Nelson Minar <[EMAIL PROTECTED]> wrote: > I'm now working on doing some active queries to NTP pool members. What > data would you recommend I track? Am I right in thinking ntpdc's > "sysinfo" gives most of the data I'd be interested in? (I get confused > by ntpq vs ntpdc).
The reference ID gives most of what you want, unless you want to start walking peers and the ntpdc monlist list, in which case I think you're already the world expert :-). The peerlist is useful (see below on how to determine what PPS really is). > Tim Shoppa wrote: > > While it would increase the number of stratum 1 servers, it would in > > fact reduce the diversity of stratum-0 refclocks being used. > > I found 80 stratum 1 servers in the pool today. Of those, 57 would > answer my ntpdc sysinfo request. Here's what they list for reference ID. > > 19 PPS > 17 GPS > 4 DCFa > 4 CDMA > > Does PPS tell us anything about the actual source of time? Or does that > just mean "time from a serial port"? 99% of clocks identifying themselves as "PPS" will be GPS-based. Usually you can confirm this by looking at the peerlist and finding the other stratum-0 entry. There are some folks with Rubidium or Cesium Beam clocks at home etc. doing PPS. See TVB's "Atomic Watch" page at http://leapsecond.com/pages/atomic-bill/ > BTW, there were a bunch of singleton reference IDs as well, including > the ever-popular LCL. That host seems to have decent time though. > > 1 hPPS > 1 WWVB > 1 Rb > 1 PG0A > 1 MSFa > 1 MSF > 1 LCL > 1 GPSi > 1 DCF > 1 CHU2 > 1 ACTS > 1 131.188.3.223 > 1 131.188.3.222 I'm the "CHU2". Other times I'll be WV5, WV10, or WV15. Tim. _______________________________________________ timekeepers mailing list [email protected] https://fortytwo.ch/mailman/cgi-bin/listinfo/timekeepers
