David J Taylor wrote:
Serious upset is a fair description.
I had no problem at all since I have a DCF77 receiver and thus ntpd had
advance leap second warning.
Jan 1 00:59:59 linux kernel: Clock: inserting leap second 23:59:60 UTC
My two GPS receivers also remained on-time but the driver I use for them
provides no advance leap second warning, have to look into that sometime.
However, the 4 NTP servers of my ISP (XS4ALL) were seriously disrupted.
And so were many others :-)
As Tim rightly remarked, leap seconds do not fit within the UTC-based
time model of Unix and Windows.
GPS time does not have this issue, but in return there is difficulty
when converting GPS to UTC time (which you would notice when writing a
driver for certain GPS receivers in binary mode). Extracting the
GPS-UTC offset from raw almanac data is no fun.
Rob
_______________________________________________
timekeepers mailing list
[email protected]
https://fortytwo.ch/mailman/cgi-bin/listinfo/timekeepers
