Minutes IETF120: 6man: Tue 20:00
minutes-120-6man-202407232000-00
| Meeting Minutes | IPv6 Maintenance (6man) WG | |
|---|---|---|
| Date and time | 2024-07-23 20:00 | |
| Title | Minutes IETF120: 6man: Tue 20:00 | |
| State | Active | |
| Other versions | markdown | |
| Last updated | 2024-07-29 |
IETF 120 6man
Tuesday, 23 July 2024, 13:00-15:00 PDT (UTC-7)
Room: Regency C/D
Chairs: Bob Hinden, Jen Linkova, Ole Trøan
Minute taker: Darren Dukes
Jabber Scribe: None
Jabber Room: 6man@jabber.ietf.org
Meetecho (Full Client):
https://meetings.conf.meetecho.com/ietf120/?session=33020
Meetecho (Onsite Tool):
https://meetings.conf.meetecho.com/onsite120/?session=33020
Agenda
Introduction, Agenda Review, and Document Status , Chairs, 10 min.
++Working Group Drafts++
Prioritizing known-local IPv6 ULAs through address selection policy, draft-ietf-6man-rfc6724-update , Tim Chown, 15 min.
SNAC Router Flag in ICMPv6 Router Advertisement Messages, draft-ietf-6man-snac-router-ra-flag , Ted Lemon, 10 min.
Signalling DHCPv6 Prefix Delegation Availability to Hosts, draft-ietf-6man-pio-pflag , Lorenzo Colitti, 15 min.
Deprecation of The IPv6 Router Alert Option, draft-ietf-6man-deprecate-router-alert , Ron Bonica, 10 min.
Entering IPv6 Zone Identifiers into User Interfaces, draft-ietf-6man-zone-ui , Brian Carpenter, 15 min.
++Active Individual Drafts++
IPv6 Node Requirements, draft-clw-6man-rfc8504-bis , Tim Chown, 15 min
++If Time Allows++
Internet Control Message Protocol (ICMPv6) Loopback, draft-mcb-6man-icmpv6-loopback , Tal Mizrahi, 5 min.
LOOPBACK6: A Utility For Detecting IPv6 Extension Header Changes, draft-he-6man-icmpv6-extensions-ipv6-ext-header , Ron Bonica, 5 min.
IPv6 Addresses for Ad Hoc Networks, draft-templin-6man-mla , Fred L. Templin, 5 min.
Clarification of IPv6 Address Assignment Policy, draft-carpenter-6man-addr-assign , Brian Carpenter, 5 min.
SLAAC Prefixes with Variable Interface ID (IID), draft-mishra-6man-variable-iids , Gyan Mishra, 5 min.
Introduction, Agenda Review, and Document Status , Chairs
- No RFCs published.
- Three drafts approved by the IESG and in the RFC editor queue
awaiting publication (see datatracker). - Reminder to collaborating working group members: please bring
content to 6man once there is some consensus in the source working
group
Working Group Drafts
Prioritizing known-local IPv6 ULAs through address selection policy, draft-ietf-6man-rfc6724-update , Tim Chown, 15 min.
- Presented updates in -09
-
Open questions before last call:
- Multiple questions were presented with list consensus (see
presentation)
- Multiple questions were presented with list consensus (see
-
Authors are close to closing open questions
- Authors ask for more conversation around open questions on the list.
-
Lorenzo:
- Has implemented draft.
- The 8 step algorithm was complex, can it be simplified?
- Jenn noted her agreement.
-
Some discussion Lorenzo and Tim around fc00::/8 vs fd00::/8
- Bob suggests treating all ULAs the same in this draft.
- Suresh K noted his agreement.
- Bob suggests treating all ULAs the same in this draft.
-
Eric K
- Asked about policies via specification vs policy via
administration - Suggests to not overspecify fc00::/8 and fd00::/8
- Asked about policies via specification vs policy via
-
Updates are expected to the document.
SNAC Router Flag in ICMPv6 Router Advertisement Messages, draft-ietf-6man-snac-router-utAC usecase.
- Need a way to differentiate RA's from SNAC routers vs infra routers.
-
Ask for early allocation of the flag
- implementations are deployed squatting on other flag bits.
- early allocation is asked
- WG last call
-
Lorenzo suggested Must be set to zero by non stub routers
- Tim Winters agrees
-
David Farmer
- Asked about RA guard
- Ted SNAC RAs are filtered with RA guard so this draft does not
affect RA guard.
-
Tim Winters
- What happens when a SNAC router is not a SNAC router?
- Ted - can be noted in SNAC draft
-
Phillip Tiesel
- What happens if a configurable router is also a SNAC router?
- Ted - The SNAC draft notes this
Signalling DHCPv6 Prefix Delegation Availability to Hosts, draft-ietf-6man-pio-pflag , Lorenzo Colitti, 15 min.
- Title changed
-
Goal - signal the network supports PD per device.
- A flag signals this, reasons given in presentation
-
WG LC
- Some recent discussion on the list
- Authors think discussion was on deployment model, not the
protocol - Protocol spec has not changed
- Reviewed some issues discussed - see slides.
-
Closing WGLC?
- Request early allocation is next.
-
Suresh
- can router turn this off and on? Think about it and write an
update
- can router turn this off and on? Think about it and write an
-
Tom
- this is necessary, the alternative is more NAT66
-
Erik K, Nick
- support the work, necessary
-
Tim Chown
- How many flags are left? 4.
-
Ole
- In favor of solving network extension problem, but the draft
leaves too many corner cases unresolved. - Note we cannot leave network extensions only to large enterprise
issues. - Notes we should not split documents among working groups.
- Notes a corner case on router transition to PD capable.
- Ole has sent them to the list.
- Lorenzo disputed some concerns are answered on the list, but
there may be some disconnect some how - Ole committed to send questions to the list for resolution.
- In favor of solving network extension problem, but the draft
-
Erik K
- yes consider corner cases - some were stated
- reminds participants to discuss issues to resolution
-
Suresh K
- Suggests throw issues in the issue tracker
- Ole notes they are on the 6man tracker
https://github.com/ietf-6man/pio-pflag/issues
-
Bob
- Notes the WG appears close to consensus, but will look and
determine next actions.
- Notes the WG appears close to consensus, but will look and
Deprecation of The IPv6 Router Alert Option, draft-ietf-6man-deprecate-router-alert , Ron Bonica, 10 min.
-
Summary
- Protocols may continue to use the option
- Future protocols must not use the Router Alert option
-
Motivation was restated
-
Can Router Alert option be replaced by other means?
- Ron gave two examples and their alternatives (see slides)
-
David
- notes a snooping use case is missing from the draft
- Ron notes alternates exist
-
Eric V
- clarified the examples (RSVP-TE & MLDv2) were only examples not
necessary to implement
- clarified the examples (RSVP-TE & MLDv2) were only examples not
-
Tim notes depricate was confusing with existing uses persisting
- Bob notes this is normal
-
Suresh
- Some older HW may only have RA as the means of punting to slow
path - Should we talk to some vendors?
- Ron and Suresh to followup offline.
- Some older HW may only have RA as the means of punting to slow
-
Bob
- Question for the list - how do you do MLD without router alert?
- Erik K notes RFC3810bis draft may do this?
Entering IPv6 Zone Identifiers into User Interfaces, draft-ietf-6man-zone-ui , Brian Carpenter, 15 min.
-
Brian notes failure to modify syntax of URI for local addresses
- notes IESG discusses
-
Reminded the WG of the only technical content
- A UI should be able to identify link local address and zone
identifier (paraphrased - see slides)
- A UI should be able to identify link local address and zone
-
Brian notes main changes during adoption call
-
Should this be a formal update to RFC4007?
- Authors think no
-
Bill Fenner
- Notes the problem is hard
- Is there any point to this if it doesn't apply to web browsers?
- Brian: yes, there are use cases that are not web browsers.
- Brian believes this will spur browsers to consider change.
-
Erik K
- should we socialize it with browser folks now? When WGLC is done
Erik will.
- should we socialize it with browser folks now? When WGLC is done
-
Discussion with Lorenzo and Brian regarding specification of the
link local text format -
Doug M - is there a difference between a browser and any other app?
- Difference is internally an IPv6 addr is a zone ID plus 128bit
addr. - Doug - is a web browser app well defined?... discussion ensued.
- Difference is internally an IPv6 addr is a zone ID plus 128bit
-
David Farmer
- Let IESG decide if there should be an update
- LL addrs does not need to have an interface ID.
- If the correct default zone is built then zone IDs are not
needed - David proposes fixing zone identifiers.
- David indicates this work does provide guidance to applications
that ignored the problem.
Active Individual Drafts
IPv6 Node Requirements, draft-clw-6man-rfc8504-bis , Tim Chown, 15 min
- initial -00 is published with minor edits to to RFC8504
- Tim asked for input from WG on other edits that may influence this
work. - Noted things that can be added/removed - it was a big list.
-
Lorenzo:
- Notes requirements on v6 nodes supporting v4
-
Tim Winters
- feedback that v6only section is useful
-
Eric V
- Continue the work, re DHCPv4 option 108, this is OK in a
specific section
- Continue the work, re DHCPv4 option 108, this is OK in a
-
Tom Hill
- supports the work
- May want to add SR6
-
Jen L
- What a host OS should do for APIs that render CLIs?
-
Gyan Mishra
- Should v4 as a service be added and the like?
- Authors: OK, lets consider that.
If Time Allows
Internet Control Message Protocol (ICMPv6) Loopback, draft-mcb-6man-icmpv6-loopback , Tal Mizrahi, 5 min.
- Described the technical aspects of the draft
- Asked for feedback and whether there is support for adoption.
LOOPBACK6: A Utility For Detecting IPv6 Extension Header Changes, draft-he-6man-icmpv6-extensions-ipv6-ext-header , Ron Bonica, 5 min.
- A means of providing extension headers in the ICMP message
- allows the receiver to determine if EH's have been modified on the
path - Ron suggest his and Tal's draft get merged.
- David made a technical suggestion for the merged draft and packet
sizes - Ron appeared amenable.
IPv6 Addresses for Ad Hoc Networks, draft-templin-6man-mla , Fred L. Templin, 5 min.
- Fred noted an issue with addressing in Ad Hoc Networks
- Defines new addresses as Multicast local addresses
- Fred provided address prefixes for these new addresses.
- Jen - looking for comments on the list.
Clarification of IPv6 Address Assignment Policy, draft-carpenter-6man-addr-assign , Brian Carpenter, 5 min.
- Brian notes problems in address assignment policy
- Brian proposes a new BPC to identify IETF review required for new
assignment (not just IESG and IANA).
SLAAC Prefixes with Variable Interface ID (IID), draft-mishra-6man-variable-iids , Gyan Mishra, 5 min.
- outlined the case for variable iid space
- Bob encouraged anyone interested to listen to this in V6OPS session
on Thursday morning.