Re: [Ntp] WGLC: draft-ietf-ntp-using-nts-for-ntp

[Miroslav Lichvar <[e-mail]>]

On Tue, Nov 06, 2018 at 08:46:12PM +0000, Karen O'Donoghue wrote:
> Folks,
> 
> This message initiates a three plus week working group last call for: 
> 
> Network Time Security for the Network Time Protocol
> https://datatracker.ietf.org/doc/draft-ietf-ntp-using-nts-for-ntp/

I support advancing this document. It's well written and it secures
the most important mode of NTP - client-server.

>From an implementor's point of view, I'd prefer if TLS 1.3 was
required, but I don't know much about TLS or crypto in general.

I still think it would be good to specify the maximum length of the
cookie. This would simplify the client implementation. From a previous
discussion, it looks like 256 octets should be enough.

In 5.2 there is "Possibly, some additional extension fields which are
neither encrypted nor authenticated.  These are discarded by the
receiver." I'd suggest to change the last sentence to "These are
normally discarded by the receiver". This is explained later in the
document for both client and server sides.

In 5.6 there are two instances "zero-padded to a word (four octets)
boundary" and two instances of "bytes". The size of a word and byte is
specific to the architecture, so just use "to a four-octet boundary"
and "octets".

In 5.7 there are typos in "but MUST be encrypted from sent from
server to client." and "comply with RFC 5095, Section 7.4".

In 5.7 there is "In that case, it MUST be able to detect and stop
looping between the NTS-KE and NTP servers.". This could be explained
in a more detail. The detection here means that the client should
check if there were no good NTP packets received between NTS-KE
exchanges?

-- 
Miroslav Lichvar