it can be an internal server that has the NTP service open for clients, or an public NTP server. To be sure, use 2 servers. To check if you can access it, run ntpdate timeserver.ntp.ch
Edit the /etc/ntp.conf file
Rename the current file to ntp.bak.conf and make a small new one: restrict default ignore
server timeserver.ntp.ch # Swiss time
server ntp.ucsd.edu # Univ of California, San Diego
restrict timeserver.ntp.ch mask 255.255.255.255 nomodify notrap noquery
restrict ntp.ucsd.edu mask 255.255.255.255 nomodify notrap noquery
server 127.127.1.0 # local clock
fudge 127.127.1.0 stratum 10 #so it only takes over if the rest fails
restrict 127.0.0.1 driftfile /etc/ntp/drift broadcastdelay 0.008 authenticate no
Set your system clock right
Run the following command a couple of times: ntpdate -u timeserver.ntp.ch # or whatever server you want to use
You will see the initial diffence in time go away afer the 2nd or 3rd time.
Set hardware clock
/sbin/hwclock --systohc
Run the ntpd daemon
service ntpd start
Add ntpd to the services started at boot time
chkconfig ntpd on
Check the NTP results
ntpd -p
will show you what the difference is between your clock and that of the servers you added. You are looking for lines like
remote refid st t when poll reach delay offset jitter
==========================================================================
LOCAL LOCAL 10 l 30 64 377 0.000 0.000 0.004 *
192.168.246.107 192.168.246.88 3 u 41 128 177 0.313 5.598 0.345
and not lines like
remote refid st t when poll reach delay offset jitter
==========================================================================
192.168.246.126 LOCAL 11 u 37 128 375 0.204 6082.02 6069.84
Jitter is too high!
Recent Comments