[Pool] Off by a second leap second failures

Tapio Sokura oh2kku at iki.fi
Tue Jul 7 13:19:22 UTC 2015


On 3.7.2015 21:13, Ask Bjørn Hansen wrote:
> Here’s a daily count of servers that were off by ~1 second (+/- 100ms) during that day.

I also came across one stratum 1 server that was keeping time fine until
the leap, but after the leap it was at a +800 ms offset for a few days.
It stayed that way until it was (I assume manually) fixed. I don't know
exactly what happened, maybe a 200 ms wide PPS pulse was involved and
for some reason the wrong edge was being observed after the leap (and on
the wrong second).

There might have been others as well that were in a non-integer second
offset, I was only monitoring a small number of servers. My gut feeling
is that this time more of the servers leaped properly than during the
previous leap second(s).


More information about the pool mailing list