0

I have the following ntp check on my server for Nagios via NRPE

/usr/local/nagios/nrpe_plugins/check_ntp_time -H localhost -w 0.5 -c 1 -q

When changing the date manually via the date command to a date far in the past:

date --set="16 JAN 2017 15:00:00"

The check is always giving an OK state output

NTP OK: Offset -1.442432404e-05 secs|offset=-0.000014s;0.500000;1.000000;

Despite the offset 

remote           refid      st t when poll reach   delay   offset  jitter
==============================================================================
 10.7.100.65     176.58.109.199   3 u 4971   64    7    0.246  6136468 4339138
 10.8.100.65     188.114.116.1    3 u 4971   64    7    1.571  6136468 4339138
 10.7.91.41      217.114.59.66    3 u 4971   64    7    0.256  6136468 4339138

I don't understand the behavior here.

I am using the latest version of the check_ntp_time, but I have the same behavior with other versions as well.

check_ntp_time v2.2.0 (nagios-plugins 2.2.0) Copyright (c) 2006 Sean Finney Copyright (c) 2006-2014 Nagios Plugin Development Team

Thank you in advance for your help,

Virtuose
  • 537
  • 1
  • 6
  • 13

1 Answers1

2

From the check_ntp_time --help output:

This plugin checks the clock offset between the local host and a remote NTP server.

This plugin compares the time on localhost with the time on the host specified via -H. When you run it as check_ntp_time -H localhost, you're comparing it to itself. (This is why the difference is nearly zero.)

It seems like you probably want check_ntp_peer instead.

Alternatively, you can just use check_ntp_time to compare remote hosts (that run NTP servers) against your Nagios box, or run it via NRPE to compare those hosts against some other valid NTP server (e.g., time.nist.gov, time.apple.com, pool.ntp.org, whatever).

Keith
  • 4,637
  • 15
  • 25