[ntp:questions] What went wrong with the leap second

Rob van der Putten rob at sput.nl
Mon Jan 2 13:18:44 UTC 2006


Hi there


Some log entries from various boxes (GMT + 1);

Jan  1 01:02:03 lvp ntpd[1477]: synchronisation lost
Jan  1 01:18:31 lvp ntpd[1477]: time reset 1.001842 s
Jan  1 01:18:31 lvp ntpd[1477]: synchronisation lost
Jan  1 01:49:38 lvp ntpd[1477]: time reset -0.712867 s
Jan  1 01:49:38 lvp ntpd[1477]: synchronisation lost
Jan  1 01:57:08 lvp ntpd[1477]: time reset -1.319332 s
Jan  1 01:57:08 lvp ntpd[1477]: synchronisation lost
Jan  1 02:46:45 lvp ntpd[1477]: time reset 0.153601 s
Jan  1 02:46:45 lvp ntpd[1477]: synchronisation lost
Jan  1 04:24:46 lvp ntpd[1477]: synchronisation lost
Jan  1 04:29:04 lvp ntpd[1477]: synchronisation lost
Jan  1 04:33:02 lvp ntpd[1477]: synchronisation lost
Jan  1 04:37:20 lvp ntpd[1477]: synchronisation lost
Jan  1 05:16:51 lvp ntpd[1477]: synchronisation lost

Dec 31 10:57:49 sput ntpd[1387]: kernel time sync enabled 0001
Dec 31 11:31:59 sput ntpd[1387]: kernel time sync enabled 0011
Jan  1 01:12:22 sput ntpd[1387]: no servers reachable
Jan  1 01:26:27 sput ntpd[1387]: time reset +1.006216 s
Jan  1 01:42:33 sput ntpd[1387]: time reset -0.244206 s
Jan  1 01:42:33 sput ntpd[1387]: kernel time sync enabled 0001
Jan  1 01:51:08 sput ntpd[1387]: no servers reachable
Jan  1 02:00:53 sput ntpd[1387]: time reset -1.997449 s
Jan  1 02:30:00 sput ntpd[1387]: no servers reachable
Jan  1 02:49:14 sput ntpd[1387]: time reset -1.055841 s
Jan  1 03:19:21 sput ntpd[1387]: time reset -0.824090 s
Jan  1 03:36:32 sput ntpd[1387]: time reset -0.208863 s
Jan  1 03:49:25 sput ntpd[1387]: no servers reachable
Jan  1 04:02:20 sput ntpd[1387]: time reset +0.307601 s

Dec 31 11:53:55 pc4 ntpd[5503]: kernel time discipline status change 11
Jan  1 01:33:37 pc4 ntpd[5503]: synchronisation lost
Jan  1 02:30:28 pc4 ntpd[5503]: time reset 0.461177 s
Jan  1 02:30:28 pc4 ntpd[5503]: kernel time discipline status change 1
Jan  1 02:30:28 pc4 ntpd[5503]: synchronisation lost
Jan  1 02:55:18 pc4 ntpd[5503]: synchronisation lost
Jan  1 03:11:25 pc4 ntpd[5503]: time reset -0.276457 s
Jan  1 03:11:25 pc4 ntpd[5503]: synchronisation lost
Jan  1 03:22:22 pc4 ntpd[5503]: time reset 0.448885 s
Jan  1 03:22:22 pc4 ntpd[5503]: synchronisation lost
Jan  1 03:49:12 pc4 ntpd[5503]: time reset -1.725224 s
Jan  1 03:49:12 pc4 ntpd[5503]: synchronisation lost
Jan  1 04:07:33 pc4 ntpd[5503]: synchronisation lost
Jan  1 04:16:15 pc4 ntpd[5503]: time reset 1.066471 s
Jan  1 04:16:15 pc4 ntpd[5503]: synchronisation lost

Dec 31 11:07:05 pc5 ntpd[22456]: kernel time sync disabled 0001
Dec 31 11:32:45 pc5 ntpd[22456]: kernel time sync enabled 0011
Jan  1 01:21:51 pc5 ntpd[22456]: kernel time sync enabled 0001
Jan  1 01:38:55 pc5 ntpd[22456]: no servers reachable
Jan  1 04:00:05 pc5 ntpd[22456]: time reset -0.298050 s
Jan  1 04:36:45 pc5 ntpd[22456]: time reset +0.410064 s
Jan  1 05:16:31 pc5 ntpd[22456]: time reset -0.276530 s

Apearently some NTPDs will increase the second length by 1/86400 to avoid
the leap second. Could this be the cause?
Is there a difference in how GPS and DCF77 handle the leap second?
Do some NTP implementations insert the leap second at 23:59:59 local time
rather then UTC?


Regards,
Rob
-- 
+----------------------------------------------------------------------+
|             Poland does not belong in the European Union             |
|             http://www.sput.nl/politics/poland-sux.html              |
+----------------------------------------------------------------------+




More information about the questions mailing list