IETF-79 Routing Area Open Meeting
=================================

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

Scribe: Deborah Brungard 

Administriva and Area Status
----------------------------
Stewart opened, Adrian reviewed the Note Well. Reminded to try and use Jabber. Reminded about Routing Area
mailing list and wiki.

Routing Area Working Group Reports
----------------------------------

BFD (Dave W) - Not meeting this week, a number of documents being discussed on list.

CCAMP (Lou) – Met this morning and will again tomorrow afternoon. Full agenda. Two areas primarily WSON
and G709. One joint last call with mpls and pwe3 on TP Control Plane Framework. Today, primary focus on
G.709, tomorrow will be WSON. Had five new RFCs. Another Ethernet
in pipeline. Couple documents, almost ready for IESG.

Forces (Jamal) - Two new rfcs -applicability statement and implementation report. Thanks to Adrian.
Two more almost done, hopefully when done, can shut down, maybe March.

IDR (John) - Meeting Thursday- very full agenda! Drafts being edited post WGLC: link-bandwidth,
draft-ietf-idr-rfc4893bis (AS 4 byte). Drafts being readied for IESG: draft-ietf-idr-bgp-identifier,
draft-ietf-idr-mrai-dep, draft-ietf-idr-bgp-bestpath-selection-criteria. Fairly close:
draft-ietf-idr-bgp-mibv2, draft-ietf-idr-as4octet-extcomm-generic-subtype, draft-ietf-best-external. Ready
for WGLC: draft-ietf-idr-bgp-issues. New WG drafts: draft-wkumari-deprecate-as-sets,
draft-ietf-idr-fsm-subcode-00. Active: add-paths, draft-ietf-idr-ext-opt-param, error handling.

ISIS (Dave W) - A number of doucumets went to last call. Path forward now for ISIS Trill, and hearing about
802.q. Doing some operational enhancements.

KARP (Joel) - Working on main documents. Making progress on sorting out core security issues. Trying to
keep the order of events and work items sensible.

L2VPN (Gilles) - Update on WG status: L2VPN signaling stuck in RFC editor queue due to reference to
segmented PW draft, 2 other drafts in IESG evaluation, 6 other active WG drafts. Potential L2VPN/L3VPN
merge. Either way we'll need to recharter. Two key items that may be added are E-Tree and MAC VPN. Shane
Amante is stepping down as WG chair after five years, and being replaced by Nabil Bitar, so many thanks
to Shane.
 
L3VPN (Ben) - The WG has reached consensus on an updated charter & milestones and it is now with the ADs
waiting for IESG approval. draft-ietf-l3vpn-mvpn-spmsi-joins which plugs a gap with IPv6 support in
multicast L3VPNs has been approved by IESG. We had some discussion (as part of a first stage of information
gathering) on possibly merging L2VPN & L3VPN together, there will be another discussion in L2VPN on
Thursday. 

MANET (Stewart) - Meeting Thursday afternoon. Both chairs could not be here. One of the authors will be
substitue chair for the day. One draft which found technical problem after handed in to IETF publishing
which needs to get fixed.

MPLS (Loa) - We have two meetings this time. We are very busy with core MPLS and MPLS-TP work, but we are
making good progress with a number of drafts progressing through WG last call, and some good constructive
discussions in the meeting and on the list.

OSPF (Acee) - New RFCs: RFC 5838 OSPFv3 Multiple Address Family. Active Drafts: OSPF Multi-Instance, OSPF
Transport Instance. Other WG Drafts: CCAMP-Multiple WDM Drafts, OTN Drafts, RFC57867bis.
And L3VPN - draft-ietf-l3vpn-ospfv3-pece-06. OSPF WG meeting tomorrow with a full agenda. Presentations
will cover security improvements, IPv6 transition, WSON, as well as some extensions to basic OSPF mechanisms.

PCE (Julien) – New RFCs on PCEP for P2MP and Synchronized VECtor. Hot topics for Thursday afternoon are
GMPLS extensions for WSON and interdomain/interAS for multi-area applications. And one on introducing state
in PCE for stateful applications.
Adrian - when will you co-charter? Julien - I am discussing with my co-chair.

PIM (Mike) - Met yesterday. Most interesting discussion on rechartering to add igmp work. Will take to
WG to confirm. Other primary discussion was on erratim for 4651 spec. Have identified someone to do it.
Thanks to Adrian for helping with queue mapping.

PWE3 (Matthew) - One new RFC (RFC5994) since last IETF, one in RFC Ed Queue, and one in IESG evaluation(AD
review). Three drafts have recently completed WG last call. PWE3 will be holding a service provider
implementation survey regarding the deployment of the PW control word and the exception mechanisms used for
VCCV. (Refer to slides uploaded)

ROLL (Adrian) - One document went thru, many issues being addressed. Other documents coming thru. Starting
to think about what to do with their charter going forward.

RTGWG (John) - Meeting Friday: Composite Link Requirements (WG doc), Composite Link Framework (proposed),
draft-lu-fast-notification-framework-00, draft-kini-ospf-fast-notification-00.

SIDR (Sandy) - Will meet Thursday afternoon. 4 documents in last call at the moment. We have about 10 drafts
that have new versions just announced yesterday. Several ready for last call. Should have ready for ADs soon.

VRRP (Adrian) - Not meeting this time. One draft keeps timing out and being renewed on a mib for vrrp. Once
complete will close the wg.


Update from LISP (Joel Halpern)
-------------------------------
Reviewed slides. Addresses routing scalability so of interest to this group. Not for standards track,
will be Experimental specs. Most interesting change based on running stuff, was that the MTU problem is real so
strongly recommended MTU handling. Alternative way to resolve - built around bgp. versioning - proxy is important
part - need to be able to work with.
Aiming for Prague to have ready for rfc. Another document on deployment which is very important to here.
Our first idea was wrong so needs now improvement, but this will be important going forward and are
discussing.
John - did you already say what the goal of lisp is?
Joel - Experimental rfcs - can help with scaling of IP addressing, which some believe, there are problems with
and others don't see problems.
Fred - Do know what to do with the MTU issue?
Joel - They do - you should look at it.
Adrian - Implications for any of our routing protocol?
Joel - None that I know of - the basis is to run as is - don't change routing protocols.


Public Discussions/Private Stitch-Ups (Stewart Bryant)
------------------------------------------------------
Folks have said to me that we do too much private discussions in the early phase of documents. Should we do
something about it? Reviewed slides. Primary way for IETF is our mailing list. Others from other SDOs
don't necessarily know this. Adrian reviewed some thoughts of why folks may do private discussions first.
What can we do about this? We can police aggressive intimidation.
Dimitri - There is no common goal. How can we restore a community goal so to have a public discussion?
Richard Gray - Some of the working groups are huge so the message volume could be huge. People want to
discuss in a smaller group.
Stewart - Should we have a subsetting of the wg lists?
Adrian - Couldn't we do that with the subject lines of our emails?
Loa- I'd advise against it as we have mails copying to tp list, mpls, pwe3, ccamp. So not very efficient.
John - I don't see it as a problem. People want to have a discussion early in the development. If the
document has some thought to it, I see that as good. As mailing lists are huge, it is good to have some
development first.
Loa - I'm not concerned on private discussion. But concerned on inability to bring key discussion to the
mailing list.
Matthew - I'm not so concerned in the early days of a draft, but I am concerned on wg drafts.
Especially when drafts are just changed, without discsusion on list. So people feel then they can not
influence wg drafts, if change just happens. That's not good.
Dan king - Authors have a responsibility to communicate to the list on updates of a draft, on what's been
changed, and summarize outstanding issues. This will help also generate interest.
Tony Li - Part of the problem here is the expectations people have. Some expect everything done on list and
offended if not on list. Had a group member threaten me. Suggest should document in the Tao what are the
standard procedures.
Dave Ward - Some public lists suffer from dos attacks, so need to do private discussions a bit to get some
progress done.
Stuart - What can we do about dos attacks?
Dave - We don't really have tools to handle - can try to change subject or use the heavy hammer to stop.
Stuart - So tony seems to say that we shoudl document this.
Acee - I think it is a good point. People have said if had disucssion on the list would not have the discussion
again. People need to decide. Chairs need to decide if discussion should go to list. If can doucment why made
a decision then maybe wouldn't have years later people questioning again.
Dimitri - Maybe when a document is first submitted, have a list of open issues. Then can focus discussion
around the document. Also helpful if have discussions between meetings.
Ben - I don't have a problem with private discssion if people don't think that they can just rubber stamp.
Think useful to have private discussion to get some maturity.
Adrian - I'll wrap this up. Specific actions: get your subject lines correct in emails and update them so
can filter. When do an update to a draft say to the list what/why. Chairs pay attention to dos attacks on
list. Be aware of newcomers that could contribute and ask specific questions for help/engagement to bring
disucssion to the list.
Acee - Anyone submitted a better tool for doing better searches of the mail archives?
Stuart - Seems Google has a better view than our internal tool.
Adrian - Should tell the tools team of desire.


The Danger of Codepoints in I-Ds (Adrian)
-----------------------------------------
Adrian reviewed slides.

Stuart - Another method used in the past as some do need a new registry. One can write a draft with a registry
and someone acts as register. We created and maintained this in pwe3.
Loa - Two comments. First, if need it, then need to say this in the draft's IANA section. Also, this is
not fail safe as another wg may use it.
Stuart - So this is if an existing register. But not if a new register.
Shane - 2nd what said. If wg keeping a registry, then I think this will lead to more problems, so prefer
other alternatives suggested in your slide.
Ben - As someone who just got an early allocation, the process is easy, don't be worried about it.


Open Discussion
---------------
No discussion