Re: [Ntp] NTPv5: big picture

[Philip Prindeville <[e-mail]>]

> On Jan 1, 2021, at 6:54 PM, Paul Gear <ntp=[e-mail]> wrote:
> 
> It's great to aim for making NTP more suitable for high-accuracy
> environments which can currently only be served by PTP, but we shouldn't
> do this at the cost of making NTPv5 less viable for adoption by those
> billions by making it harder to deploy (e.g. by requiring changes to
> every firewall which allows NTP, or making it hard to work out whether
> you're looking at UTC or TAI time).


I occasionally bang on the default set of rules for OpenWRT, perhaps the most pervasive Open Source router package out there today.  What are your specific concerns about what’s not being done?

The current model for OpenWRT is that it’s a suborned peer to more accurate (lower stratum) servers out there, and then in turn provides time to the clients inside its perimeter.  It typically doesn’t support queries from external sources (which helps avoid spoofing attacks).

If there’s something we’re not doing that we should be, please call it out now.  Or conversely, something we’re doing that we shouldn’t be...

Thanks