Re: [ippm] Last Call: <draft-ietf-ippm-twamp-yang-07.txt> (Two-Way Active Measurement Protocol (TWAMP) Data Model) to Proposed Standard

["tom p." <[e-mail]>]

Al, see end...

Tom Petch

----- Original Message -----
From: "MORTON, ALFRED C (AL)" <[e-mail]>
Sent: Friday, April 13, 2018 1:58 PM


> Tom, see end...
> >
> > ----- Original Message -----
> > From: "MORTON, ALFRED C (AL)" <[e-mail]>
> > Sent: Friday, April 13, 2018 1:19 AM
> >
> > ...
> > > > >> "The new well-known port (862) MAY be used.";
> > > > >> This was allocated in 2008 which seems to stretch the meaning
of
> > ‘new'
> > > >
> > > > [mj] Al, do you want to comment on this?
> > > >
> > > [acm]
> > > Yes, I'll comment.
> > >
> > > The UDP well-known port will have a new allocation,
> > > changing from TWAMP-CONTROL to TWAMP-TEST.
> > >
> > > Incidentally, the progress of that draft is also a dependency,
> > > but it is ready to ship (a small typo was identified in London,
> > > and the new draft has been available since).
> >
> > Al
> >
> > In which case, I think that that draft
> >
> >  draft-ietf-ippm-port-twamp-test
> >
> > needs to be a Normative Reference from
> >
> > draft-ietf-ippm-twamp-yang
> >
> > and should appear in the description clause, replacing the reference
to
> > 'new port'
> >
> > along with a note up-front to the RFC Editor asking them to replace
all
> > mentions of
> > draft-ietf-ippm-port-twamp-test
> > with the RFC number of that I-D when it is allocated.
> >
> > I find that the use of 'new' is rarely a good idea - road signs near
me
> > talk of a 'new' road layout that is now 10 years old - unless there
is a
> > clear date, explicit or implicit associated with it; and the
> > reallocation of the use of a port is IMHO a significant change that
> > needs calling out - the YANG module makes references to RFC 5357 and
I
> > think it is asking too much for users to track down another RFC that
> > updates RFC 5357 in order to see the change.
> >
> > Tom Petch
>
>
> [acm]
> I suggest:
> The well-known port (862) allocated in RFC YYYY MAY be used.
>
> RFC Editor: replace RFC YYYY with the RFC number of
> draft-ietf-ippm-port-twamp-test when it is allocated.

Yes, that is fine for the description clause; RFC YYYY will still need
to be in the References of the I-D, since it appears in the YANG module
and those are the rules for a YANG module, and so will need to appear
somewhere in the body of the text as well, lest it is flagged by nits as
an unused reference (Mahesh is already fixing this for RFC1305 and
RFC2330).

I did check with the RFC Editor and they said that they preferred notes
asking them to do replacements to be up-front, in one place, and not
spread throughout the I-D - hence my comment before about up-front.

Tom Petch

> > > Al