[ntp:questions] Frequency Error

robert sebestyen sebestyenrobert at hotmail.com
Wed May 30 14:58:27 UTC 2007

Hi folks
I created 2 NTP-Server Stratum 2 and 3 NTP-Server Stratum 1.
The Stratum 2 servers are synchronized by the Stratum 1.
Of course you will say, what else!?
My question is now following, i am not sure if i understood the 
documentation of
Dr. Mills right.
The frequency error is the difference between system-time and hardwaretime 
due to the fact that CPU´s are never equal.
The so called wander of both CPU´s is only 0,0705ppm on each!!!
But one server is showing 21.5 in average, the other one 39.6
I would like to decrease these values, somewhere around ZERO (if possible)
I tried to do that in following way.
I stopped ntp-service
I installed a cron-job making every 3rd hour a hwclock --systohc
(Cron-job is suppossed to run for more than 24hrs)
I edited in a startupscript following:
/usr/sbin/hwclock --adjust
/usr/sbin/ntpdate -b IP-Adress (of one of the Stratum 1 Servers)
/usr/sbin/hwclock --systohc

Running cronjob for more than 24 hrs
Shutting down the server and bring him up again.
Starting ntp-service

Is that a possible way to solve my problem?
Does anybody have experience how to handle frequency error in practice?
Would be nice to get a kick from your side!

Wir hören Musik, wir retten die Welt - Live Earth, das Konzert der Konzerte. 

More information about the questions mailing list