[ntp:questions] Leap second question

unruh unruh at invalid.ca
Wed Mar 20 19:13:56 UTC 2013

He was saying that yes, your assumption that leap seconds are added only
is a very reasonable one. However, the leapsecond agreement says that
they can be both added and deleted, and ntpd should certainly allow for
both. Note that many of the refclocks do allow for the leapsecond to be
deleted as well as added.

On 2013-03-20, Stephen Yu <syu at symmetricom.com> wrote:
> Chuck,
> I guess I did not state my question clearly enough. I would like to confirm this was not an omission in the code to check for condition that can lead to "sys_leap = LEAP_DELSECOND". Searching the code, it seems "sys_leap" can only be LEAP_NOTINSYNC, LEAP_NOWARNING, or LEAP_ADDSECOND.
> Thanks,
> Stephen
> -----Original Message-----
> From: Chuck Swiger [mailto:cswiger at mac.com] 
> Sent: Wednesday, March 20, 2013 10:18 AM
> To: Stephen Yu
> Cc: questions at lists.ntp.org
> Subject: Re: Leap second question
> Hi--
> On Mar 19, 2013, at 6:15 PM, Stephen Yu wrote:
>> The question is why "sys_leap = LEAP_ADDSECOND" is unconditional. Is this based on the prior knowledge that earth always rotates slower? In other word, could it ever be "sys_leap = LEAP_DELSECOND"?
> Over the long term, the earth's rotation is slowing over time primarily due to tidal friction from the moon.  It's possible for a large-scale event like a big earthquake or meteor strike to increase the earth's rotational speed slightly, but even a ~9.0 richter earthquake resulted in a speedup of only a few milliseconds.
> Regards,

More information about the questions mailing list