I think it's about Windows (and/or the NTP implementation). On FreeBSD, with the same GPS receiver there where no leap second problems.
Have a look to: https://ecoca.eed.usv.ro/mrtg/ntp1usvro_offset_server2-day.png The "critical" moment was at 2 o'clock (UTC+2) - nothing special to observe :). This is FreeBSD with GPS. But please have a look to: https://ecoca.eed.usv.ro/mrtg/ntp3usvro_offset_server2-day.png Interesting things ... This in Windows with GPS. I have 3 Windows 2000 servers and all three had the same problem as yours - but I restarted the ntp service (about 15-20 minutes after 23:59:60UTC) and, with a STEP correction the problem was automaticaly solved. _______________________________________________ questions mailing list [email protected] https://lists.ntp.isc.org/mailman/listinfo/questions
