[ntp:questions] Re: ntp servers reporting leap second erroneously?

Danny Mayer mayer at ntp.isc.org
Wed Oct 26 12:47:58 UTC 2005


Martin Burnicki wrote:
> Dave,
> 
> David L. Mills wrote:
> 
> 
>>It may in fact be easier to have a separate process whose only mission
>>to step the clock back one second at a designated time and activated by
>>a cron job or whatever the equivalent in Windows. Don't do a slew at
>>this point; that would drive the NTP clock discipline nuts.
> 
> 
> The leap second is an exception, so it should be handled as an exception. If
> an additional offset is applied in adj_host_clock() for a specific
> interval, just to account for the leap second, then the variables of the
> clock discipline algorithm should not be touched at all.
> 

That sounds like a reasonable proposal. I don't like Dave's suggestion
of a separate process doing anything to the clock, I'd rather keep it
all in one place.

Danny
> 
> Martin




More information about the questions mailing list