[Ntp] NTPv5: Operational use case for server/node identification

["Marco Davids (IETF)" <[e-mail]>]

Dear NTP Working Group,

As we develop NTPv5, I would like to raise an operational use case 
regarding server/node identification.

In large-scale deployments, operators frequently need to determine the 
exact node responding to an NTP query. This is particularly useful when 
troubleshooting synchronisation issues, monitoring anomalies, or 
operating anycast and load-balanced infrastructures where multiple 
servers may share the same address.

In the DNS world, operational mechanisms such as NSID (RFC5001) and 
ID.SERVER (RFC4892) are commonly used for this purpose.

To my knowledge, NTP currently lacks a standard mechanism for an 
operator to explicitly query the identity of the responding server.

Has the working group considered providing an optional node 
identification facility in NTPv5?

One possible approach could be an optional Extension Field that allows a 
client to request an administrative identifier from the server. The 
server could then return a configured identifier, such as a hostname, 
node name, or opaque operator-defined string.

Any such mechanism would of course need to take into account the privacy 
and reconnaissance considerations described in RFC7384 Section 3.2.12, 
as well as the monitoring requirements described in 
draft-ietf-ntp-ntpv5-requirements-04 Section 5.3. Consequently, it 
should remain lightweight.

I believe the protocol could benefit from such a capability and would be 
interested in hearing the working group's thoughts.

Kind regards,

-- 
Marco Davids
Research Engineer

SIDN | Meander 501 | 6825 MD | Postbus 5022 | 6802 EA | ARNHEM
T [phone] | www.sidnlabs.nl | Twitter: @marcodavids
https://mastodon.social/@marcodavids | Matrix: @marco:sidnlabs.nl
Nostr: 11ed01ff277d94705c2931867b8d900d8bacce6f27aaf7440ce98bb50e02fb34