Session Date/Time: 25 Jan 2024 13:00

ROLL

Summary

The ROLL WG held an interim meeting with low initial attendance, but proceeded after additional participants joined. The discussion focused on the current status of several drafts, including Projection, NSAA Extension, Enrollment, MOPEX, and Border Router Failure Detection (RFTD), detailing open issues and proposed next steps. Key technical points were raised regarding the "experimental" status for Projection, a non-normative language issue in RFC 6719 relevant to NSAA Extension, and various review comments for RFTD. The group also discussed meeting cadence and the need to prioritize document progression given participant availability.

Key Discussion Points

Draft Status Overview:
- Projection: Submitted to IESG. Version 34 is expected to address routing directorate comments. An open point remains regarding whether the draft should be experimental or standard track.
- NSAA Extension: Submitted, but returned to the working group due to open issues (ticket #28 reopened).
- Enrollment Priority: All issues are addressed, and the draft is considered ready to move forward.
- MOPEX: In progress, addressing open issues (2 tickets).
- RFTD (Border Router Failure Detection): Four reviews received (Anan, CARS, Security Directorate, Routing Directorate) with associated tickets. These reviews raise points on large-scale deployments, energy consumption, use of overhearing, threshold values, rationale for prime numbers, and CRFC description. Many of these require clarification on design choices rather than fundamental technical disagreements.
- RPL Capabilities and Storing Route at Acknowledge (MVR): Changes have been implemented as suggested in previous IETF discussions.
Meeting Cadence & Future Interims:
- A proposal was made for regular virtual meetings every two months due to typically low participation.
- The next interim meeting is proposed for February, with a Doodle poll to be shared on the mailing list.
- A virtual interim in a different time zone was suggested as an alternative to an IETF Brisbane meeting to potentially attract more participants from the APAC region, given current low attendance.
Errata for RFC 6719 (Mr. Hoff):
- Dominique raised an editorial errata for RFC 6719 (misspellings, forgotten words). The RFC editor referred this to John, and further discussion may occur on the mailing list.
- A correct point to notice regarding RATA (RPL Information at the RPI) typically being non-zero when arriving at the DoAR was highlighted.
Projection Draft - Experimental Status Discussion:
- Discussion ensued regarding the "experimental" status proposed by the Routing Directorate reviewer. Pascal expressed concern that past experience with experimental drafts in ROLL (e.g., P2P) indicated low engagement.
- A sense of those present indicated that a standard track publication would be preferred, with John confirming this was acceptable.
- Pascal confirmed that some other minor fixes had been made based on reviewer feedback, and he believed the reviewer was generally satisfied, but an explicit email confirmation was lacking.
MOPEX Draft - Reusing Old Values:
- A previous consensus on accurately describing the reuse of old values for the new MOPEX field was recalled.
- Dominique offered to contribute text to address this point in the GitHub repository.
Enrollment Draft - Trickle Timer Reset:
- Dominique raised a question from a previous action item: whether a new reason for resetting the Trickle timer needs to be specified in the document, or if it aligns with existing RFC 6550 conditions.
NSAA Extension Draft - RFC 6719 Normative Language:
- Dominique identified an issue in RFC 6719, noting a lack of normative language explicitly stating that a neighbor with a cost equal to MaxPathCost shall not be chosen as a parent, despite informal hints in the text.
- Two approaches were considered:
  1. Rely on the informal hinting in RFC 6719.
  2. Make the NSAA extension independent of RFC 6719 by removing direct reference to MaxPathCost and instead describing it as a supplement to an objective function for parent selection.
- Aris (co-author) had previously committed to proposing new text in a separate GitHub branch to address this by decoupling the NSAA extension from RFC 6719's MaxPathCost. The group will await this text for further discussion.
Working Group Direction:
- A suggestion was made to focus the group's energy on progressing the most important documents through the publication process in the coming months.
- Dominique indicated that he would be available to contribute significantly for the next few months but would be unavailable during the summer.

Decisions and Action Items

Decision: Proceed with a quick meeting despite low initial attendance.
Decision: Propose a next interim meeting in February; a Doodle poll will be shared on the mailing list.
Action Item: Pascal to ping Susan to confirm agreement on the changes made in Projection v34 and clarity on the "experimental" status discussion.
Action Item: Dominique to contribute text to the MOPEX draft on GitHub regarding the reuse of old values for the new MOPEX field.
Action Item: Create a ticket to track the question of whether the Enrollment draft needs to update RFC 6550's Trickle reset criteria.
Action Item: The WG will await Aris's proposed text in a separate GitHub branch for the NSAA Extension draft to address its dependence on RFC 6719's MaxPathCost concept.
Decision: The working group will focus on advancing key documents to publication within the next three months.
Decision: Explore holding a virtual interim in a different time zone to potentially attract new members from the APAC region.

Next Steps

The chairs will share a Doodle poll for the proposed February interim meeting.
Pascal will follow up with Susan regarding the Projection draft.
Dominique will contribute text to the MOPEX draft.
A ticket will be created for the Trickle timer reset question in the Enrollment draft.
Aris is expected to provide proposed text for the NSAA Extension draft's decoupling from RFC 6719.
Authors of the RFTD draft will continue addressing review comments in preparation for a Working Group Last Call.
Internal reviewers will be assigned for the Enrollment draft, with reviews hoped for by early February.