IETF-86 Routing Area Open Meeting
=================================

http://rtg.ietf.org 
ADs: Stewart Bryant, Adrian Farrel

Scribe: Deborah Brungard 

Adrian opened the meeting, reviewed slides, IPR policy. Prompted by discussions at the Plenary, we will be looking at having
mentors for assisting newcomers, possibly a help mail list, and a buddy scheme for mutual support. Would like a diverse
pool for mentoring and most likely not chairs, intent is to involve others. Will contact folks and also folks can
volunteer. Welcome ideas.

Jeff: Why not have more than 1:1 e.g. use jabber?
Adrian: Good idea, would have to actively promote.
Lou: Maybe instead of 1:1, as may not work out timewise, do by mail list (mentors subscribe, and anyone available can
answer questions).
Stewart: Seems to be a good idea, a mailing list, but some don't realize to signup, so 1:1 also good.
Jeff: May scare off (using a mailing list).
Acee: I wouldn't be able to do this (time wise for me difficult).
Adrian: First thing, I want you to do is be a chair. Hoping we can use this to expose others from area, so not really for chairs.
In the plenary, talked about recognition,and see this as one way to do that. Exactly how will do, need to work out. A wiki approach
does seem good to organze.
Michael R.:Don't see as really helpful for new people, see more this is for us. When sec registers a new attendee, let us know,
and then we can designate a greeter.
Ross: Should invite newcomer and their menter to the newcomers reception, and menter can introduce.
Adrian: Ok, so we shouldn't solve for IETF, should do for our area. And remember, a newcomer is not nessary new to IETF, but new to
coming to a RA meeting. I have about 6 people volunteered and will float more on the list. Also thinking of a buddy system, someone
to talk to, with similar background.
 

Administriva and Area Status
----------------------------
 

BFD (Jeff) - A few documents active. BFD over LAG, IEEE seems to be happy with. Had interop with Juniper and Cisco at BT's
lab. Will do with Alcatel-Lucent soon. So will be updating.
Alcatel-Lucent has a partial implementation of P2MP draft. MIBs not complete yet. Others (Crypto) a bit stalled waiting for
implementations.
CCAMP (Lou) - Meeting twice. 3 recent RFCs. One a mib - took very long time. One incoming liaison from BBF. One Errata.
Friday meeting all day-discussing with ITU's SG15 Q6 representatives. Topics related to Q6. They will be staying on for IEEE.
So taking advantage of that.
Forces (Adrian did) - Met, working on recharter. 5 items made it, short horizon. One item didn't make list, will continue as
individual work. Will go to IESG in short time.
IDR (Sue)- Short meeting. Link-state and TE info using BGP discussion. Low bits in attribute flag discussion. Two documents
at RFC editor, 1 WG LC, and 1 adoption. Need implementations to progress documents. Let us know.
I2RS (Alia) - Met this morning, alot of work to get done, tight milestones. Looking at functionality, avoiding kitchen sink.
Getting real discussion going on mailing list.
ISIS (Chris) - Four new drafts presented. Flooding scope PDUs, ISIS TE Metric extensions, ISIS support for unidirectional links,
ISIS Homenet and Data Centers. 
Adrian: how is the interaction with Homenet going? Chris - very new yet.
Karp (Joel) - Moving very well on charter. Have documents in front of ADs. Presentations on charter deliverables. Ready for last
call. Acee told us how they are progressing in OSPF our related drafts. Don't expect to recharter, should be able to disband.
L2VPN (Gilles) - 20 some Working Group drafts. 8 have been Last Called. Working on cleaning up. Not so many new drafts coming in.
More nvo3 space, maybe came to us as not in nvo3 charter.
L3VPN (Thomas) - Since last IETF we had 2 drafts adopted relating to multicast VPN, one draft soon to be pushed to IESG, and our
milestones were updated. Our session is this afternoon and our agenda includes items on multicast VPN and new proposals describing
how l3vpn techniques can be applied in and at the edge of datacenters.
Manet (Joe) - Some improvements to previous IESG submitted on-going - few in AD review, one in IESG processing. MIB document in
progress, need to clear discusses. Two WG last calls just finished. Few documents expired, looking to revive.
MPLS (Ross) - Three new rfcs. Starting a gap analysis, what is needed to deploy MPLS in a pure IPv6 environment. Some issues with
dealing with hashes with special use labels. Had put together a MPLS reveiw team when had a large queue of documents, not so many
now in queue. Have some ITU-IETF reviews of ITU documents going on.
Adrian - On the rview team, if others interested, contact MPLS chairs on how did/how worked out. 
Ross - Also new policy, we make sure authors say no new IPRs need disclosing, before taking to IESG. Send to whole Working
Groupp, so anyone knows can disclose.
NVO3 (Stewart) - Met. No RFCs yet, problem statement and framework drafts completed WG Last Call, some comments to address.
Met on Monday. Topics included architecture design team feedback, updates on on-going documents, use cases, and gap analysis.
Protocol work will be done in appropriate Working Group. Looking to recharter/disband.
OSPF (Acee) - Didn't meet in Atlanta. Four new RFCs. Two implemented first, then drafts written, so pretty easy for process. One
informational ready for Last Call (IPv4-IPv6 translation). Some stuff close to last call - security related to karp. Ongoing:
OSPFv3 Autoconfiguration. Proposed Work: Asymmetric Hello  For HA scenarios and auto-config, OSPFv3 LSA Extension (revived work),
Source/Destination/Flow Routing, Scaling: Topology Transparent Zone.
PCE (Julien) - One new RFC, focused on optical. One document just passed Last Call. Expect others to follow soon. Discussion on
stateful PCE. Many drafts on it. Will discuss further on list. Will propose recharter soon, working on.
PIM (Stig) Proposing to transition one to Internet standard. Will probably adopt draft and quickly take to Last Call. Took
on igmp work, variety of extensions being discussed, in regards to mobility.
PWE3 (Stewart) - One new RFC. Met tuesday. Topics: RFC4447bis, PW Endpoint protection, TRILL PW, PW availability signaling.
Roll (Michael) - Met on Tuesday. RPL-P2P/measurement and terminology at IESG. Expect security-threads to hit IESG within 2-3 weeks.
mcast-trickle to be called mickle, will get WG LC. Adopted 2 applicability statements as WG
documents, may remove via recharter the 3rd item as no support for standardizing now.
RTGWG (Alvaro) - Two documents ready for last call. On agenda, first hour on normal business, second hour on power aware networking.
As part of charter taking on new topics, problems, one solution. Good feedback in room on interest, but still considered
experimental work. Don't think should take on in IETF. Will ask on the list.
SIDR (Sandy) - Sent late a slide. Two meetings. At last minute, ended up with filling time of both. Two new RFCs. 3 in RFC Ed Queue.
5 drafts about ready for publication (past WG LC). About a dozen others. Some discussion on IPR recently filed at monday meeting.
Extensive discussion on APNIC trust anchor design. Multiple suggestions on how to model performance. From an operator deploying,
had a report of problems deploying. Plan an interim meeting.


Trust Routers (Margaret)
------------------------------------------
Adrian - A good way to get my attention in IETF is to publish a draft with router in the title. Often, it's about concepts of routers,
used in other ways. Today, have Margaret to talk about trust routers. Hope to get you excited so you can add your experience.

(Margaret presented slides)

ABFAB trust router almost done so looking at new work, motivated by that. Similar to federations of adfab. Problem statement document.
Solution document. Like routing, not IP, it's information distribution. It would be great to get your help vs. our limited knowledge.

curtis - Want topology help - look to ospf. For admin attributes- look at mpls lsp.
Acee - Probably need to take a step back on explaining the requirements and I didn't get it all. Especially the policy if 1:1, 1:n, m:n for
these various identities, best way to identify what models can help.
Margaret - Our framework document has that - we can discuss that.
Adrian - Thanks for sharing. As you were talking, I was writing down - how big a problem? Do you have a feeling if successful, how big?
Margaret - Federations are limited by scaling properties - so 1000s, not ten 1000s. Could over time, if use for banking, comercial use.
Won't rivel Internet scale.
Adrian - We'll never knew how big the Internet will be.
Margaret - Don't think will scale to BGP size, though would be good if good.
Adrian - We don't want to boil the ocean, if don't need. Is there dynamics in the network?
Margaret - Not that high, as links represent signed agreements between organizations.
Sandy - See that you have object security - not what carrying. Some of our security problems in the routing protocol is this. Have you
considered that the trust paths could be bad?
Margaret - We have decided to trust the infrastructure, want to trust all the routers in the path vs. man in the middle. It's a design
tradeoff.
Sandy - Trusting the information?
Margaret - The information is a bit different. When do peering relationship, are deciding what information will trust from them. Not like
BGP. The trust router is set up to filter incoming stuff.
Sandy - So you are presuming all trust routers are trusted. If one bad, then all your assumptions will fail?
Margaret - It would be for a community. If don't agree with assumption, come and try to convince us otherwise.
Jeff - Seems this is bound to trust router. Seems could also be used to distribute a set of objects along a path.
Margaret - I guess could be used that way.
Jeff - It's a different direction. One issue in SIDR is mechanisms to get the security keys. A flooding mechanism with some idea of
trust among a community, if your protocol can do that, and scale, this may be applicable.
Margaret - It would be good to work together as we are interested in those attributes.
Curtis - Different from BGP as don't have a peer unless both sides agree. so a matter of distributing topology info - BGP issues seem
similar.
Margaret - Could have more than one interface - not routers though - these are servers. But not physically as in an actual router.
Curtis - If want a reliable server, probably want more than one interface.
Margaret - yes, though I don't deploy, that's for those who do to decide how to do.
 
Other business
--------------
Stewart: Discussed recently the lack volunteers for being an AD - one thing that takes a huge amount of time is reviewing documents to
get ready for IESG and following up with authors. But that's really the Shepherd's job. Some chairs do better than others. So a large
part of this could be done by Shepherds.
Joel - You said do/could do and negotiate with IESG on (their) list. But Shepherds can't do that part. IESG discuss among themselves - and
expect the other AD to do the issue.
Stewart - Can look at minutes.
Joel - Can't get out of the minutes what it (the issue) is sufficiently.
Stewart - Ok, understand. Also, the Shepherd doesn't have to be the WG chair. So please help in finding Shepherds and doing the job.
Jeff - How much is document work flow which tools can help?
Adrian - Not sure if being prodded by a tool gets you to do or better by a person.
Joel - I find it helps to paraphrase and work with the authors. So tool can not really do. And to help the authors push back on an AD.
Jeff - I was interpreting not the interpretation of doing, but the process part to say still open and list of what needed to do. I was
thinking the painful part, not the human part.
Adrian - There's a continuous dribble of requirements to the tools team, and it is getting better. But our concern is not so much picking up
the ball, but actually doing.
Stewart - Also please realize you (chairs) don't have to do this yourself. And that's the next generation of leadership, so good.

Open Discussion
---------------
None.