[ntp:questions] Re: minpoll and maxpoll

Brad Knowles brad at stop.mail-abuse.org
Tue Dec 21 16:15:40 UTC 2004

At 3:15 PM +0000 2004-12-21, srui at fe.up.pt wrote:

>  good enough to keep time between power cycles. In a normal situation with
>  a big clock skew the NTP daemon crashes, so I was told that using a low
>  value for minpoll would solve this and it would be the correct equivalent
>  to a forced initial correction.

	The daemon doesn't actually crash.  What happens is that it sees 
the difference as being too large, and it decides that it can't fix 
this problem and quits.

>                                   Is this true? What solutions do I have
>  for this?

	Whoever told you that was very, very wrong.  Minpoll and maxpoll 
have to do with how long ntpd will wait between sending volleys to 
the selected time servers, to try and get an idea of what they think 
the current time is and how that compares to the others, etc....

	If you want to solve this problem, start up ntpd with a "-g" 
option, so that ntpd is allowed to step the clock by large amounts 
(only on startup).  You can help your time server settle down to a 
"good" value of time by putting "iburst" at the end of your 
configuration lines, and you want to make sure to specify a "drift" 
file to record the values that ntpd calculates for how much your 
system clock is off from "real"time, which will allow it to settle 
back down much faster on restart.

	We've tried to address all these issues in the Community 
Supported documentation at 
<http://ntp.isc.org/bin/view/Support/WebHome>.  You will want to pay 
special attention to sections #4 and 5.

Brad Knowles, <brad at stop.mail-abuse.org>

"Those who would give up essential Liberty, to purchase a little
temporary Safety, deserve neither Liberty nor Safety."

     -- Benjamin Franklin (1706-1790), reply of the Pennsylvania
     Assembly to the Governor, November 11, 1755

   SAGE member since 1995.  See <http://www.sage.org/> for more info.

More information about the questions mailing list