NTP Monitoring

By default, LogicMonitor will attempt an NTP query against all Linux hosts to determine whether their time is synchronized against an NTP server. Proper time synchronization is very important in most data centers, and easy to set up.

Your monitored host should be running NTPd or XNTPd, and be configured to get its time from an NTP server. It should also be configured to allow the collector host to query it via NTP.

Note that the NTP datasources query the devices to which they are applied, not the NTP client, itself. For troubleshooting purposes, this is important to distinguish due to the fact that an NTP client may be functioning properly, but the machine on which it runs may not.

If NTP is correctly configured but the collector still cannot monitor NTP status, a firewall or host based firewall may be the issue. Ensure UDP port 123 is open between the collector and the monitored device.

Sample ntp.conf:

driftfile /var/lib/ntp/drift/ntp.drift
# path for drift file
logfile /var/log/ntp
server ntp1.cs.wisc.edu
server ntp1.tummy.com
server ntp2.usno.navy.mil
#If you wish to impose ntp access restrictions,
#note they need to be  done via numeric IP, not by name
#If so, uncomment following lines.
#restrict default ignore
#restrict mask
#restrict mask
#restrict mask
#restrict mask

For more information about NTP configuration on Linux devices, see this page.