IETF
anima
anima@jabber.ietf.org
Monday, July 20, 2015< ^ >
Room Configuration
Room Occupants

GMT+0
[06:45:13] Toerless Eckert joins the room
[06:45:57] <Toerless Eckert> good morning
[06:49:14] mcr joins the room
[06:49:21] <mcr> hi.
[06:50:07] <Toerless Eckert> good morning
[06:50:12] <Toerless Eckert> is audio working ?
[06:50:39] <Toerless Eckert> because we didn't do anything here, so i hope ietf has set it up correctly ;-)
[06:52:12] whatdafuq joins the room
[06:52:31] Terry Manderson joins the room
[06:53:20] Brian Carpenter joins the room
[06:55:05] <Toerless Eckert> anybody who is willing to say audio is working ?
[06:55:27] Max Pritikin joins the room
[06:55:28] <mcr> not sure, the links aren't filled in on the agenda, digging for URL. I thikn it is: http://ietf93streaming.dnsalias.net/ietf/ietf937.m3u
[06:56:26] Michael Richardson joins the room
[06:56:55] <Toerless Eckert> somehow i can't bring up the audio client from the URL, so can't check myself...
[06:57:05] Meetecho joins the room
[06:57:25] <mcr> okay, I see chairs at table, but audio quality is not great. Maybe you are really chatting with Behringer, and not room.
[06:57:58] <Max Pritikin> yes, they are just chatting locally
[06:58:35] <mcr> good sound.
[06:59:09] Markus Stenberg joins the room
[06:59:15] <Michael Richardson> hummmmm
[06:59:29] <Toerless Eckert> thanks
[07:01:18] Zongpeng Du joins the room
[07:01:25] <Brian Carpenter> audio OK now?
[07:02:30] joel jaeggli joins the room
[07:03:06] <Max Pritikin> I'm not sure what you all need from the "jabber scriber" let me know if i'm not providing something you want
[07:03:31] Matthieu joins the room
[07:03:42] <mcr> max, you get to relay comments from the jabber that are prefixed with mic:
[07:03:59] <Max Pritikin> mcr: thx :) will do
[07:04:16] <mcr> and it's nice if you tell us what slide things are on, but as I'm using the meetecho, it ought to work out find, and the meetecho guys sometimes insert slide numbers.
[07:04:17] <Max Pritikin> meanwhile: find agenda and materials here: https://datatracker.ietf.org/meeting/93/materials.html
[07:06:56] <Max Pritikin> sldie 3
[07:08:29] <Max Pritikin> slide 4
[07:08:33] Laurent Ciavaglia joins the room
[07:08:54] Zongpeng Du leaves the room
[07:09:24] zongpeng Du joins the room
[07:09:43] <Max Pritikin> slide 5
[07:09:48] <mcr> btw: camera is only at chairs desk, not at speaker.
[07:10:56] <Meetecho> fixed :)
[07:11:00] <Toerless Eckert> missing slides for today now uploaded
[07:11:15] <Toerless Eckert> let me know if there is an issue to access them.
[07:11:17] <Max Pritikin> slide 6
[07:12:33] <Max Pritikin> slide 7, fyi
[07:12:54] zongpeng Du leaves the room
[07:13:37] <Max Pritikin> slide8
[07:15:05] <Max Pritikin> 9
[07:15:35] <Laurent Ciavaglia> hello. slide is not displaying on the meetecho screen...
[07:15:36] TAHIR JAMEEL joins the room
[07:15:54] <Max Pritikin> just this slide or none of the slides?
[07:15:56] <Michael Richardson> I see the slides just fine in meetecho.
[07:16:17] <Laurent Ciavaglia> should be on my side. slides were ok until 1 or 2 minutes ago
[07:16:37] <Max Pritikin> laurent: you can find all the slides at this link: https://datatracker.ietf.org/meeting/93/materials.html
[07:16:47] <Laurent Ciavaglia> thanks ;)
[07:16:48] <Michael Richardson> slide 10 now.
[07:16:58] <Max Pritikin> go down to anima, "Slides from ANIMA Signaling Design Team <https://www.ietf.org/proceedings/93/slides/slides-93-anima-1.pdf>"
[07:17:38] <Max Pritikin> 11
[07:17:38] Laurent Ciavaglia leaves the room
[07:18:03] Laurent Ciavaglia joins the room
[07:19:04] <Max Pritikin> 12
[07:19:42] whatdafuq leaves the room
[07:20:57] joel jaeggli leaves the room
[07:21:23] <Max Pritikin> slide 13
[07:23:00] whatdafuq joins the room
[07:23:28] <Max Pritikin> slide14
[07:25:28] <Max Pritikin> slide15
[07:29:05] aretana joins the room
[07:30:00] <Michael Richardson> explore the CBOR approach.
[07:30:05] <Michael Richardson> hummmmm
[07:30:44] <Max Pritikin> not voicing your contribution in room because everybody else agreed
[07:32:24] Brian Carpenter leaves the room
[07:32:30] Brian Carpenter joins the room
[07:33:06] whatdafuq leaves the room: Replaced by new connection
[07:33:08] whatdafuq joins the room
[07:33:13] Toerless Eckert leaves the room
[07:34:59] Toerless Eckert joins the room
[07:35:06] <Toerless Eckert> slide 5
[07:38:49] <Toerless Eckert> slide 6
[07:40:02] <Toerless Eckert> slide
[07:40:04] <Toerless Eckert> slide 7
[07:41:04] <Toerless Eckert> slide 8
[07:43:54] Simon Romano joins the room
[07:46:00] <Toerless Eckert> slide 11
[07:47:36] Max Pritikin leaves the room
[07:48:40] <Michael Richardson> could the chairs fullscreen the slides, as they are pretty small.
[07:48:46] joel jaeggli joins the room
[07:52:56] <Toerless Eckert> slide 12
[07:54:39] <Toerless Eckert> slide 13
[08:03:19] <mcr> mic: and this visit to the vendor after failing to find a local resource is also very consistent with the netconf-zerotouch documents.
[08:05:23] Max Pritikin joins the room
[08:08:03] <Max Pritikin> humm for signaling doc. no opposition.
[08:08:12] <Laurent Ciavaglia> hummmmm
[08:08:21] <Max Pritikin> humm for bootstrapping doc. no opposition.
[08:08:35] <Max Pritikin> [laurent, i'm assuming you came in positive? timing was unclear]
[08:08:52] <Laurent Ciavaglia> yes in favor of adoption for bootstrapping draft
[08:08:54] <Max Pritikin> thx
[08:09:37] aretana leaves the room
[08:10:07] <Toerless Eckert> slide 3 of ACP draft
[08:10:48] <Max Pritikin> ACP slides https://www.ietf.org/proceedings/93/slides/slides-93-anima-4.pdf
[08:12:38] hosnieh joins the room
[08:14:51] <mcr> maybe we have our own version of: https://en.wikipedia.org/wiki/Omnipotence_paradox <https://en.wikipedia.org/wiki/Omnipotence_paradox>: Can ACP make a network so complex that even ACP can't guarantee it won't shoot itself in the foot.
[08:15:25] hosnieh leaves the room
[08:15:33] Hosnieh Rafiee joins the room
[08:15:49] Hosnieh Rafiee leaves the room
[08:16:46] aretana joins the room
[08:21:29] <mcr> (not for mic) We have a dozen different ways to tunnel IPv4 over IPv6, so as long as the ACP can transport v6 packets, some stateless can give one 4to6 gives you IPv4.
[08:22:32] <Brian Carpenter> @mcr: Sure, Make the legacy user suffer ;-)
[08:23:25] <Max Pritikin> legacy users, by definition, enjoy suffering. (as do first adopters)
[08:25:05] <mcr> who was last speaker about the weekend?
[08:25:22] <Brian Carpenter> Juliusz
[08:25:49] <Max Pritikin> at mic: henning rogge
[08:26:38] <Brian Carpenter> To be complete, Juliusz Chroboczek
[08:29:21] <mcr> I don't understand the Data Plane based ACP.... does it mean that GRASP has to provide an application layer (objecting) routing function?
[08:30:19] <mcr> er, Object!
[08:30:24] <mcr> er, object routing function.
[08:31:11] <Max Pritikin> sorry. i just read. :)
[08:31:25] <Terry Manderson> @mcr: Stop objecting ;)
[08:33:05] <TAHIR JAMEEL> hummmmm
[08:34:41] <mcr> I still don't know if the Data-Plane ACP will layer-3 forward GRASP packets, or if GRASP will forward things hop-by-hop.
[08:35:14] <Laurent Ciavaglia> hummmmm
[08:35:24] <Laurent Ciavaglia> in favor of adoption
[08:35:45] <mcr> I am for adopting, but I am against including the Data-Plane ACP at this time.
[08:38:14] <mcr> but, the data-plane doesn't exist yet.
[08:39:41] <mcr> But, I said GRASP on purpose, it's the protocol that will get the addresses to make the data plane to work, so before that exists, how does things work?
[08:40:28] <Brian Carpenter> discovery starts link-local, but off-link really needs the ACP
[08:41:15] <mcr> There is insufficient detail in the document to have any clue about what this data-plane ACP is about....  
[08:41:31] <Toerless Eckert> mcr: i think our assumption is that GRASP is not needed to get ACP addressing set up
[08:41:53] <Toerless Eckert> mcr: but i think it's an importand discussion to define these interactions so we do not create circular dependencies.
[08:50:37] <Laurent Ciavaglia> In the ACP presentation, the ACP is tagged "autonomic".
[08:51:15] <Laurent Ciavaglia> Michael just mentioned that the infrastructure (incl. the ACP) are not autonomic function... is there an inconsistency here?
[08:51:15] aretana leaves the room
[08:51:38] Toerless Eckert leaves the room
[08:51:55] Simon Romano leaves the room
[08:51:57] Terry Manderson leaves the room: Replaced by new connection
[08:52:02] Markus Stenberg leaves the room
[08:52:05] joel jaeggli leaves the room
[08:52:07] Terry Manderson joins the room
[08:52:09] Max Pritikin leaves the room
[08:52:10] joel jaeggli joins the room
[08:52:10] Matthieu leaves the room: Replaced by new connection
[08:52:11] Matthieu joins the room
[08:52:36] Toerless Eckert joins the room
[08:53:33] Matthieu leaves the room: Replaced by new connection
[08:53:34] Matthieu joins the room
[08:53:42] Max Pritikin joins the room
[08:53:53] joel jaeggli leaves the room
[08:54:02] joel jaeggli joins the room
[08:54:13] <mcr> Just remember that light *bulbs* are not particularly energy constrained.  It's the possible future light switches.
[08:54:25] Brian Carpenter leaves the room
[08:54:29] Brian Carpenter joins the room
[08:54:59] Terry Manderson leaves the room: Replaced by new connection
[08:55:14] Terry Manderson joins the room
[08:55:38] Max Pritikin leaves the room
[08:55:52] <Toerless Eckert> mcr: did you see my answer on ACP without virtual separation ?
[08:55:58] <Toerless Eckert> mcr: i had wifi/vpn issues
[08:56:15] <mcr> Toerless Eckert: mcr: i think our assumption is that GRASP is not needed to get ACP addressing set up
[08:56:15] <Brian Carpenter> @remote people - wifi has been bad in the room for a few minutes
[08:56:38] Balazs joins the room
[08:56:39] Max Pritikin joins the room
[08:56:54] <Toerless Eckert> mcr: one instance of GRASP is needed IMHO to help build ACP, but the main instance of GRASP would leverage ACP.
[08:56:57] Terry Manderson leaves the room
[08:57:01] <Max Pritikin> [my apologies for dropping in/out .. the network in this room is flaky]
[08:57:39] Toerless Eckert leaves the room
[08:59:38] Max Pritikin leaves the room
[08:59:50] Max Pritikin joins the room
[09:01:42] Balazs leaves the room
[09:02:10] Brian Carpenter joins the room
[09:03:22] <mcr> comment: 802.11AR is rather vague about what kind of name is in the IDevID.  Basically just punts back to IETF documents.  The discussion needs to bifurcate into what are initial names, and what kind (if any) of LDevID names are useful.
[09:04:35] <Max Pritikin> personal opinion: 802.1AR punts to the device vendor for the "serial number" name. We could carry that forward into the LDevID although collisions are technically possible.
[09:04:43] <Max Pritikin> thus the random name discussion
[09:05:43] <mcr> do we depend upon the vendor to put a unique number into the 1AR "MIC"?
[09:05:54] <mcr> if so, we might need to state at least that.
[09:05:55] <Max Pritikin> unique to the vendor but not globally unique
[09:06:04] <mcr> yes, unique to the Issuer.
[09:06:06] aretana joins the room
[09:06:44] aretana leaves the room
[09:06:51] whatdafuq leaves the room
[09:07:33] Matthieu leaves the room: Replaced by new connection
[09:07:33] Matthieu joins the room
[09:08:03] <Max Pritikin> yes, that uniqueness is required by 802.1AR/PKIX. And for LDevID's the same requirement exists. But 802.1AR doesn't have any position on this discussion here: does ANIMA care about the naming structure or what
[09:10:25] Brian Carpenter leaves the room
[09:13:14] Toerless Eckert joins the room
[09:13:24] <Toerless Eckert> $#*& wifi
[09:14:02] <Max Pritikin> ietf-v6ONLY appears more stable
[09:14:26] Brian Carpenter joins the room
[09:14:55] Brian Carpenter leaves the room
[09:16:04] whatdafuq joins the room
[09:16:23] joel jaeggli joins the room
[09:16:23] joel jaeggli leaves the room
[09:16:33] joel jaeggli joins the room
[09:17:26] joel jaeggli leaves the room
[09:23:45] <mcr> I prefer each autonomic entity just ask for a number of DeviceIDs, and leave it to the registrar to think about aggregating addresses. The VM instances on the blades might not even get booted until the Intents are distributed and some node figures out that the network needs a FOOBAR ASA, and it happens to have a FOOBAR VM that it can run.
[09:24:55] Matthieu leaves the room
[09:24:56] <Toerless Eckert> mcr: my addressing suggestion was primarily for  easily aggregated routing
[09:25:53] <Toerless Eckert> mcr: an ASA in a VM of an autonomic node may get as you suggest its address dynamically, but the autonomic node could pick it from the prefix it has. In result the node only needs to put the prefix route into the routing protocol. As opposed to one /128 for every ASA running on it
[09:26:02] <mcr> My thought is that the knowing how big the N to subtract from the 64 in advance is hard, and the registrar can easily give out DeviceIDs that increment by 16 or something if it likes.
[09:26:42] <Toerless Eckert> mcr: but the node should know what prefix it did get ?
[09:26:46] <mcr> Toerless, if doubling the number of /128s in the routing protocol kills us, then we are dead already.
[09:27:07] <Toerless Eckert> mcr: well, i was thinking i may have 20 ASA per node.
[09:29:02] <mcr> And I'm thinking that you actually won't know how many (and which) ASA to start until the Intents are distributing, so you'll always have to ask for a chunk of address space, and your network evolve over time, anyway...
[09:30:31] <mcr> thanks!
[09:30:44] <Laurent Ciavaglia> I agree with Michael R. and also the AF are not bound to a single node...
[09:31:09] <Laurent Ciavaglia> Seems more feasible to have the address (poll) assigned to the AF rather than to the ANode
[09:31:11] Max Pritikin leaves the room
[09:31:33] whatdafuq leaves the room
[09:31:43] <Laurent Ciavaglia> I mean address pool (not oll)
[09:31:58] <mcr> it's also much easier to run multiple ASA on a single OS if each one can be a separate process, and don't have to coordinate processing the TLV/CBOR with each other.
[09:32:20] Meetecho leaves the room
[09:32:30] Toerless Eckert leaves the room
[09:32:40] <Laurent Ciavaglia> I think ti is up for discussion on the mailing list (buit very interesting indeed)
[09:32:44] <mcr> So I prefer to hand out address per AF, and if some are in VMs are some are not, who cares....  And the VMs might even (live) migrate destroying all the aggregated blocks anyway.
[09:32:55] Michael Richardson leaves the room
[09:33:37] <Laurent Ciavaglia> Agreed, let's conitnue on the mailing list  please (I will disconnect from the meetecho stream).
[09:33:40] <Laurent Ciavaglia> thanks
[09:33:43] <Laurent Ciavaglia> bye
[09:34:25] Laurent Ciavaglia leaves the room
[09:38:22] joel jaeggli leaves the room
[09:43:25] Brian Carpenter leaves the room
[09:51:14] Max Pritikin joins the room
[09:52:07] Max Pritikin leaves the room
[09:55:19] joel jaeggli joins the room
[09:59:22] joel jaeggli leaves the room
[10:06:41] Simon Romano joins the room
[10:22:14] Simon Romano joins the room
[10:22:25] Simon Romano leaves the room
[10:39:55] Simon Romano leaves the room
[10:44:32] Max Pritikin joins the room
[10:44:33] joel jaeggli joins the room
[10:47:00] Max Pritikin leaves the room
[10:48:40] joel jaeggli leaves the room
[10:50:44] Simon Romano joins the room
[10:53:45] Simon Romano leaves the room
[11:04:04] Brian Carpenter joins the room
[11:10:41] Brian Carpenter joins the room
[11:10:56] Brian Carpenter leaves the room
[11:15:00] Max Pritikin joins the room
[11:16:38] aretana joins the room
[11:16:56] Max Pritikin leaves the room
[11:17:27] Max Pritikin joins the room
[11:18:40] Brian Carpenter joins the room
[11:19:24] Max Pritikin joins the room
[11:19:46] Max Pritikin leaves the room
[11:23:56] Brian Carpenter leaves the room
[11:24:07] Brian Carpenter joins the room
[11:27:52] Toerless Eckert joins the room
[11:28:41] Max Pritikin leaves the room
[11:33:26] Brian Carpenter leaves the room
[11:39:12] Max Pritikin joins the room
[11:40:10] aretana leaves the room
[11:42:26] Brian Carpenter leaves the room
[11:44:07] aretana joins the room
[11:45:13] Max Pritikin leaves the room
[11:48:49] Matthieu joins the room
[11:49:53] Brian Carpenter joins the room
[11:52:01] Max Pritikin joins the room
[11:53:26] Matthieu leaves the room
[11:55:18] Brian Carpenter leaves the room: offline
[12:00:19] Max Pritikin leaves the room
[12:00:42] Matthieu joins the room
[12:01:03] Matthieu leaves the room
[12:28:42] aretana leaves the room
[12:30:46] Toerless Eckert leaves the room
[12:31:16] Toerless Eckert joins the room
[12:33:35] Toerless Eckert leaves the room
[13:11:29] Max Pritikin joins the room
[13:23:23] aretana joins the room
[13:33:25] Toerless Eckert joins the room
[13:34:06] Matthieu joins the room
[13:35:53] Matthieu leaves the room
[13:40:47] Max Pritikin leaves the room
[13:50:44] Max Pritikin joins the room
[13:50:52] Toerless Eckert leaves the room
[13:57:35] Toerless Eckert joins the room
[14:11:52] aretana leaves the room
[14:12:27] aretana joins the room
[14:15:00] Toerless Eckert leaves the room
[15:20:22] aretana leaves the room
[15:30:59] Max Pritikin leaves the room
[15:35:10] Max Pritikin joins the room
[15:38:14] Max Pritikin leaves the room
[15:39:45] Toerless Eckert joins the room
[15:45:09] Max Pritikin joins the room
[15:46:32] mcr leaves the room
[15:49:03] aretana joins the room
[15:58:07] Toerless Eckert leaves the room
[16:15:04] Robert Cragie joins the room
[16:15:11] Robert Cragie leaves the room
[16:36:51] Toerless Eckert joins the room
[16:39:07] aretana leaves the room
[16:40:46] Toerless Eckert leaves the room
[16:57:17] aretana joins the room
[16:58:31] aretana leaves the room
[16:59:40] Max Pritikin leaves the room
[17:06:11] Max Pritikin joins the room
[17:12:50] Max Pritikin leaves the room
Powered by ejabberd Powered by Erlang Valid XHTML 1.0 Transitional Valid CSS!