Hi,

On my two linux-boxes I've got this in my logs:

Box 1: Jul 01 01:59:59 [kernel] [11613806.049909] Clock: inserting leap
second 23:59:60 UTC

Box 2: Jul 01 01:59:59 [kernel] [976439.291563] Clock: inserting leap second
23:59:60 UTC

(I'm on UTC+2)

That's the only thing in my logs I could find indicating that something
happened tonight ;) 

-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of Dave Hart
Sent: Saturday, June 30, 2012 01:40
To: Daniel Norton
Cc: [email protected]
Subject: Re: [Pool] Leap Second UTC

On Fri, Jun 29, 2012 at 8:40 PM, Daniel Norton <[email protected]>
wrote:
> From http://www.ntp.org/ntpfaq/NTP-s-algo-real.htm#AEN2499
>
> 5.3.4. What happens during a Leap Second?
>
> The theory of leap seconds in explained in Q: 2.4.. In reality there 
> are two cases to consider:
>
> If the operating system implements the kernel discipline described in 
> Section 5.2, ntpd will announce insertion and deletion of leap seconds 
> to the kernel. The kernel will handle the leap seconds without further 
> action necessary.

Although exactly how it's handled is platform-defined.  One can imagine
simply stepping the clock, pausing the system for a second, or more subtle
schemes that try to ensure always-increasing clock readings.  I'm curious if
any don't simply step the clock.

> If the operating system does not implement the kernel discipline, the 
> clock will show an error of one second relative to NTP's time 
> immediate after the leap second. The situation will be handled just 
> like an unexpected change of
> time: The operating system will continue with the wrong time for some 
> time, but eventually ntpd will step the time. Effectively this will 
> cause the correction for leap seconds to be applied too late.

NAK on that FAQ.  Well, assuming recent ntpd that is.  I'm not sure when it
changed, but modern ntpd steps the clock back if not using the kernel loop
discipline.  That FAQ hasn't been maintained in many years and is
increasingly unhelpful.  I'd like to see it updated simply to say it's out
of date and shouldn't be relied upon.  The intended replacement is the
http://support.ntp.org collection of [t]wikis.

Cheers,
Dave Hart
_______________________________________________
pool mailing list
[email protected]
http://lists.ntp.org/listinfo/pool

_______________________________________________
pool mailing list
[email protected]
http://lists.ntp.org/listinfo/pool

Reply via email to