NTP howto

Revision as of 15:20, 18 June 2008

This is a short description on how to get two computers have the same clock using NTP (Network Time Protocol)

Introduction - how

Assuming two computers are to be synchronized, one called master and the other slave. Here the following IP's are used:

• master IP 192.168.1.10
• slave IP 192.168.1.20

Are master and slave running OK

Test that master is running (as root on the master):

ntpq -p
     remote           refid      st t when poll reach   delay   offset  jitter
==============================================================================
*LOCAL(0)        .LOCL.           6 l   39   64  377    0.000    0.000   0.001

The important things are that there is a '*' at the start of the line and the 'st' (stratum) value is less than 15. This should be the situation about 4 minutes after boot of the master.

Test the slave (as root on the slave)

ntpq -p
     remote           refid      st t when poll reach   delay   offset  jitter
==============================================================================
*192.168.1.10    LOCAL(0)         7 u   19   64   17    0.254  -20.070   1.857

The important things are the '*' in the first column, the stratum 'st' beeing less than 16 (stratum of master minus 1) and that the offset is acceptable (in ms).

If offset is too big

(regardless of stratum or '*') Update clock:

root@smr19:/rhome/demo# ntpdate -u 192.168.1.10

This will update the clock on this computer to the clock on the 192.168.1.10 computer (if the difference is above 0.5 second - else nothing is changed). The accuracy is typically better than 1 ms.

If stratum is 16 or no '*' in first column

(and ntpd has been running for at least 4 minutes) Restart the ntpd daemon on the slave (as root):

/etc/rc.d/ntpd restart

After another 4 minutes, the situation should be as shown above.

If not read next section

Sync takes time

When ntpd is started it will decide which server is the better, this takes typically about 4-8 minutes (requires about 4-8 samples spaced 64 seconds - the 'reach' field (octal) should have about 4 ones (017 octal) as shown below - may reach 0377 (8 times '1' in octal)).

Test progress

as root the status can be monitored by, e.g.:

root@smr19:/rhome/demo# ntpq -p
     remote           refid      st t when poll reach   delay   offset  jitter
==============================================================================
 192.168.1.10    .INIT.          16 u    -   64    0    0.000    0.000   0.000
*pc89.iau.dtu.dk LOCAL(0)         3 u   19   64   17    0.254  -20.070   1.857

This shows that the pc89 computer is the time source (the * as first character) the offset is -20.070 ms The last poll was performed 19 seconds ago and the poll interval is 64 seconds.

After a few hours the clock difference will reduce to about the magnitude of the jitter (in ms - here about 1.8 ms).

The other source 192.168.1.10 shows 16 in the field 'st' (stratum - a measure of clock quality), ntpd will never try to synchronize to a source with stratum 16. See configuration of master clock source to ensure a low stratum.

Configuration of slave computer

The ntpd configuration is typically at

/etc/ntp.conf

On the computer that should be synchronized the configuration file could look like (some comments removed):

# file /etc/ntp.conf
#
server  192.168.1.10
server pc89
#server kalman
#
driftfile /etc/ntp/drift
multicastclient                 # listen on default 224.0.1.1
broadcastdelay  0.008

The server part shows in this case two clock sources, the 192.168.1.10 and the PC89.

The server pc89 is the one to use in this example, as the 192.168.1.10 is not available in this test setup. In the 'ntpq -p' above the 192.168.1.10 has never had a successful update.

The driftfile is the place where ntpd saves the adjustment to the local clock to maintain it at the same speed as the reference computers clock. NB! there is only one drift value, so if the computer has to keep sync with different computers, where these computers may drift at a a different rate, then ntpd will use 'very' long time to readjust to the new source. That is once is gets used to the errors in the PC89 clock source, and then is reconfigured to a sutuation where 192.168.1.10 is available and PC89 is not, then this will probably drift with a quite different value - and require maybe 8-20 hours to get back into sync.

The restrict part and other parts of the configuration file are commented out.

Start of slave

The ntpd may be started automatically - on a slackware distribution the a starter script must be present in the /etc/rc.d directory, like this

$ ls /etc/rc.d/rc.ntpd -l
-rwxr-xr-x 1 root root 8331 2007-01-22 16:53 /etc/rc.d/rc.ntpd

If it is executable, then it is started at boot time.

Configuration of master computer

The ntpd configuration is typically at

# /etc/ntp.conf on clock master
#
server 127.127.1.0 #local clock
fudge 127.127.1.0 stratum 2
#
driftfile  /etc/ntp/drift

This shows one server only, being the local real-time clock. The 'fudge' line tels ntpd to fake local clock as stratum 2 (a high quality) clock source.

The ntpd need to be started on this computer, and a ntpq -p would show something like:

pc89 jca # ntpq -p
     remote           refid      st t when poll reach   delay   offset  jitter
==============================================================================
*LOCAL(0)        .LOCL.           2 l   39   64  377    0.000    0.000   0.001

The users of this clock source should then obtain stratum level 3, as shown above.

If it fails

no server suitable for synchronization found

To test the time difference between two computers where ntpd is running, the following query can be used:

root@smr15:~# ntpdate -q pc89
server 192.38.66.189, stratum 16, offset -4437.744951, delay 0.02577
16 Jun 15:28:00 ntpdate[7235]: no server suitable for synchronization found

In this case the PC89 computer is 4437 seconds offset relative to this computer, shat is OK, but the no server suitable for synchronization found is bad. This means - in this case - that the server clock quality is too bad, stratum is 16, as the PC89 server is just restarted. Wait a few minutes and try again:

root@smr15:~# ntpdate -q pc89
server 192.38.66.189, stratum 3, offset -4437.762400, delay 0.02577
16 Jun 15:31:54 ntpdate[7237]: step time server 192.38.66.189 offset -4437.762400 sec

Now the PC89 clock source is OK (stratum 3 is less than 16 and thus OK), ready for a synchronization:

root@smr15:~# ntpdate -u pc89
16 Jun 14:19:39 ntpdate[7238]: step time server 192.38.66.189 offset -4437.770068 sec

and now the query shows

root@smr15:~# ntpdate -q pc89
server 192.38.66.189, stratum 3, offset -0.000240, delay 0.02577
16 Jun 14:19:43 ntpdate[7239]: adjust time server 192.38.66.189 offset -0.000240 sec

Right to access the server

In the configuration file there may be some restrict lines, that can make the master unavailable as a clock source. In these configuration all restrict lines are changed to comments.

Firewalls may further prohibit access (consult google for details)

--Jca 12:37, 16 June 2008 (UTC)