IESG Narrative Minutes

Narrative Minutes of the IESG Teleconference on 2008-06-05. These are not an official record of the meeting.

Narrative scribe: John Leslie (The scribe was often uncertain who was speaking.)

Corrections from: Russ Housley

1 Administrivia

Roll Call 1134 EDT Amy:
- Loa Andersson---regrets
- Jari Arkko---y
- Marc Blanchet---no
- Ron Bonica---y
- Ross Callon---y
- Michelle Cotton---y
- Spencer Dawkins---no
- Lisa Dusseault---late
- Lars Eggert---regrets (but called in later)
- Pasi Eronen---y
- Marshall Eubanks---no
- Sandy Ginoza---y
- Russ Housley---y
- Cullen Jennings---y
- Olaf Kolkman---
- John Leslie---y
- Cindy Morgan---y
- Chris Newman---y
- Ray Pelletier---regrets
- Jon Peterson---y
- Tim Polk---y
- Dan Romascanu---y
- Mark Townsley---y
- Amy Vezza---y
- Dave Ward---y
- Magnus Westerlund---regrets
Bash the Agenda
- new: p2p BOF
- NAT-PT made it to agenda
Approval of the Minutes of the past telechat
- May 22 minutes---approved
- May 22 narrative minutes---updated based on comments; approved
Review of Action Items from last Telechat
- Magnus -- not with us

2 Protocol Actions

2.1 WG submission

2.1.1 - New Items

Traffic Engineering Extensions to OSPF version 3 (Proposed Standard)
draft-ietf-ospf-ospfv3-traffic-11.txt
Token: David Ward
Balloting
Ballot Comments:
1. Jari Arkko: Comment [2008-06-03]: does not specify what type of IPv6 address is legal or illegal...
2. Pasi Eronen: Comment [2008-06-04]: RFC 2119 "MAY" in Section 4; Reference [OSPFV3] question
3. Tim Polk: Comment [2008-06-04]: secdir review noted two issues that really should get addressed.
4. Dan Romascanu: Discuss [2008-06-05]: missing information concerning manageability
Telechat:
- Amy: open positions; a discuss
- Dave: working with authors, AD followup
Two-Document ballot (Proposed Standard)
draft-ietf-idr-route-filter-16.txt
draft-ietf-idr-bgp-prefix-orf-04.txt
Token: David Ward
Balloting
Ballot Comments:
1. Jari Arkko: Discuss [2008-06-04]: one simple mistake (Minlen & Maxlen)
  Comment [2008-06-04]: two small edits
2. Lars Eggert: Comment [2008-06-04]: RFC2119 terms: SHOULD vs. MUST, etc.; some citations
3. Pasi Eronen: Comment [2008-06-04]: SecDir review: need to apply the filters locally?
4. Russ Housley: Comment [2008-06-03]: removing ORF entries
5. Tim Polk: Comment [2008-06-05]: security considerations section lacks a reference to the unchanged information
6. Dan Romascanu: Discuss [2008-06-04]: should include a list of any configuration or statistics objects that must be part of the instrumentation...
Telechat:
- Amy: discusses
- Dave: revised ID needed
Format for using TLVs in PIM messages (Proposed Standard)
draft-ietf-pim-join-attributes-03.txt
Token: David Ward
Balloting
Ballot Comments:
1. Jari Arkko: Discuss [2008-06-03]: one error (address field is variable length)
  Comment [2008-06-03]: can an attribute can disable processing of any attributes after itself?
2. Ron Bonica: Comment [2008-06-03]: Support Jari's DISCUSS
3. Lisa Dusseault: Comment [2008-06-03]: expand acronym
4. Lars Eggert: Comment [2008-06-04]: Doesn't describe IANA allocation procedure for new values
5. Russ Housley: Discuss [2008-06-03]: need response to Gen-ART Review
6. Chris Newman: Comment [2008-06-04]: "TBD" scattered throughout the text -- good idea to tell the RFC editor
7. Tim Polk: Discuss [2008-06-04]: (preliminary) expected to see [security concerns] references to RFC 5015 and 3973...
  security directorate review -- three specific issues not addressed
Telechat:
- Amy: open positions
- Mark: come back to me
- Dave: revised ID needed, extensive discussion with authors
- Mark: no objection
SEC Textual Conventions for Syslog Management (Proposed Standard)
draft-ietf-syslog-tc-mib-08.txt
Token: Pasi Eronen
Balloting
Ballot Comments:
1. Jari Arkko: Discuss [2008-06-02]: a MUST requirement that may be ambiguous
  Comment [2008-06-02]: mix of numbers and names in listing the Facilities
Telechat:
- Amy: a discuss
- Pasi:
- Jari: moved to a yes
- Amy: approved
Two-Document ballot (BCP & Informational)
draft-ietf-ipr-3978-incoming-09.txt
draft-ietf-ipr-outbound-rights-06.txt
Token: Russ Housley
Balloting
Ballot Comments:
1. Jari Arkko: Comment [2008-05-21]: The legend URL is currently non-existent.
2. Lisa Dusseault: Comment [2008-06-03]: definition of "RFC" (section 1.k) implies that all RFCs are IETF documents
3. Lars Eggert: Comment [2008-05-21]: [empty]
4. Cullen Jennings: Discuss [2008-06-04]: (maybe my confusion)
  1. how do outbound rights allow a person to create a bis version?
  2. can we, e.g., take email contributions and include them in a draft? Hard to understand some restrictions...
  3. it seems we need the Trust to grant the right to create derivative works: not clear that this is provided in outbound
  4. In outbound, the term contributions never seems to be defined
  5. does outbound section 4.2 cover Contributions or only IETF documents?
  6. outbound section 4.3: Clearly we don't want people to be able to use it in a way that removes the liability limitation in license legend.
  Comment [2008-06-04]: good to add an appendix that contains the initial text that needs to be placed at the legends URL
  outbound, section 4.3: consider adding "pseudo code" to the list
  Section 4.5 of outbound may need to be more explicit - perhaps an example.
5. Tim Polk: Comment [2008-06-05]: support adding [WG charters] to the definition for "Contribution"; iab has requested that iab stream be in scope; can't find a reference that clearly specifies that -iesg documents are in scope
Telechat:
- Amy: a discuss
- Russ: wrote note to Jorge; he's on vacation
- Cullen: seems to not allow things we always do
- Russ: definition of contribution, wait for Jorge
- Jari?: don't know legal review process
- Russ: Jorge co-author of one, reviewed other
- Tim: per Magnus discuss-discuss: I read contribution as broader; not clear it covers all it needs to
- Cullen: most things come through email
- Tim: do we want something about draft-iesg
- Russ: draft-ietf and draft-iesg are both part of the IETF stream; AD followup
Bidirectional Forwarding Detection (Proposed Standard)
draft-ietf-bfd-base-08.txt
Token: Ross Callon
Balloting
Ballot Comments:
1. Lars Eggert: Discuss [2008-06-04]: BFD is congestion-unresponsive... BFD aims to be a generic protocol... cannot simply punt on congestion control
  Comment [2008-06-04]: mailing-list no longer appropriate
2. Russ Housley: Discuss [2008-06-04]: IANA has questions
  Comment [2008-06-04]: should indicate intended status; s/must/MUST/; Section 6.6 wording
3. Chris Newman: Discuss [2008-06-05]: If password is human enterred, need to state a charset and any mandatory-to-implement preparation
  need to define the key management for keys
Telechat:
- Amy: 4 discusses
- Ross: Dave, do you think we need to discuss today
- Dave: I propose we add section on BFD timers to deal with other issues (to avoid link loss due to congestion, etc.)
- Lars?: depends on text
- Dave: can't tell what is cause of packet loss; bootstrapping;
- Lars?: shouldn't be application issue; BFD can generate considerable traffic; should be recommended-way in doc; ramp-up?
- Dave: choose larger of two ends; ramp-up is different way, better to have each end set limit
- Lars?: defend against clueless admins (in addition to limits)
- Dave: peer will detect packet loss
- Lars?: collateral damage
- Dave: working up new approach, take this offline
- Lars?: other option, limit the applicability
- Dave: consider MPLS (backchannel is generally routed);
- Dave: like to lump security concerns: Pasi, Russ; Tim
- [discussion of charset of passwords]
- Dave: point of cleartext passwords is to interop with many existing implementations; SHA stuff wasn't there when they implemented
- Ross: implementors want to implement same way they already use
- Dave: stuff is passed down -- pluggable library; this is what people are coding to; bunch of MD5 drafts, but not widely deployed
- Tim?: can we say these ways aren't very good
- Pasi: password length
- Dave: I have a note to clarify length; revised ID needed
BFD for Multihop Paths (Proposed Standard)
draft-ietf-bfd-multihop-06.txt
Token: Ross Callon
Balloting
Ballot Comments:
1. Jari Arkko: Discuss [2008-06-05]: lacks detail on encapsulation of the echo packets; should provide guidance on how BFD endpoints should deal with congestion across a path; should decide how applicable BFD documents are for the general Internet use
2. Lars Eggert: Comment [2008-06-04]: My congestion-control DISCUSS on the base document is especially relevant to multi-hop paths
3. Pasi Eronen: Discuss [2008-06-03]: How would BFD Echo work on multihop paths?
  Comment [2008-06-03]: say port 4784 has already been assigned by IANA
4. Russ Housley: Discuss [2008-06-04]: following SecDir Review, appeared to be agreement to add text to Security Considerations
Telechat:
- Amy: 3 discusses
- Dave: Jari's applicability request
- Jari: want to understand whole package, don't want to change ground rules later
- Dave: DHCP, etc., found no base-spec mods needed; I feel text is clear what BFD is trying to accomplish
- Jari: would like more on general characteristics of BFD, example special addresses, how is it implemented
- Dave: not sure how to proceed, could walk through public domain implementations
- Jari: usage of particular addresses unclear
- Dave: tried to make it bootstrapping-protocol-agnostic; are you asking for description of working with DHCP
- Jari: RFC 1122, what kind of addresses to use, what about redirect, can't see any way of configuring addresses without problems
- Dave: you want addressing rules, interaction with ARP; MPLS case
- Jari: state why you can't use 127.0...
- Dave: for wrong addresses, session will never come up
- Jari: what would normally work can fail with BFD
- Dave: take this offline
- Ross: we don't normally tell folks how to implement
- Dave: in IPv4, it just works (v6 slight issue)
- Jari: still don't understand ARP and ND
- Ross: don't routers already have to deal with all this?
- Jari: setting up filter, how to
- Dave: specifically negotiated echo mode with you
- Jari: should note that filters are likely to break this
- Dave: rest of discusses in email, revised-ID needed; Pasi is right that echo mode won't work in multi-hop
BFD for IPv4 and IPv6 (Single Hop) (Proposed Standard)
draft-ietf-bfd-v4v6-1hop-08.txt
Token: Ross Callon
Balloting
Ballot Comments:
1. Jari Arkko: Discuss [2008-06-05]:
  1. support Pasi's discuss and want to track its resolution.
  2. wonder how this can be made to work at all, except in very special circumstances; question what the special conditions must be; need to discuss applicability of in the general Internet
2. Lars Eggert: Discuss [2008-06-04]: Source ports for control packets have earlier been defined to be in the range of 49152-65535, can the same rule be applied here?
3. Pasi Eronen: Discuss [2008-06-03]: need some more text about how BFD Echo packets are actually sent
  Section 4.2 does not mention ND Redirects
  Comment [2008-06-03]: soy ports 3784 and 3785 have already been assigned by IANA
4. Russ Housley: Comment [2008-06-05]: title page header should indicate intended status; Gen-ART Review and SecDir Review both point out need to clarify second paragraph in section 2
Telechat:
- Amy: three discusses
- Dave: revised-ID needed
Generic Application of BFD (Proposed Standard)
draft-ietf-bfd-generic-04.txt
Token: Ross Callon
Balloting
Ballot Comments:
1. Jari Arkko: Discuss [2008-06-05]: had expected more information about WHEN BFD can be applied and in particular when it cannot
2. Lars Eggert: Comment [2008-06-04]: mailing-list no longer appropriate
3. Pasi Eronen: Comment [2008-06-05]: SecDir review clarification/editorial changes should be fixed
4. Dan Romascanu: Discuss [2008-06-05]: concerned about whole set of BFD documents brought today
  1. could find no information about operational considerations
  2. section 8: unclear what 'more aspects' are exercised
  3. would like clearer description of the interaction of BFD with management systems
  4. believe that there is a need to define a high level protocol independent information model for management
Telechat:
- Amy: discusses
- Dave: think we've covered everything; revised-ID needed
BFD For MPLS LSPs (Proposed Standard)
draft-ietf-bfd-mpls-05.txt
Token: Ross Callon
Balloting
Ballot Comments:
1. Jari Arkko: Discuss [2008-06-05]: agree with Pasi's Discuss.
  I do not think removing the RFC 1122 requirement on not using 127/8 on the wire is appropriate in general
2. Pasi Eronen: Discuss [2008-06-03]: Using 127/8 addresses violates a MUST in RFC 1122
  I can't find text saying how BFD Echo works in MPLS
  Comment [2008-06-03]: SecDir review had some editorial fixes and suggestions
3. Russ Housley: Discuss [2008-06-05]: do not actually see a response that deals with the content of Carlos Pignataro Last Call comments;
  per Gen-ART Review, it is unlcear whether BFD Demand mode is allowed (Section 6);
  per Gen-ART Review, What is supposed to happen in an IPv6-only environment? (Section 6);
  Comment [2008-06-05]: title page header should indicate intended status
4. Tim Polk: Comment [2008-06-05]: I expected to find an MPLS specific BFD ECHO packet format.
Telechat:
- Amy: discusses
- Dave: discussed in email; revised-ID needed
Resource ReSerVation Protovol (RSVP) Extensions for Emergency Services (Proposed Standard)
draft-ietf-tsvwg-emergency-rsvp-08.txt
Token: Magnus Westerlund
Balloting
Ballot Comments:
1. Ross Callon: Discuss [2008-06-04]: It seems to me that this relies on RSVP signaling, thus on the router alert option, which most ISPs don't allow...
  if this capability were widely deployed, routers would need to accept router alerts coming from whereever the emergency provider happens to be... a vector for DOS
2. Pasi Eronen: Discuss [2008-06-03]: Section 1 needs clarifications about scope and applicability
3. Cullen Jennings: Discuss [2008-06-04]: don't understand how authorization would work outside of a transitive trust walled garden environment
4. Tim Polk: Comment [2008-06-05]: I support Ross's concerns regarding the router alerts and resulting security vulnerabilities
5. Dan Romascanu: Discuss [2008-05-22]: applicability of extensions defined in this document is not described in terms consistent with other IETF work
6. Mark Townsley: Comment [2008-06-05]: I support the discuss comments about the use of router alert options.
7. David Ward: Comment [2008-06-04]: I am abstaining for reasons that Ross and Ron outlined.
Telechat:
- Amy: Magnus not here; open positions
- Lars: (trying to cover for Magnus)
- Lisa: no position, waiting for discussion of Ron's issue
- Chris: no position
- Jon: no position
- Ron: in private, quite a few say they do filter
- Russ: seems we need in-person meeting in Dublin
- Ross?: worried about false sense of security
- Pasi?: not clear what's driving this work
- Jon: these issues were discussed five years; needs to be public and ubiquitous
- Lars: place in AD-followup

2.1.2 Returning Items

(none)

2.2 Individual Submissions

2.2.1 New Items

Contexts for IMAP4 (Proposed Standard)
draft-cridland-imap-context-05.txt
Token: Chris Newman Note: Alexey Melnikov is the document shepherd
Balloting
Ballot Comments:
1. (none)
Telechat:
- Amy: only Magnus open
- Chris: some edits pending implementor comments; in this case the Document Shepherd should get notice of AUTH48
- Amy: approved
Sieve Email Filtering: Date and Index Extensions (Proposed Standard)
draft-freed-sieve-date-index-12.txt
Token: Lisa Dusseault
Balloting
Ballot Comments:
1. Jari Arkko: Comment [2008-06-02]: I would change the parameter order in the examples to match the syntax.
2. Chris Newman: Comment [2008-06-03]: A few issues I noticed while reviewing this: Section 4.1, 4.2, 4.3
Telechat:
- Amy: no active discusses; approved

2.2.2 Returning Items

(none)

3 Document Actions

3.1 WG Submissions

3.1.1 New Items

Host Threats to Protocol Independent Multicast (PIM) (Informational)
draft-ietf-pim-lasthop-threats-04.txt
Token: David Ward
Balloting
Ballot Comments:
1. Tim Polk: Discuss [2008-06-04]: (discuss-discuss) document references sparse mode and bidirectional PIM, but not dense mode.
Telechat:
- Amy: one discuss
- Dave: discussed with author, dense mode out of scope, RFC ed note
- Tim: cleared
ECC Support for PKINIT (Informational)
draft-zhu-pkinit-ecc-04.txt
Token: Tim Polk
Balloting
Ballot Comments:
1. Russ Housley: Comment [2008-06-05]: Please update reference to point to RFC 5280
2. Chris Newman: Comment [2008-06-03]: normative reference to the publication of a consortium with only one listed member seems questionable
Telechat:
- Amy: one discuss
- Tim: expect RFC-ed note can take care of it; AD followup

3.1.2 Returning Items

(none)

3.2 Individual Submissions via AD

3.2.1 New Items

Atom Bidirectional Attribute (Experimental)
draft-snell-atompub-bidi-06.txt
Token: Lisa Dusseault
Balloting
Ballot Comments:
1. Lars Eggert: Comment [2008-05-20]: Why didn't this come through the ATOMPUB WG?
2. Pasi Eronen: Comment [2008-05-19]: considerations text should have a pointer to UTR #36
3. Russ Housley: Comment [2008-05-19]: Suggestion: come up with a shorter reference handle for: [W3C.REC-xml-names-19990114]
4. Cullen Jennings: Discuss [2008-06-04]: should specify what takes precedence when there is both unicode control character and a bidi attribute;
  should specify what takes precedence when there is html with internal bidi control indication along with an atom level bidi attribute.
  Comment [2008-06-04]: Would be great to have a specific example of where this was needed.
5. Chris Newman: Discuss [2008-05-21]: cover the case of "dir" attribute with an empty-string value
  Comment [2008-05-21]: look forward to results from the experiment.
Telechat:
- Amy: couple of discusses
- Lisa: revised-ID needed
A Uniform Resource Name (URN) Namespace for the 3rd Generation Partnership Project (3GPP) (Informational)
draft-monrad-sipping-3gpp-urn-namespace-02.txt
Token: Lisa Dusseault
Balloting
Ballot Comments:
1. Pasi Eronen: Comment [2008-06-03]: References [RFC1123] and [RFC3958] not actually cited; remaining references should be normative
2. Cullen Jennings: Comment [2008-06-04]: disturbing that the plan is to have non unique URN in some case;
  hope this work replaces draft-montemurro-gsma-imei-urn and that we will not be seeing two requests here.
Telechat:
- Amy: no discusses;
- Lisa: AD followup please

3.2.2 Returning Items

(none)

3.3 Independent Submissions via RFC Editor

3.3.1 New Items

(none)

3.3.2 Returning Items

(none)

1254 EDT break

1300 EDT back

4 Working Group Actions

4.1 WG Creation

4.1.1 Proposed for IETF Review

(none)

4.1.2 Proposed for Approval

Data for Reachability of Inter/tra-NetworK SIP (drinks)
Token: Jon
Telechat:
- Amy: any objection?
- Jon: working on WGCs
- Amy: charter approved pending WGCs; will mailing-list change?
- Cullen: should change

4.2 WG Rechartering

4.2.1 Under evaluation for IETF Review

Common Control and Measurement Plane (ccamp)
Token: Ross
Telechat:
- Amy: any objections
- Ross: add T-MPLS (jabber comment, s/b MPLS-TP)
- Various: most MPLS documents not scheduled to go to ccamp
- Ross: milestones need work; probably will fix later
- Mark: my preference is to go out without waiting
- Amy: we could approve pending AD mods
- Ross: do we want to add Dan's text?
- Dan: like to leave door open, NETCONF
- Ross: would adding this confuse the LastCalls; I'll add that, L2VPN, milestones, then External review
Path Computation Element (pce)
Token: Ross
Telechat:
- Amy: any objections
- Ross: simpler, was ambiguous about multicast, this explicitly says point-to-multipoint; added a few milestones
- Ross: after talking with WGCs, will add MIB milestone
- Amy: will send recharter announcement
Multiprotocol Label Switching (mpls)
Token: Ross
Telechat:
- Amy: any objection
- Ross: list of additional milestones a bit larger, some Loa milestones not added yet; do I want to add L2VPN
- Amy: similar to ccamp, will go for external review pending updates from Ross

4.2.2 Proposed for Approval

EAP Method Update (emu)
Token: Pasi
Telechat:
- Amy: any objection
- Pasi:
- Amy: approved, send update annoucement

5. IAB News We can use

Amy: neither Loa nor Olaf here

6. Management Issues

Shepherding NAT-PT (Mark Townsley)
Telechat:
- Mark: concerned about IPv6/v4 as compatible as possible (his phone dropped out)
- Mark: note when I call Paris number, it drops at exactly two hours
- Mark: first step; thinking about steps after that; fallacy to think adding IPv6 experts to BEHAVE will fix this; need for full integration; worried about cross-area reviews happening too late; can't have both sides ignoring "responsibilities" of different areas
- Russ: could we give the WG an AD from each area?
- Mark: problem is too few people tracking
- Mark: thing changing the most is IPv6
- Russ: what about you and Lars as co-ADs
- Lars: expertise in NAT is missing from IPv6 groups
- Ross: too easy for clueful folks to skip meetings outside their area
- Jari: could we force two WGs to both have agenda item for overlap
- Dave: could assign specific aspects to different groups
- Jon: don't want to bet future of IPv6 on changing our culture
- Dave: need to decide how much belongs in one WG
- Jon: one deliverable vs. turning a corner (getting IPv6 people to BEHAVE)
- Jari: think we need work being done in two places
- Russ: out of time; need to set separate call time
p2p BOF
Telechat:
- Lars: I see two work items: use of excess bandwidth, sharing
- Chris: things that are possible within a single application
- Lars: queue on cable router independent of application
- Lars: third item: selecting among multiple peers; touching many areas, not just transport
- Chris: rather wait on point 3
- Jon: even before conference we had BOF proposal
- Jari: third item of different nature; for critical mass belongs in one place
- Jon: fundamentally different expertises
- Chris: don't want to couple these proposals, very different interests driving requirements
- Jon: not comfortable about overlapping top-down vs. grassroots
- Chris: could have separate BOFs

7. Agenda Working Group News

Jari Arkko (Internet)--- struggling -- candidates for chairs, advice welcome
Ron Bonica (O & M)---
Ross Callon (Routing)--- (gone)
Lisa Dusseault (Applications)--- pass
Lars Eggert (Transport)--- looking for TCP-M chairs; offering time to Larry Roberts to talk about problem he's tackling
Pasi Eronen (Security)--- (gone)
Russ Housley (General)--- pass
Cullen Jennings (RAI)--- pass
Chris Newman (Applications)--- pass
Jon Peterson (RAI)--- pass
Tim Polk (Security)--- (gone)
Dan Romascanu (O & M)---
Mark Townsley (Internet)--- (gone)
Dave Ward (Routing)--- pass
Magnus Westerlund (Transport)--- (absent)

1425 EDT Adjourned