Skip to main content

Minutes IETF114: ntp

Meeting Minutes Network Time Protocols (ntp) WG Snapshot
Title Minutes IETF114: ntp
State Active
Other versions markdown
Last updated 2022-08-10


NTP WG Meeting @ IETF 114
Wednesday, 27 July 2022,1000-1200 (1400-1600 UTC)
Independence C

  1. Administrative and Agenda Bashing

    • Karen: Note Well and IETF 114 Meeting Tips
    • Minutes: Dieter
    • No agenda bashing
  2. NTP/TICTOC WG Document Status Review/Update (Chairs)

    • YANG Data Model for NTP has been published as RFC 9249. Thanks
      to the authors, especially to Dhruv who really persisted through
      the entire process.

    • Two documents in IESG process

      1. Interleave Modes
      2. Mode 6 cmds
    • WGLC for the Chronos document has been issued. Please review and
      comment on the mailing list.

    • Two documents waiting for shepherd write-up

      1. Update registries. In the last interim we agreed that the
        document is ready to be passed to the IESG. During shepherd
        write-up review it was decided to change it from
        informational to standards track because it updates several
        standard tracks documents. We will do a very short WGLC on
        this specific issue. There was no opposition with this plan
        of action.

      2. PTP Enterprise profile. We've gotten approval from the IEEE
        to share the underlying specification (IEEE 1588) for the
        purpose of reviewing this document. There is going to be a
        terminology issue with this document as we pass it through
        the IESG. Its terminology is based on the 1588 specification
        which uses the concept of master and slave clocks. This
        probably will be problematic during review. The 1588 working
        group has an active PAR that is addressing this issue and is
        making progress. For the purpose of moving this document
        forward we will reference the IEEE PAR with the
        understanding that this IETF document is a profile of the
        1588 specification. We will be using the language that is in
        the current 1588 specification with the understanding that
        the 1588 working group is updating the language. This
        document can be updated at a later date to reflect the
        updated language.

        • Denis: (also member of the 1588 working group). The
          current text still uses the old language, and we will
          note in the write-up that if the alternative terminology
          amendment gets approved, we will change it?
        • Karen: We are not committing to go back and update the
          document. We are just saying that we apply the current
          language used and that we understand that this language
          is going to be updated. We advise implementors to refer
          to the work of 1588 wg to see what the updated language
          would be.
        • Karen: If the 1588 wg publishes new terminology before
          this document goes forward, we can update the
          terminology of this document. It is even possible to
          update is later as a new RFC if someone is willing to do
          the work.
  3. NTP v5 use cases and requirements

    • James: New version published as working group document. Some
      editorial changes and nits. Please provide comments on the
      mailing list.
    • Dieter: Did you consider the thread model of RFC 7384 "Security
      Requirements of Time Protocols in Packet Switched Networks" for
      the section 5 of this document?
    • James: I wasn't aware of it. I will consider it.
    • Karen: That was the requirements document that was done prior to
      NTS. Would be worth to look at.
  4. NTP Registry update draft

    This topic was already considered in topic 3 (WG Document Status

  5. Work with no updates:

    • Roughtime

      • Originally wanted to have a hackathon effort on this
      • No further comments or questions
      • Karen: Interested to get implementations experience on it.
        It would be possible to plan a hackathon effort during IETF
    • NTP v5

      • Karen: The requirements draft is adopted as wg document. We
        could consider this document for adoption call.
      • Mirsolav: There have been some discussions about conformity
        between this draft and the requirements draft. There have
        been some issues I already forgot.
      • James: I did some analysis of what Doug has been provided.
        There were only two or three things where the draft didn't
        meet the requirements. That was against the previous
        version. I will re-evaluate that against the current
        version. We can talk about that on the mailing list.
      • Karen: Do you think it is worth to have an adoption call now
        or do you want to wait longer?
      • Miroslav: Not sure.
      • Karen: I think we should go ahead and do a call for
        adoption. Then we have both documents side-by-side.
      • James: According to Doug's spread sheet there are five
        things that are outstanding. With that I think the document
        is suitable for adoption. These issues can be worked out
      • Karen: will proceed with an adoption call for the NTPv5
    • NTP Over PTP

      • Karen: Next steps?
      • Mirsolav: It is ready for adoption. There is work needed for
        some issues with the PTP sequence id.
      • Karen: Adoption call will be issued.
    • NTS for PTP

      • There have been two proposals on the table. These shall be
        merged into a single document. That has not been done yet.
        We are waiting for the primary authors.
  6. AOB

    • Miroslav: NTPv5 - Local timescale

      • Does it make sense to support local timescale for NTPv5?
        Request from some users.
      • Karen: Any comments?
      • James: Put it on a list as a topic to be considered.
      • Denis: It is worth discussing. We need to understand the
        different use case involved. I can understand the motivation
        behind that. There is more than one way to do that. Do we
        want to transfer different time scales over the wire or will
        it be sufficient to have a field on the client side that
        provides the translation?
      • Karen: This topic shall be discussed on the mailing list
    • Karen: Virtual interims

      • We plan to have a virtual interim monthly.
      • We need to look for a hackathon effort in November (IETF
    • Karen: Based on the next few virtual interims we will need to
      decide if we request meeting time at IETF 115

      • Erik: do want people to meet on IETF 115
      • Karen: A number of regular participants are not here.
        Difficult to answer this question.
      • Karen: This will be decided in the next month or so.
    • It would be good to ask for some implementor or operator reports
      for NTS at a future meeting.