On Thu, Jul 10, 2014 at 4:20 AM, David Taylor <david-tay...@blueyonder.co.uk.invalid> wrote: > Simply that different folk have different needs.
This true but not nearly as much as people think. But yes different folks have different needs. Despite that in the next sentence > You later say: "It's a Reference Clock not an instance of NTPD.", in which > case I would consider one of these: you again drag out the RPi hammer that appears to make everything look like a nail to you. It's a *network attached* reference clock. For me this started when I commented on the USB redesign issues with the RPi. Despite that fact that was reported by many people and was documented in the RPi design docs your response was (essentially) -- it works for me, you must have a rubbish power supply. > Different folk, different needs. Yes but unless you're doing NTP(ref) R&D or NTP(ref) specific work the tracking you're so fond of is an impediment. This is your bias: ".. where it continues to be called an "NTP Server", which it not." When he actually wrote "I call it SNTP and not NTP because time is served from GPS alone and no other NTP servers are consulted" and: "I did speak with the originator many moons ago about this project, hoping that it would conform to NTP management standards. I felt it was a missed opportunity" The point is time transfer, not having access to NTP(ref) tools intended to monitor/manage NTP(ref). It's not a missed opportunity any more than SNTP/Chrony/OpenNTP or other proprietary yet NTP wire protocol compatible products are missed opportunites. Broader management and network design issues are far more important than clockstats/loopstats/peerstats or if your S1 server responds to NTP control messages. 99.999% of the people looking for a low power/low cost, GPS based S1 server would be well or best served by something like a Laureline. Full disclosure: I run an RPi based S1 server. And two BeagleBone Blacks servers and two x86 servers and a previous generation Laureline (while I wait for my new one). _______________________________________________ questions mailing list questions@lists.ntp.org http://lists.ntp.org/listinfo/questions
