We have a group of Linux servers running SUSE Linux and we would like to monitor the NTP status. If we manually login to the servers we would do something like:

# ntpq -nc peers
which would return something like this:
     remote           refid      st t when poll reach   delay   offset  jitter
==============================================================================
*000.000.000.000     000.000.000.1     2 u   47  128  377    1.271    0.575   0.124

We can then see the offset and see if it is excessive or not.

We tried adding the SNTP sensor to one server as a test to see how it went, but it returns with a sensor timeout error.

Can you confirm that we should be able to monitor linux servers with the SNTP sensor. If not, can you confirm the best way in which to monitor NTP, I am guessing via some sort of custom SSH script sensor?

Thanks

Article Comments

Hello,

thank you very much for your KB-Post. If the normal SNTP-Sensor does not work with the Suse-Linux Servers in this case, we would indeed recommend, using a self-written script, that is then used by the SSH Script Sensor. You could have the sensor then show the time offset in PRTG.

best regards.


May, 2014 - Permalink

You are getting the session time out error because NTP on linux is configured to ignore external query requests by default.

You can read the documentation http://support.ntp.org/bin/view/Support/AccessRestrictions#Section_6.5.1.

Edit the /etc/ntp.cfg file and add a line to allow the PRTG server to query the NTP server.

Replace the IP with your IP address IPv4: restrict 192.168.1.10

Hope this helps,


Jul, 2015 - Permalink