:02] <Joel Jaeggli> in favor
[21:05:06] <Joel Jaeggli> no opposed
[21:05:15] <Joel Jaeggli> short prefix vs /128
[21:05:28] <Joel Jaeggli> preference for /64
[21:05:36] <Joel Jaeggli> arguement for a bcp makes sense
[21:05:36] Carlos Martinez leaves the room
[21:06:08] <Joel Jaeggli> once the prefix length. Lets take this to a working group last call, and move it ahead.
[21:06:14] Dave Thaler joins the room
[21:06:36] fgont leaves the room
[21:06:49] victor.kuarsingh leaves the room
[21:06:52] <Joel Jaeggli> fred - dhcpv6 behcet sarikaya
[21:06:58] Nick Hilliard leaves the room
[21:07:36] fgont joins the room
[21:07:50] <Joel Jaeggli> lat time it was presented was 02 − 07 was idnits and final revison
[21:09:53] <Joel Jaeggli> slide dhcpd for stateful address configuration
[21:10:36] EuiJong Hwang leaves the room
[21:11:09] Ruri Hiromi leaves the room
[21:12:12] <Joel Jaeggli> fred - when we asked about after updates it seemed like there was support for it
[21:12:22] <Joel Jaeggli> wg do you want to see this as a wg item?
[21:12:28] Ruri Hiromi joins the room
[21:12:49] <Joel Jaeggli> heard a couple each way
[21:14:05] <Joel Jaeggli> show of hands show not strong support
[21:14:49] velt leaves the room
[21:15:05] <Joel Jaeggli> nor for support for indiviual submission.
[21:15:14] EuiJong Hwang joins the room
[21:16:34] gigix73 leaves the room
[21:16:44] vincent-afnic leaves the room
[21:16:45] <Joel Jaeggli> meeting over… which drafts are remaining for tomorrow?
[21:17:02] geir leaves the room
[21:17:06] Atarashi Yoshifumi leaves the room
[21:17:07] kfantonio leaves the room
[21:17:11] behcet.sarikaya leaves the room
[21:17:14] <Joel Jaeggli> port sharing, rogers, wireline individual, gundavelli and v6nd
[21:17:24] Ruri Hiromi leaves the room
[21:17:25] <Joel Jaeggli> thank you
[21:17:30] jinmei leaves the room
[21:17:41] EuiJong Hwang leaves the room
[21:17:42] josoinin leaves the room
[21:17:54] Akira Kato leaves the room
[21:18:03] dudisaki leaves the room
[21:18:06] arifumi leaves the room
[21:18:06] jweil leaves the room
[21:18:24] Ole Troan leaves the room
[21:18:32] mcharlesr leaves the room
[21:18:33] mcharlesr joins the room
[21:18:36] ahpj@jabber.org/Meebo leaves the room
[21:18:42] juampe.cerezo leaves the room
[21:19:48] danwing leaves the room
[21:20:21] Brian Haberman leaves the room
[21:20:32] miyano leaves the room
[21:20:35] Joel Jaeggli leaves the room
[21:20:37] Bill leaves the room: Computer went to sleep
[21:22:43] Tomas Podermanski leaves the room
[21:23:10] dougm.work leaves the room
[21:24:29] sm leaves the room
[21:25:36] Wes George leaves the room
[21:25:36] Dave Thaler leaves the room
[21:25:36] PETRESCU-EXOPC leaves the room
[21:28:36] shinmiyakawa leaves the room
[21:29:22] Barbara S leaves the room
[21:31:11] keith.brazington leaves the room
[21:31:27] Joel Jaeggli joins the room
[21:32:02] geir joins the room
[21:33:52] Tomas Podermanski joins the room
[21:34:18] Tomas Podermanski leaves the room
[21:35:36] fgont leaves the room
[21:37:29] tsavo_work@jabber.org/Meebo leaves the room
[21:39:00] Dave Thaler joins the room
[21:39:52] Wes George joins the room
[21:40:54] jinmei joins the room
[21:41:03] jinmei leaves the room
[21:42:12] Dave Thaler leaves the room
[21:45:10] PETRESCU-EXOPC joins the room
[21:46:36] geir leaves the room
[21:47:44] Wes George leaves the room
[21:49:52] jweil joins the room
[21:52:30] arifumi joins the room
[21:57:22] Ruri Hiromi joins the room
[22:03:29] jweil leaves the room
[22:05:05] arifumi leaves the room
[22:09:58] Bill joins the room
[22:10:01] Bill leaves the room
[22:12:52] Bjoern A. Zeeb leaves the room
[22:23:51] mcharlesr leaves the room
[22:24:53] Joel Jaeggli leaves the room
[22:40:18] Joonhyung Lim leaves the room
[22:41:52] Joel Jaeggli joins the room
[22:45:47] danwing joins the room
[22:47:17] danwing leaves the room
[22:47:22] danwing joins the room
[23:00:03] Ruri Hiromi leaves the room
[23:21:30] Ole Troan joins the room
[23:24:04] Joel Jaeggli leaves the room
[23:26:22] danwing leaves the room
[23:32:10] Ole Troan leaves the room
[23:39:07] PETRESCU-EXOPC leaves the room