Re: [Ntp] Last Call: <draft-ietf-ntp-mode-6-cmds-08.txt> (Control Messages Protocol for Use with Network Time Protocol Version 4) to Historic RFC

[Harlan Stenn <[e-mail]>]

On 6/12/2020 10:20 AM, Brian Haberman wrote:
> Hi Pete,
> 
>      While I am responding to your note, I will capture thoughts related
> to what others have said so far...
> 
> On 6/2/20 1:17 PM, Pete Resnick wrote:
>> This use of the "Historic" status seems utterly bizarre to me. It's
>> certainly not how it's described in 2026.
>>
> 
> It is a bit strange, but was done for a reason...
> 
>> 1305 has already been obsoleted by 5905. This document simply gives
>> information about what those control messages were and indicates that
>> they ought no longer be used. This sounds like a fine candidate for
>> Informational status. When 5905 (or it's successors) advance to
>> Standard, this document can be included in the STD.
>>
>> If eventually you want to move a document to Historic, it would be 1305.
>>
>> I don't think publishing this straight to Historic makes sense.
>>
> 
> There are two intents with this document.
> 
> The first is to capture mode 6 commands that were not defined within RFC
> 1305, Appendix B. If you compare the command sets, you will see several
> additions in this draft. Several people have asked since the publication
> of 5905 about some of those commands as they were not documented anywhere.
> 
> The second intent is to signal that mode 6/7 commands aren't/shouldn't
> be used going forward. If you do a quick search for NTP vulnerabilities,
> you will find many that relate to abuse of these commands. Additionally,
> I/we received input (e.g., chronyd) no longer use these commands to
> accomplish their stated function. That is where making the document
> Historic seemed to make sense.

Shennanigans.

That chronyd doesn't support them is their policy choice.

This document is about mechanism.

Many people use mode6 to perform what would be standardized monitoring
and control of the reference implementation of NTP.

To my knowledge, there is no defined alternative to the mode6.

As for mode7, by definition that is vendor-specific space.  The purpose
of the specification is to make it easy to identify these packets.

> In the grand scheme of things, either Historic or Informational is a
> viable document status. I (as document author) did not see consensus for
> Standards Track, but that is not my call.

I disagree.  If I was to speak for the NTP Project (which I can and am
happy to do), that would be a "disagree/no" from there, too.

> Regards,
> Brian
> 
> 
> _______________________________________________
> ntp mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/ntp
> 

-- 
Harlan Stenn <[e-mail]>
http://networktimefoundation.org - be a member!