lhommedumatch wrote:
I'm working on a oceanographic vessel ntpgmtaceb is our reference clock that is synchronised with a gps at the beginning of a cruise. The drift of this clock is <1ms each day. It's connected on the lan board of our vessel.I have removed these two lines server 127.127.1.0 fudge 127.127.1.0 stratum 10 Then: service ntpd stop ntpdate ntpgmtaceb 12 Apr 21:46:22 ntpdate[9860]: adjust time server 192.168.21.5 offset -0.000252 sec service ntp start ntpq -p remote refid st t when poll reach delay offset jitter ============================================================================== *ntpgmtaceb .1PPS. 1 u 20 64 377 0.144 -0.373 0.863 So it's seems to be really better, isn't it?
Delay on that connection is 0.144 vs 29.431 your original message so it's probably "better" because of that rather than removal of LOCAL(0). If original figure from delay was correct, eg due to network load, then you'd need to get that fixed. If using a recent ntpd you can have "tos orphan 10" to fallback to local clock if it's sufficiently reliable, and other tos options to set limits on what time if any is accepted and reported. What version of ntpd are you running? David _______________________________________________ questions mailing list [email protected] http://lists.ntp.org/listinfo/questions
