Thanks Greg for your review comments. Right, the RFC 6374 DM and LM messages can carry either sequence number or timestamp but not both. There might be some cases where both are desired. The proposed sequence number TLV is optional and can be used with timestamps in DM and LM messages. We can clarify the text in the next update of the draft-gandhi-spring-udp-pm <https://datatracker.ietf.org/doc/draft-gandhi-spring-udp-pm/>.
Thanks, Rakesh On Tue, Nov 6, 2018 at 1:06 AM Greg Mirsky <[email protected]> wrote: > Dear Authors, > in your presentation of this draft at IPPM WG meeting I've pointed that > assertion in Section 6 of the draft: > The message formats for DM and LM [RFC6374] do not contain sequence > number for probe query packets. > is not accurate. RFC 6374 allows interpretation of the Timestamp field as > a sequence number. Section 3.4 explains that QTF and RTF values could be 0, > 1, 2, or 3, with 1 identifying the sequence number: > 1: Sequence number. This value indicates that the timestamp field > is to be viewed as a simple 64-bit sequence number. This provides > a simple solution for applications that do not require a real > absolute timestamp, but only an indication of message ordering; an > example is LM exception detection. > > Regards, > Greg > > _______________________________________________ > ippm mailing list > [email protected] > https://www.ietf.org/mailman/listinfo/ippm >
_______________________________________________ spring mailing list [email protected] https://www.ietf.org/mailman/listinfo/spring
