[ntp:questions] Leap second preparedness

A C agcarver+ntp at acarver.net
Sun Jul 1 00:09:44 UTC 2012


It appears that my system did not respond well to the leap second.

The GPS has not yet received the leap second notification via the 
satellites (it is currently still reporting an offset of 15 seconds). 
The leap second was inserted by ntpd according to the leap file and then:

Jun 30 23:59:58 sunipx2 ntpd[271]: 0.0.0.0 011b 0b leap_event 
                                                                   x
Jul  1 00:00:21 sunipx2 ntpd[271]: SHM(0) 961a 8a sys_peer 
                                                                   x
Jul  1 00:00:21 sunipx2 ntpd[271]: 0.0.0.0 0413 03 spike_detect 
+1.009783 s                                                               x
Jul  1 00:05:09 sunipx2 ntpd[271]: 0.0.0.0 041c 0c clock_step +1.015057 
s                                                                 x
Jul  1 00:05:10 sunipx2 ntpd[271]: 0.0.0.0 0415 05 clock_sync 
                                                                   x
Jul  1 00:05:11 sunipx2 ntpd[271]: 0.0.0.0 c418 08 no_sys_peer 
                                                                   x
Jul  1 00:05:11 sunipx2 ntpd[271]: 184.105.192.247 8014 84 reachable 
                                                                   x
Jul  1 00:05:11 sunipx2 ntpd[271]: 130.207.165.28 8014 84 reachable 
                                                                   x
Jul  1 00:05:17 sunipx2 ntpd[271]: 131.144.4.10 8014 84 reachable 
                                                                   x
Jul  1 00:05:17 sunipx2 ntpd[271]: 66.228.35.252 8014 84 reachable 
                                                                   x
Jul  1 00:05:19 sunipx2 ntpd[271]: 173.160.143.19 8014 84 reachable 
                                                                   x
Jul  1 00:05:26 sunipx2 ntpd[271]: SHM(0) 8024 84 reachable 
                                                                   x
Jul  1 00:05:26 sunipx2 ntpd[271]: SHM(0) 903a 8a sys_peer 
                                                                   x12
Jul  1 00:05:27 sunipx2 ntpd[271]: PPS(0) 8084 84 reachable 
                                                                   x
Jul  1 00:05:27 sunipx2 ntpd[271]: PPS(0) 909a 8a sys_peer 


It lost PPS lock because SHM(0) was now reporting a full one second 
offset.  Eventually ntpd forced the clock steps.

The billboard currently shows that the Internet servers are still a 
second behind after the above resync:

o127.127.22.0    .PPS.            0 l   12   16  377    0.000   -7.519 
  2.663
*127.127.28.0    .GPSD.           4 l   13   16  377    0.000   -3.686 
10.446
  184.105.192.247 216.218.254.202  2 u  204 1024    1   38.373  -1014.3 
   0.122
  173.160.143.19  131.107.13.100   2 u  196 1024    1   75.144  -1019.4 
   0.122
  66.228.35.252   209.51.161.238   2 u  198 1024    1  112.360  -1010.5 
   0.122
  130.207.165.28  130.207.244.240  2 u   96 1024    1   93.724  -1015.8 
   4.515
  131.144.4.10    65.212.71.102    2 u  101 1024    1   96.809  -1014.8 
   4.166


Previously the Internet servers were only a few milliseconds offset but 
SHM(0) showed +1030 or so seconds offset.


More information about the questions mailing list