[ntp:questions] Mixed maxpoll values for mixed LAN/Internet servers - sensible?

David J Taylor david-taylor at blueyonder.neither-this-bit.nor-this.co.uk
Sat Feb 28 07:52:05 UTC 2009

Dave Hart wrote:
> On Feb 27, 5:38 pm, Dave Hart <daveh... at gmail.com> wrote:
>> When Dr. Mills first announced the change allowing the other peers'
>> intervals to go up with a refclock configuration, I believe he
>> observed a benefit from the mixed poll intervals, but he was thinking
>> at least two levels above my pay grade and the reasoning didn't
>> stick. I can't find the message to refer to right now, but I'll try
>> again later.
> Here it is:
> https://lists.ntp.org/pipermail/hackers/2003-November/000474.html
> I haven't come across any followup on the question of slower
> correction of temperature-induced drift swings, but I'll keep my eyes
> peeled.
> Cheers,
> Dave Hart

Thanks for that, Dave.  It's a pity that a change available in 2003 hadn't 
made it into my 2005 download!


More information about the questions mailing list