Eugen COCA wrote: > As I understood, leap seconds means the time is adjusted instantly with > one second. This has nothing to do with the drift file! Please see my > responses on yesterday messages to see some graphs on FBSD and Win. > > As I observed until today, no Windows system changed the time correctly > at 23:59:60UTC. Am I wrong?
The ntpd version [EMAIL PROTECTED] which we have published only a few days before the leap second event should have handled the leap second gracefully under the condition that its upstream servers also behaved well. I'll put a log taken from one of our machines online tomorrow when I'm back in the office. This version of ntpd is the only one I'm aware of which is capable of slewing the Windows system time quickly without interfering with the NTP loop filters. Those who watch these newsgroup know that we couldn't publish that version earlier due to stability problems of the Windows version of ntpd. One of this problems has been solved, but there's still another one around which has to be determined and fixed. Eugen, I know that version of ntpd doesn't work on your server and I'm really sorry for this. I wish we would have had more time to find the problem and make a version available which is stable on all systems. Martin -- Martin Burnicki Meinberg Funkuhren Bad Pyrmont Germany _______________________________________________ questions mailing list [email protected] https://lists.ntp.isc.org/mailman/listinfo/questions
