[13:04:30] --- OleTroean has joined
[13:05:12] --- Glenn Parsons has joined
[13:05:32] --- Glenn Parsons has left
[13:05:40] --- OleTroean has left
[13:05:49] --- amalis has joined
[13:05:56] --- amalis has left
[13:06:06] --- Glenn Parsons has joined
[13:06:07] --- OleTroean has joined
[13:06:25] --- yjs has joined
[13:06:43] --- jlcjohn has joined
[13:07:12] --- bkhabs@jabber.org has joined
[13:09:38] --- yushun has joined
[13:15:05] --- Bill has joined
[13:16:24] <yjs> Milestones - completed base protocol spec as WG document accept problem statement and applicability as WG work item start work with routing area WGs on TRILL extensions by July
[13:17:32] <yjs> one year from now work should be completed and will need to recharter
[13:17:44] <yjs> Problem and Applicability Statement (draft-ietf-trill-prob-00.txt) presented by Joe Touch
[13:17:58] <yjs> changes: diagram cleanup, scale estimates (intended NOT to increase scale)
[13:18:09] <yjs> which version of Ethernet is baseline for Trill? 802.1D - yes W - yes (transparently)
[13:18:19] <yjs> delivery requirements in-order and no-duplication required MSTP regions as a way to limit STP scale/reconfig 802.1Q MSTP ?
[13:18:54] <yjs> which addresses are NOT forwarded? BPDU frames, Pause frames ?
[13:19:13] <yjs> added risks to TRANSPARENT
[13:19:40] <yjs> Next step - summary of problem statement and requirements
[13:19:48] <yjs> Address remaining notes in doc
[13:19:59] <yjs> target - final revision by end of July
[13:20:13] <yjs> LC early August and then forward to IESG
[13:20:32] <yjs> mailing list has been silent - please discuss on list.
[13:20:40] <yjs> Is this document useful?
[13:20:49] <yjs> (Silence)
[13:20:59] <yjs> Chair: how many have read doc?
[13:21:03] <yjs> Answer: 7-8
[13:21:15] <yjs> Doe any have an opinion as to usefullness of doc?
[13:21:30] <yjs> (Silence) bring to list.
[13:22:04] <yjs> Joe: such lack of feedback that hard to know what to do.
[13:22:44] <yjs> Andy Malis: I read and saw a history of Ethernet, but didn't understand the applications for TRILL.
[13:23:07] <yjs> Joe: Not sure, except to say when STP is a problem (in draft)
[13:23:40] <yjs> Once again, this is another statement of the fact that we need a good statement at the front of the doc.
[13:23:56] <yjs> NEXT: The Architecture of an RBridge Solution to TRILL
[13:24:06] <yjs> raft-gray-trill-rbridge-arch-01.txt
[13:25:07] <yjs> changes to draft: remove 2119 standards language from informational draft
[13:25:23] <yjs> intro has been expanded, definitions clarified
[13:25:45] <yjs> removed options and aligned with protocol spec, removed references to "campus"
[13:25:53] <yjs> clarifying examples
[13:26:06] <yjs> Intro includes text on "strategy"
[13:27:34] <yjs> Definitions have been changed. Campus replaced with "CRED", designated RBridge fixed, etc.
[13:28:26] <yjs> Shows very complex and colorful slide
[13:29:22] <yjs> Is this a "campus"? No-one considers a campus to extend beyond the "domain" but does it include the routers?
[13:29:35] <yjs> Now CRED has all cooperating RBridges.
[13:30:18] <yjs> When there are several paths between RBridges, STP would shut down, but this would lose BW
[13:30:42] <yjs> role of desginated RBridge clarified. There is always a designated RBridge.
[13:31:12] <yjs> Still need to discuss - need "pseudo-node" use ? is this too detailed for architectures?
[13:31:39] <yjs> Should we remove distinctins between hubs, switches and bridges?
[13:31:57] <yjs> Need a better security statement, even though not a normative doc.
[13:32:26] --- washad has joined
[13:32:29] <yjs> (But don't want too much to be configured)
[13:32:40] --- washad has left
[13:33:05] --- Glenn Parsons has left
[13:33:34] <yjs> Stewart: the wiring closet problem. is this still an issue?
[13:34:32] <yjs> Eric: there was a discussion
[13:34:54] --- touch has joined
[13:34:58] <yjs> Stewart: IEEE people have said that RBridge needs to address, otherwise inferior to existing implementation
[13:35:29] <yjs> Eric: need to discuss how to solve.
[13:35:44] <yjs> Radia: Maybe not solvable. Eric: that would be nice to know too.
[13:35:52] <yjs> Stewart: Need to admit if can't be solved.
[13:36:30] <yjs> Eric: need more comments
[13:36:38] <yjs> accept as WG doc?
[13:36:45] --- ric@jabber.org.au has joined
[13:37:12] --- Glenn Parsons has joined
[13:37:13] <yjs> Chair: how many have read?
[13:37:20] <yjs> (very few)
[13:37:48] <yjs> Of those who have read - how many for and against? (2 yeses)
[13:38:18] --- cabo has joined
[13:38:31] <yjs> Jon opposes just for it to be 2:1 rather than 2:0
[13:38:41] <yjs> next- Base Protocol Specification
[13:38:46] <yjs> Radia presenting
[13:39:04] <yjs> a few minor things left to do
[13:40:03] <yjs> Stewart suggested using MPLS-like header (4 bytes) rather than 6 byte header originally proposed
[13:40:15] <yjs> for this we need to use nicknames
[13:41:16] <yjs> Also there are 3 differnt kinds of IS-IS packets (L3 IS-IS, RBridge for core instance, desginated RBridges per VLAN)
[13:42:06] <yjs> first and third are forwarded by RBridges, second always consumed
[13:42:50] --- becarpenter has joined
[13:43:17] <yjs> encap format: ETH header, hop count, original packet (with L2 header)
[13:43:38] <yjs> shim header has 19-bit label + 1-bit specifying if ingress or egress RBridge
[13:44:14] <yjs> MPLS header has EXP and S bits, but never need S bit as hierarchy shouldn't be needed
[13:45:09] <yjs> Shahram: Is there a field for OAM packets?
[13:45:59] <yjs> Radia: can use BFD if needed
[13:46:34] <yjs> Dino: Is there a new Ethertype?
[13:46:42] <yjs> Radia: yes need a new one.
[13:47:31] --- nm has joined
[13:47:31] --- yjs has left: Lost connection
[13:48:01] --- yjs has joined
[13:49:34] <yjs> Radia: should we reserve fields in MPLS label (e.g. priority, S-bit) or define them?
[13:49:58] <yjs> Radia: we probably do not want to use Ethertype for MPLS
[13:51:44] <yjs> Yaakov S: IETF asked ITU NOT to use a separate Ethertype for T-MPLS
[13:52:17] <yjs> Mark T: probably want to keep different, although this would make less reason to insist on same shim header
[13:53:17] <yjs> Radia: regular L3 IS-IS packet just like normal traffic (what about OSPF?)
[13:53:47] <yjs> Dino: IS-IS needle in a haystack but may need prioritization.
[13:53:58] <yjs> Radia: good reason to use priority field
[13:54:07] <yjs> but don't need to prune the flood.
[13:54:50] <yjs> per VLAN packets: endnodes and multicast receivers for IGMP-joined groups
[13:55:16] <yjs> decided to differentiate, so need 3 different multicast addresses
[13:55:43] <yjs> 1) ordinary flooded) 2) core IS-IS 3)
[13:57:01] <yjs> shows format (for core instance payload is entire IS-IS packet as emitted)
[13:57:46] <yjs> correction: core instance is generated by RBridges, so there is an IS-IS packet without Ethernet header
[13:58:50] <yjs> Per VLAN: also don't need inner layer 2 header, but need VLAN tag (+ align to 16 bits)
[13:59:04] <yjs> outer header is for classical bridges to function correctly
[13:59:55] <yjs> Asks if this is OK
[14:00:34] <yjs> New algorhyme written by Radia's son (written in 1 hour)
[14:02:27] <yjs> Radia suggests that the poem should be the abstract to the spec
[14:02:51] <yjs> and furthermore, that all RFCs should have rhyming abstracts
[14:03:26] <yjs> Chair: need to request multicast address from IEEE registration authority
[14:03:58] <yjs> should we request a small block?
[14:04:20] --- HP has joined
[14:04:21] --- psavola has joined
[14:05:12] <yjs> Donald Eastlake plans to meet (informally) with 802.1 chair to discuss 802.11s multicast addresses
[14:05:30] <yjs> he will bring up TRILL requirements
[14:09:03] <yjs> Dino: IEEE gave a block of 256 to IETF for IPv6 multicast
[14:10:15] <yjs> Also need an Ethertype.
[14:10:43] <yjs> last item on agenda (ahead of schedule) : TRILL Routing Requirements in Support of Rbridges
[14:10:54] <yjs> eric gray
[14:11:06] <yjs> There have been changes after detailed view.
[14:11:21] --- FDupont has joined
[14:11:33] <yjs> clarified purpose of document and importance of multicast
[14:11:42] <yjs> reduced size of sections 3 and 4.2
[14:11:48] <yjs> security section somewhat expanded
[14:12:10] <yjs> Purpose - why we want a routing protocol for multicast?
[14:12:29] <yjs> Still need to discuss - STP termination (need to clarify or remove)
[14:12:47] <yjs> Need to number requirements? (somewhat asked for this)
[14:13:18] --- FDupont has left
[14:13:21] <yjs> Obtain address prefix for IS-IS (in order to make it configuration-free)
[14:13:46] <yjs> Radia: Area address needs to be zero. Eric: may be reserved.
[14:14:04] <yjs> Dino: need to specify size of address (6+something)
[14:14:13] --- nm has left: Replaced by new connection
[14:14:13] --- nm has joined
[14:14:13] --- nm has left
[14:15:35] <yjs> Dino: no longer a body that does NSAP address allocation.
[14:15:50] <yjs> Radia: make sure that doesn't clash with something else
[14:16:03] <yjs> Dino: doesn't matter since different transport
[14:16:33] <yjs> Eric: need one more round of editing - please send comments
[14:16:41] <yjs> Can we accept as a WG document?
[14:17:40] <yjs> Chair : How many have read the draft? 3-4
[14:17:42] <ric@jabber.org.au> one hand
[14:18:14] <yjs> Chair said "three or four" under his breath, and I believed him
[14:18:31] <yjs> Chair: anyone want to bring up anything else?
[14:18:46] <yjs> Mark: lot of people in room - please participate.
[14:20:18] <yjs> Jon: How is multicast traffic handled?
[14:20:55] <yjs> Radia: multicast traffic forwarded to all links which have multicast RBridges or addresses that have joined
[14:21:45] <yjs> Dino: haven't seen anyone implementing
[14:22:08] <yjs> Chair: yes please tell mailing list if implement - otherwise this will all die
[14:22:27] <yjs> Dino: have you polled the list?
[14:22:43] --- touch has left
[14:22:45] <yjs> Chair: not explicitly, but have heard large companies mention customer demand
[14:22:56] <yjs> DONE
[14:23:01] --- HP has left
[14:23:17] --- ric@jabber.org.au has left: Logged out
[14:23:20] --- bkhabs@jabber.org has left
[14:24:11] --- yjs has left
[14:31:05] --- Bill has left
[14:31:27] --- OleTroean has left
[14:31:31] --- touch has joined
[14:35:20] --- psavola has left
[14:40:31] --- touch has left
[14:46:48] --- becarpenter has left
[14:57:00] --- Glenn Parsons has left
[15:04:10] --- cabo has left: Logged out
[15:13:31] --- yushun has left
[15:20:27] --- jlcjohn has left
[19:00:30] --- LOGGING STARTED