Skip to main content

Minutes IETF115: ntp: Fri 12:00

Meeting Minutes Network Time Protocols (ntp) WG Snapshot
Title Minutes IETF115: ntp: Fri 12:00
State Active
Other versions markdown
Last updated 2022-11-22


NTP WG Meeting @ IETF 115

11 November 2022, 1200 UTC

Draft Agenda

  1. Administrative and Agenda Bashing
  2. NTP/TICTOC WG Document Status Review/Update
  3. Chronos (finalize WGLC results)
  4. NTPv5

    • NTPv5 Hackathon project @ IETF 115
    • Use cases and requirements
    • NTPv5 specification
    • Timescales
  5. Roughtime

  6. Ongoing work needing updates
  7. AOB
  8. Way Forward
    • Future meetings
    • Discussion of status of NTS deployment


  1. Administrative and Agenda Bashing

  2. NTP/TICTOC WG Document Status Review/Update

    • RFC9327 published as historic document
    • Interleaved modes:

    • One abstain, in IESG

    • Eriks suggestions:

      • Either add note to explain that it is not as scary as
        suggested in some of the comments

        • Depends on the type of change whether this would need to
          go by the working group or whether this is purely
          editorial and can be done immediately.
      • Defer interleaved mode to NTPv5

    • Miroslav:

      • Is the option of changing document status to informational
      • Erik: Probably, but underlying problem is Ops is not
        convinced interleaved mode is safe to do, probably need some
        reassurance. might not be fixed by changing document type.
    • Erik action item: Miroslav indicated that there is a new
      version, Erik will try to get the relevant people within the
      IESG to look at the new text. Reevaluate after second attempt.

      • NTP registries: Waiting on Erik to check
      • Enterprise Profile for PTP: Still needs shepherd writeup
  3. Chronos (finalize WGLC results)

    • Karen: Name changed due to conflict issues, final textual
      mistakes fixed, Dieter will do shepherd writeup
  4. NTPv5

    • NTPv5 Hackathon project @ IETF 115
    • Use cases and requirements

      • David: We might need to say something in the requirements
        draft on tradeoffs between security vs synchronization
      • Miroslav: Probably ok to have some fields unauthenticated,
        especially for intermediate residence time headers, impact
        is the same as a delay attack
      • David: Also consider tradeoffs caused by crypto-induced
      • Miroslav: This probably doesn't have that much impact.
      • Rick van Rein: Please make crypto very strong, it is needed
        for a lot of other cryptosystems to have good knowledge of
    • NTPv5 specification

    • Timescales

      • David:

        • Two approaches, timestamps either always in the same
          timescale, or allow different timescales in the
          timestamps, each with corresponding way of communicating
          information for getting to/from different timescales.
          There is a need for a discussion on which we go with and
          particularly why.
      • Miroslav:

        • Current aproach server can have a preferred timescale.
          This allows server to use an efficient timescale to use
          for the transmit timestamp as the generation of this is
          along a critical path.
      • Karen:

        • 1588 (the ptp standard) has done some work on this, we
          should at least look at it to ensure we have options for
          interoperability, as we likely will need to be able to
  5. Roughtime

    • Karen: Update hackathon, little activity due to not many people
      working on it.
    • Karen: Document authors are not present, probably little use for
      discussion now, ask for more comments on the list.
    • Ben Laurie:
      • Interest in roughtime at google, a couple of people there
        have looked at the standard and will send some comments.
  6. Work needing updates

    • NTS for PTP:

      • Karen: There is still interest in doing this, primary author
        has not yet gotten update done, last indication was that
        there should be an update coming in november, so hopefully
    • NTP over PTP ():

      • Karen: No comments on call for adoption, it will be adopted.
  7. AOB and Way Forward

    • Erik Kline: Thanks everyone for helping with the Erata and the
      hackathon work.
    • Future meetings

      • Karen: There has been discussion between the chairs about
        keeping us moving forward at a better pace
      • Karen: Virtual interims seem better at keeping us moving,
        there will be more, tentative date given in slides.
      • Karen: We will try to have another hackathon session during
        the yokohama ietf, predicated on having people who are
        willing to spend time during in that timezone during that
    • Discussion of status of NTS deployment

Check with Erik on Errata