[12:21:31] --- gurtov has joined
[12:22:34] --- jlaganie has joined
[12:34:42] --- jishac has joined
[12:40:48] <jlaganie> Meeting started?
[12:42:12] <jlaganie> I'll have a question on draft-lindqvist-hip-tcp-piggybacking-00.txt
[12:42:31] --- jeffa has joined
[12:42:39] <jlaganie> The TCP segments piggybacked on I2 and R2 aren't protected at all?
[12:42:54] <jeffa> I will now start taking notes directly into Jabber here
[12:43:04] <jeffa> (had to get 802.11a up)
[12:43:19] <jeffa> Tim S.:
[12:43:36] <jeffa> skeptical that we will build into applications managers that are aware of multiple i/fs
[12:43:48] <jeffa> SP: Bittorrent for example
[12:44:04] <jeffa> SP = Sebastien Pierrel
[12:44:13] <jeffa> Tim S.: seems to be hard problems here
[12:44:17] <jeffa> SP: don't quite see your point
[12:44:36] <jeffa> TS: Can't imagine end users wanting to tweak these knobs, so we need something automatic
[12:44:45] <jeffa> Petri Jokela:
[12:44:50] <jeffa> we made the easy part
[12:45:04] <jeffa> the difficult is common for many other wg also, shim6, monami ...
[12:45:10] --- bew has joined
[12:45:17] <jeffa> PJ: we implemented (something) ....
[12:45:25] --- miika has joined
[12:45:29] <jeffa> PJ: i know that's a very hard problem
[12:45:37] <jeffa> PJ: that cannot be solved in two years
[12:46:00] <jeffa> Shinta: question about how the locator part is handled
[12:46:15] <jeffa> (Julien: I will ask your question at the mic once Janne presents the TCP draft)
[12:46:23] <jeffa> (hopefully the network will not go down)
[12:46:32] <jeffa> Shinta: binding flow per interface?
[12:46:51] <jeffa> SP: ... (about binding)
[12:47:05] <jlaganie> (Thanks Jeff. A reply later would also be fine if the NW is tare down)
[12:47:15] --- lars has joined
[12:47:22] <jeffa> Shinta: seems a bit redundant to establish another SA for only the purpose of having multiple locators
[12:47:26] <lars> do wwe have an agenda and are the slides online?
[12:47:36] <jeffa> ?: application may not be able to establish the interfaces
[12:47:47] <jeffa> (yes there is a draft agenda....)
[12:47:51] <jeffa> (please wait for URL...)
[12:47:52] <jishac> lars: there was an agenda
[12:48:07] <jeffa> AG: how does it work, separate SAs per HIT?
[12:48:24] <jeffa> SP: for every ....
[12:48:30] <jeffa> SP: can have multiple SAs
[12:48:41] <jeffa> Miika Komu: there should be one SPI per interface in the mm draft
[12:49:01] <jeffa> SP: here we are taking multiple interfaces into account
[12:49:08] --- woj has joined
[12:49:19] * jishac has set the topic to: http://hip.piuha.net/meetings/ietf66/agenda-hip.html
[12:49:20] <jeffa> *** agenda was sent to ML, see here: https://honor.trusecure.com/pipermail/hipsec-rg/2006-July/000330.html ***
[12:49:32] <lars> jishac: thanks
[12:49:36] <jeffa> Andrei: anyone from shim6 that can comment?
[12:49:52] <jeffa> Shinta: I was also working on shim6, as far as I see there is no work on this topic
[12:50:48] <jeffa> Shinta: common need but no particular work here
[12:51:03] <jeffa> SP: should this be a research group topic?
[12:51:13] <jeffa> Andrei: take it to the list, I think Tom has a comment
[12:51:20] <jeffa> ----------- switch presentations ------------
[12:51:39] <jeffa> Topic: IP service Discovery; Presenter: Petri Jokela
[12:52:21] <jeffa> Petri http://www.ietf.org/internet-drafts/draft-jokela-hip-service-discovery-00.txt
[12:52:28] <jeffa> Slide: General
[12:53:00] --- fparent@jabber.org has joined
[12:53:01] <jeffa> Petri: this work started because we needed to locate services
[12:53:17] <jeffa> on the ML this came up that this resembles the old BOS packet (bootstrap)
[12:53:26] <jeffa> service discovery could be used in the same manner as the BOS
[12:53:33] <jeffa> Andrei: status of BOS, removed?
[12:53:46] <jeffa> Petri: Yes it is removed. nobody wanted to work on it at that time (?)
[12:54:23] <jeffa> slide: New HIP packets
[12:54:35] <jeffa> slide: New packet: Service Discovery
[12:55:01] <jeffa> ...describes active service discovery
[12:55:26] <jeffa> the packet structure in its basic form is like an I1 with a destination HIT set to zero
[12:55:41] <jeffa> the REG_INFO from the reg draft is optional
[12:55:49] <jeffa> slide: New Packet: Service Announcement
[12:56:04] <jeffa> R1 parameters may be included to speed things up
[12:56:14] <jeffa> slide: Active Service Discovery
[12:56:34] --- seb_p has joined
[12:56:35] <jeffa> ...gives example of a HIP host joining a network
[12:56:48] <jlaganie> I don't understand why service discovery needs to be integrated in the HIP protocol. Why can't we just use DNS records, or SLP?
[12:56:58] <jeffa> ...host increasing TTL
[12:57:07] <jeffa> slide: Active Service Discovery
[12:57:19] <jeffa> (Julien -- will address your question soon ...)
[12:57:28] <jeffa> s/address/ask/
[12:57:48] <jeffa> ...going through slide animation describing the process
[12:58:21] <jeffa> new diagram showing regional model
[12:58:41] <jeffa> slide: Active Service Discovery (multiple frames)
[12:58:48] <jeffa> slide: Passive Service Discovery
[12:58:55] <jeffa> the HIP node doesn't do anything actively
[12:59:28] <jeffa> slide: Passive Service Discovery
[12:59:35] <jeffa> ...describing diagram
[12:59:46] <jeffa> ...the yellow circle has a local address space
[13:00:07] <jeffa> will not go into detail about MR operation
[13:00:19] <jeffa> slide: Implementations
[13:00:31] <jeffa> Ericsson has implemented on-path, plans for regional
[13:00:34] <jeffa> slide: Future
[13:00:46] <jeffa> haven't been looking at existing protocols
[13:01:14] <jeffa> how this could be transferred to other protocols? ...
[13:01:46] --- miika has left: Lost connection
[13:02:02] <jeffa> Jeffa: asking Julien's question
[13:02:07] <jeffa> Tim S.: I have the same question
[13:02:25] <jeffa> TS: reading the draft, I see it uses site local multicast address ... deprecated ? no
[13:02:36] <jeffa> TS: I was still unable to think of a motivating example
[13:02:52] <jeffa> TS: if you can think of a unicast address ... you are looking for middleboxes, why?
[13:03:01] <jeffa> Petri: for HIP-aware middleboxes
[13:03:16] <jeffa> Petri: we need this for the MR, a network mobility scenario
[13:03:19] --- mkomu has joined
[13:03:30] <jeffa> TS: maybe you need some motivating examples in the draft, or point to one
[13:03:39] <jeffa> TS: maybe you could give an exxample now
[13:04:22] <jeffa> Bill Atwood: one of the things that I see is ICMP packet being thrown away by routers because of the security damage
[13:04:31] <jeffa> BA: but you are relying on them
[13:04:39] <jeffa> BA: have you considered that risk at all?
[13:04:47] <jeffa> PJ: we should implement some kind of timer in that case ...
[13:04:53] <jeffa> Andrei: or do it in parallel
[13:05:04] <jeffa> BA: a whole bunch of router admins throw this stuff away
[13:05:19] <jeffa> Miika: another use would be searching for a printer or something
[13:05:30] <jeffa> MK: or in the NAT draft there is a hairpin problem
[13:05:54] <jeffa> MK: using service disc. or BOS you can figure out if two nodes are in the same network
[13:06:01] <jeffa> Petri: OK we could define
[13:06:05] <jeffa> Julien are you happy?
[13:06:15] --- shep has joined
[13:06:20] <jeffa> Andrei: what is the advantage of Active vs Passive
[13:06:31] <mkomu> passive works only on path
[13:06:40] <jeffa> Petri: passive "" ""
[13:07:08] <jeffa> Andrei: with active discovery you probe off path, how do you know which places to probe? random?
[13:07:15] <jeffa> Petri: broadcast on the link
[13:07:28] <jeffa> Andrei: what about for beyond TTL 1 ?
[13:07:36] <jlaganie> I've got to think more about that.
[13:07:54] <jeffa> Petri: we use site local address
[13:08:27] <jeffa> Tim S.: I may be repeating myself, but as a group we cannot ask questions about this unless we understand
[13:08:28] <jlaganie> If midbox detection is the issue, there is MIDCOM...
[13:08:37] <jeffa> TS: what you are trying to accomplish
[13:08:44] <jeffa> Andrei: think about Windows services
[13:08:51] <jeffa> TS: why should this be a part of HIP?
[13:09:03] <jeffa> Petri: we need to look at existing protos, do we provide an advantage?
[13:09:10] <jlaganie> If Service Discovery is the issue, still SLP, DNS, or even multicast DNS
[13:09:20] <jeffa> AG: some generic protocol might not be aware of HIPs
[13:09:36] <jeffa> TS: I'm full of questions about this ...
[13:10:12] <shep> I'm full of questions about this but I cannot figure out which questions are appropriate to ask without first understanding what the motivation for this is.
[13:10:28] <jeffa> Petri: we will look at these issues
[13:10:40] <jeffa> draft-lindqvist-hip-tcp-piggybacking-00.txt: TCP piggybacking (Janne Lindqvist)
[13:10:46] <jeffa> --------- switching presentations ----------------
[13:11:19] <jeffa> Janne Lindqvist : Establishing HIP Opportunistic Mode with TCP option
[13:11:27] <jeffa> was presented in Dalls but this is about the new draft
[13:11:48] <lars> http://www.ietf.org/internet-drafts/draft-lindqvist-hip-opportunistic-01.txt
[13:11:50] <jeffa> http://www.ietf.org/internet-drafts/draft-lindqvist-hip-tcp-piggybacking-00.txt
[13:12:01] <jlaganie> General comment: I like the idea, this is a neat hack
[13:12:24] <lars> general comment: i hate the idea, this is a bad hack
[13:12:28] <jeffa> slide: Motivation for the Approach
[13:12:31] <jeffa> slide: Basic Idea
[13:12:34] --- admcd has joined
[13:12:36] <lars> :-) but since this is the rg, lookking at this is ok
[13:12:45] <shep> Hmm... I cannot find that draft in my mirror of the internet-drafts directory.
[13:13:04] <lars> shep: do you mirror from the ietf or the rfc editor? they are sometimes out of sync
[13:13:21] <jeffa> slide: Piggybacking ...
[13:13:25] <jeffa> slide: Design Choices
[13:13:43] <shep> I think I mirrored from ftp.rfc-editor.org .
[13:13:56] <shep> I'm running rsync again now.
[13:14:11] <jeffa> slide: Piggybacking
[13:14:29] <jeffa> (it's from July 11, so ...)
[13:14:33] <lars> oh wait, the "bad hack" comment goes with janne's other draft
[13:14:41] <lars> i agree with julien :-)
[13:14:43] <jeffa> slide: w/diagram
[13:15:06] <jeffa> slide: Integrity Protection
[13:15:09] <jlaganie> Lars, so the bad hack is piggybacking?
[13:15:17] <jeffa> after hallway discussions this may not be a good option at all
[13:15:28] <jeffa> slide: Relation to Opportunistic HIP w/TCP Option
[13:15:37] <jeffa> slide: Security Considerations
[13:15:52] <jeffa> slide: Alternatives ...
[13:16:00] <jeffa> result of hallway discussions ...
[13:16:11] <jlaganie> w.r.t. security, the TCP segments piggybacked on I2 and R2 aren't protected at all?
[13:16:33] <jeffa> slide: Conclusion
[13:16:51] <jeffa> comments/questions?
[13:17:01] <jeffa> Lars: what's specific about TCP about this?
[13:17:21] <jeffa> Lars: why can't you allow other transports to piggyback ... relax the MUST to a MAY and it becomes much more genreal
[13:17:27] <jeffa> Janne: true
[13:17:34] <jlaganie> Agree with Lars. Could possibly work with any protocol
[13:17:40] <jeffa> TS: ... you would have all you need at the end of the packet
[13:17:59] <jeffa> TS: you could have a magic SPI
[13:18:11] <jlaganie> Agree with Tim as well
[13:18:18] <jeffa> TS: it could work, need more details...
[13:18:21] <jlaganie> If you use a fixed SPI you can use ESP
[13:18:40] <jeffa> JL: it could work... also ESP was removed from the base draft, so there may be more that I don't know
[13:18:40] <jlaganie> And protects piggybacked packet
[13:18:56] <jeffa> TS: Bob Moskowitz originally wanted this functionality in the protocol
[13:19:07] <jeffa> TS: you don't want to add another RT
[13:19:25] <jeffa> (shep: maybe you can fill in your question above... I missed it)
[13:19:37] <jlaganie> Issue is that some IPsec people are allergic to fixed SPI , they equate them to SKIP, and SKIP was ruled out by IETF
[13:19:48] <jeffa> TS: reason to put it in the same packet ... to protect it with ESP
[13:19:54] <jeffa> Andrei: ? (missed it)
[13:20:15] <jeffa> TS: it's unfortunate ESP was removed from the base draft, b/c it makes this harder to explain
[13:20:41] <jeffa> Lars: Still suggest that you don't just focus on ESP
[13:20:50] <jeffa> LE: what if you had simult. syn
[13:21:02] <jeffa> LE: use something that protects that stuff, this would be awesome
[13:21:19] <jeffa> Andrei: any issues with maximum packet size?
[13:21:29] <jeffa> AG: it should fit but what if not
[13:21:39] <jeffa> LE: you need to know how big the data is you want to include
[13:21:43] <jeffa> LE: there are ways to do this
[13:22:09] <jeffa> JL: I need to look at not only TCP, that sounds interesting to me
[13:22:17] <jeffa> ---------- switching presentations ---------------
[13:22:33] <jeffa> representatives of each implementation, what is the status
[13:22:40] <jeffa> Miika -
[13:22:46] <jeffa> NAT support for client
[13:22:55] <jeffa> currently implementing NAT server
[13:23:11] <jeffa> it would be nice to have a planetlab testbed and provide free rendezvous service
[13:23:35] <jeffa> I have 4 CDs that contain knoppix live CD with HIP support
[13:23:39] <jeffa> welcome any feedback
[13:23:46] <jeffa> also will post this to the project web page
[13:24:52] <mkomu> jeff: done upgrading of drafts, support for IPv6, NAT support from nec
[13:25:03] <mkomu> jeff: will be included in the next version of OpenHIP
[13:25:17] <mkomu> jeff: will be working on integrating SHIM6
[13:25:28] <mkomu> andrei: does mac support already?
[13:25:35] <mkomu> jeff: yes
[13:25:42] <jeffa> (thanks Miika)
[13:25:49] <jeffa> Petri: Ericsson implementation is up to date
[13:25:57] <jeffa> PJ: but I don't know much about that
[13:26:29] <jeffa> PJ: woodtsock test server doesn't yet support the new28-bit HIT prefix
[13:26:37] --- jlaganie has left
[13:26:44] <jeffa> LE: some of these things don't seem very researchy to me anymore
[13:26:52] <jeffa> LE: if we have running code and it ineroperates
[13:27:06] <jeffa> AG: it's too researchy, b/c it runs but crashes ....
[13:27:14] <jeffa> LE: full standard immediately
[13:27:41] <jeffa> http://www.openhip.org and http://infrahip.hiit.fi
[13:28:04] <jeffa> Andrei showing IRTF wiki
[13:28:13] <lars> s/LE: full standard immediately/LE: full standard immediately :-)/
[13:28:20] <mkomu> http://www.hip4inter.net/
[13:28:55] <jeffa> Tim S: (talking about Dave Thaler's mobility comparison draft)
[13:29:04] <jeffa> Tim S: it's a really interesting draft
[13:29:15] <jeffa> TS: you should go look at this draft...
[13:29:28] <jeffa> TS: there are some "no's" in his tables
[13:29:57] <jeffa> TS: http://www.openhip.org/irtf/wiki/index.php?title=Main_Page
[13:30:04] <jeffa> (that is the IRTF wiki page)
[13:30:15] <jeffa> (oops, not TS: for the link)
[13:30:27] <jeffa> TS: look at the draft and see if HIP is being explained accurately
[13:30:29] <seb_p> what's the draft about mobility protocols comparison?
[13:30:41] <jishac> shep: can you link to that draft?
[13:30:55] <lars> draft-thaler-mobility-comparison-01.txt
[13:30:58] <jeffa> Dave Thaler: outages point is not clear ... outages is where both ends think their locator is working but they are not
[13:31:05] <jishac> lars: thanks
[13:31:34] <jeffa> DT: based on IETF *WG* it has the signalling stuff, but not discovering the failure
[13:31:47] <jeffa> TS: it's an accurate description of the scenario
[13:31:52] <lars> was a really good presentation, too, slides at IP mobility protocol comparison
[13:32:04] <jeffa> TS: an implementation might do something here too
[13:32:26] <jeffa> TS: what would the implementation do in this situation, what would you expect your implementation do?
[13:32:35] <jeffa> Andrei: double jump
[13:32:48] <jeffa> AG: Hi3 is based on an overlay network
[13:32:56] <jeffa> TS: the double jump is a different situation
[13:33:18] <jeffa> TS: both hosts have multiple locators, each AB - YZ
[13:33:29] <jeffa> TS: A - Y can communicate, then something goes wrong
[13:33:59] <jeffa> TS: failure in the middle of the network, communication from A-Z might work, but would an impl. discover this fact?
[13:34:11] <jeffa> TS: in SHIM6 they have a story for that
[13:34:44] <jeffa> Andrei: is this purely an implementation issue?
[13:34:57] <jeffa> AG: doesn't require any interoperability
[13:35:08] <jeffa> DT: what probing mechanism are you trying to use?
[13:35:14] <jeffa> s/trying/going/ ?
[13:35:20] --- admcd has left
[13:35:27] <jeffa> TS: if we don't write this don't people won't be convinced
[13:35:33] <jeffa> TS: as to how
[13:36:04] <jeffa> DT: which probing mechanism, depends what the other side can support, not interop if the other side doesn't respond
[13:36:18] <jeffa> TS: you probe with various TCP retransmissions
[13:36:31] <jeffa> DT: that's a mod of the TCP protocol -- if you spray them across different paths
[13:36:39] <jeffa> TS: I don't think this would change the protocol
[13:36:54] <jeffa> MK: mm draft gives a mechanism to probe for active locators
[13:37:08] <jeffa> MK: using the reachability test, hasn't been nailed down in the draft
[13:37:16] <jeffa> Andrei: something added to the SIMA draft later?
[13:37:28] <jeffa> Andrei: about how to probe local interfaces
[13:37:35] <jeffa> Sebastien: OK we can look at that
[13:37:43] <jeffa> Andrei: set of drafts in WG is quite small
[13:38:02] <jeffa> in RG we have had 20 or so drafts, not easy to find ... like privacy extensions, etc
[13:38:13] <jeffa> Andrei (to DT) thanks for writing the document
[13:38:16] <jeffa> it is useful
[13:38:31] <jeffa> ?: invite you to a workshop
[13:38:37] <jeffa> deadline end of this month
[13:38:56] <jeffa> it will be colocated with GLOBCOM (?) Nov 27- Dec 1 in San Francisco
[13:39:01] <jeffa> send good topics to me
[13:39:06] <jeffa> (anyone have his name?)
[13:39:20] <jeffa> Shinta: would like to add about reachability verification
[13:39:23] <lars> ? is xiaoming fu
[13:39:29] <jeffa> (thanks)
[13:39:44] <jeffa> Shinta: SHIM6 has failure detection protocol in the works
[13:40:03] <jeffa> Shinta: designed so it could be used with other protocols, like HIP
[13:40:03] --- woj has left
[13:40:10] <jeffa> Shinta: it may be useful for implementers
[13:40:18] --- woj has joined
[13:40:18] --- fparent@jabber.org has left: Logged out
[13:40:25] --- lars has left
[13:40:31] <jeffa> Andrei: thanks for coming, we are done.
[13:40:34] --- gurtov has left: Не в сети
[13:40:52] --- seb_p has left
[13:41:04] --- bew has left
[13:41:39] * jishac has set the topic to: HIPRG Meeting ( http://www.irtf.org/charter?gtype=rg&group=hip )
[13:42:01] --- jishac has left
[13:49:57] --- jeffa has left
[13:56:04] --- shep has left
[14:05:48] --- woj has left
[14:16:28] --- mkomu has left