check_chrony_stratum queries a Chrony server using its management protocol for its current stratum.
check_chrony_peer queries a Chrony server using its management protocol for the status of its peers.
./check_chrony_stratum -H <hostname> [-p <port>] [-W <warn>] [-C <crit>]
The server is chosen using the -H <hostname> option, which may be a hostname or an IP address.
-p <port> sets the port that should be contacted. Note that this is not the regular NTP port but the port for chronyd's management interface, also used by chronyc. By default, the port is 323.
-W <warn> and -C <crit> set the thresholds for reporting a too-high stratum. If the reported stratum is greater than or equal to the given threshold, the script reports the respective (warning or critical) state. Both options are optional; if none is given, the script always reports an OK state. If the criteria for both warning and critical states are met, critical takes precedence; if both are to be used, it makes sense for the critical threshold to be greater than the warning threshold.
./check_chrony_peer -H <hostname> [-p <port>] [thresholds]
The server is chosen using the -H <hostname> option, which may be a hostname or an IP address.
-p <port> sets the port that should be contacted. Note that this is not the regular NTP port but the port for chronyd's management interface, also used by chronyc. By default, the port is 323.
The following thresholds can be specified:
| measurement | crit short | crit long | warn short | warn long | default | description |
|---|---|---|---|---|---|---|
| offset | -c |
--ocrit |
-w |
--owarn |
w=0:60 c=0:120 | the absolute time offset between Chrony and its chosen peer (in seconds) |
| jitter | -k |
--jcrit |
-j |
--jwarn |
none | the standard deviation of network packet transmission durations between Chrony and the chosen peer |
| stratum | -C |
--scrit |
-W |
--swarn |
none | the stratum of the chosen peer |
| truechimers | -n |
--tcrit |
-m |
--twarn |
none | the number of choosable peers |
If a threshold is not specified, the default threshold for that measurement is taken. If there is no default threshold for a measurement, that measurement is ignored when determining the overall status.
Exit codes match those expected of Nagios/Icinga check plugins:
- 0 = OK
- 1 = warning (
check_chrony_stratum: reported stratum >= stratum warning threshold) - 2 = critical (
check_chrony_stratum: reported stratum >= stratum critical threshold) - 3 = unknown (an error occurred while querying)
For check_chrony_peer, full thresholds (minimum and maximum) are specified; the given status is reported if the value lies outside of the boundaries given by each threshold value pair.