NTP monitoring does not work

[jdmonsalvel]

Hey there
I have a problem to monitor the NTP service from a Linux server, the service is working, but the Monitor says that the service is down .

I have the agent installed and working with other services.

you must perform additional configuration for monitoring this service?

Thanks

[LenR]

What is the item key?

[peterloron]

I am running into this as well. I'm using the built in (Zabbix 3.2) NTP template which uses an item key of "net.udp.service[ntp]".

I can run zabbix_get with that key and host on the zabbix server, and it is returning the correct value for the state of the remote service. In the UI monitor (latest data) it is always 0.

Any suggestions?

Thanks.

[LenR]

net.udp.service is a simple check, it doesn't use the agent. You could use tcpdump and wireshark to capture the udp traffic between the zabbix server (or proxy) running the test and the client to verify the zabbix test result. nmap might work to debug as wel.

I really don't know what zabbix_get does with a simple check on the agent. It appears that its running the test, and that the service is "available" to the agent on the client, or zabbix_get may be returning invalid results since it's a simple check.

If zabbix_get is valid from the client view, that could be the result of a local firewall or ntp access rules.

[peterloron]

I'll capture some traffic and see if that reveals anything.

This check is running fine with many (> 100) hosts scattered across various AWS regions. Other hosts in the same VPC and subnet are working fine with this check as the problem host.