Re: [Ntp] WGLC: draft-ietf-ntp-interleaved-modes

[Miroslav Lichvar <[e-mail]>]

On Tue, Nov 27, 2018 at 05:47:07PM +0200, Tal Mizrahi wrote:
> Hi,
> 
> I believe the draft is ready for publication, assuming the following
> comments are resolved.

Thanks for the comments.

They all make sense to me and I'll try to address them in the next
version of the draft.

> Comments:
> - The draft header should say that it updates RFC5905 (although it is
> interoperable with existing implementations, which is mentioned in the
> draft).
> - "but that is difficult to implement in the current packet" - please
> clarify that it is difficult to implement unless a hardware timestamping
> mechanism is used.
> - Please clarify what "reference implementation" refers to, including a
> citation.
> - In Section 2, please describe what happens when a client request is lost,
> and what happens when a server response is lost, i.e., how each of the two
> sides re-synchronizes to the correct request/response.
> - Security considerations section: (1) Please mention the potential
> DoS/DDoS threats that may arise from the fact that the server maintains a
> per-client state. (2) Please also consider the following threat: a spoofed
> client request may cause the server's copy of the {receive,transmit}
> timestamp pair to be updated, thereby causing DoS to the legitimate client.
> (3) I would suggest to add a general comment to the security
> considerations, something like: "The security considerations of time
> protocols in general are discussed in [RFC7384], and specifically the
> security considerations of NTP are discussed in [RFC5905].".

-- 
Miroslav Lichvar