Trigger: "{node:status.last(0)}=2" What does this check?

[ptader]

There is a trigger defined in the Template_linux that is configured like so:

Name: Server {HOSTNAME} is unreachable
Expression: {Template_Linux:status.last(0)}=2

All my systems with this Template applied to it are "grey" in the Monitoring / Overview web page. What is this trigger, triggering on?

I am getting data from other checks on the same nodes.

The document reads:

"# Example 7. Server is unreachable.

{zabbix.zabbix.com:status.last(0)}=2

Note:Note ‘status’ is special parameter which is calculated if and only if corresponding host has at least one parameter for monitoring. See description of ‘status’ for more details. "

I can't find any more references detailing the `status` parameter so any pointer will be helpful.


Thanks,
Tader

[jeenam]

dur

Status = ping

The host is pinged.

[ptader]

Hmm, are you sure? All other checks are succeeding (CPU, network, memory, etc.) and all the host(s) are ping-able from the Zabbix server at the time when the Overview screen reports the grey box.

Thanks,
Paul

[jeenam]

dur

There seems to be a bug related to this trigger. It does not always work when adding a host, and there have been times when it does not work after un-linking and then re-linking a host to a template.

The fix I found was to clone the host, give it a unique name, then check the overview page to verify the STATUS monitor was flashing green. If the STATUS monitor of the clone is not green then I create another clone. If after attemping to create 5 clones the STATUS box still does not flash, I restart zabbix and mysql on the server, then attempt the process over again. Once a cloned host actually works with the trigger, rename it to it's final desired name.

After searching the forums no one else seemed to report this issue, but it is probably very common. It's an aggravating bug as I just spent an hour this morning performing this very process to 'fix' a bunch of hosts that were accidentally removed from a template.

Hope this helps you.

[rolandsym]

status item

It's common. I don't use the status item because it doesn't really change or even update that often. If the zabbix client stops responding or sending info the status item doesn't change(usually). So it's bung in my book. I use a trigger based off of the nodata() function to check if the zabbix client died and a ping status to check if it's up.

If someone can tell me a better method I'm all ears.

Rolandsym