Leap second on 31 march
A couple of days ago this appeared in my system logs
my first reaction of course was "great! they gave us one second more of sleep! MY PRECIOUSSSS", but then I realized that yes, this year there was supposed to be a leap second, but it should have been in June, not in March.
Other people I know noticed the message, but nobody knew anything else about it, and duckduckgoing didn't find anything, so I'm asking the lazyweb: does anybody know what happened?
Mar 31 23:59:59 kernel: Clock: inserting leap second 23:59:60 UTCmy first reaction of course was "great! they gave us one second more of sleep! MY PRECIOUSSSS", but then I realized that yes, this year there was supposed to be a leap second, but it should have been in June, not in March.
Other people I know noticed the message, but nobody knew anything else about it, and duckduckgoing didn't find anything, so I'm asking the lazyweb: does anybody know what happened?
Show all 5 replies
I thought the leap seconds were coordinated through ntp services. What ntp servers are you using?
The ones in the debian pool: debian.pool.ntp.org
I've checked all the boxes I have access to (Ubuntu and Fedora), and I don't see any evidence of a leap second reported. It looks like there have been subtle bugs over time with both leap second announcements and local handling of leap seconds (depending on the source). If you find out what happened, I'd be interested to hear.
Update: it seems that this has been traced to a single layer1 ntp server.