Zabbix map status for snmp monitored host is allways green

[panika]

Hello, this problem took me quite some time to figure out. I found a few similar older threads here, but for older versions or stated differently so I decided to state the problem (and solution! )in my own words. I hope this helps somebody.


I'm monitoring many hosts that cant run Zabbix agent (cisco switches etc). So I use the SNMP interface and ICMP template to monitor availability with ping. Triggers work fine, everything is great. I did some maps to have everything in one big picture but to my surprise hosts remained with green "OK" status even when the ping failed. As a temporary solution, I put

{?last(/{HOST.HOST}/icmpping)}

to the description. In the picture you can see one host down one up, but both with OK.


After long fiddling, I changed the host type to trigger and selected Unavailable by ICMP ping, removed not necessary

{?last(/{HOST.HOST}/icmpping)}

from the description and voila:



I guess this could be done a better way, so I'm all ears if anyone knows how. But this is much better than my original solution.

[panika]

Here is a minor update, I discovered that everything works as expected in another map with the same devices. I can't find the difference so I will probably compare the serialization of the maps.