Zabbix Agent unreachable flapping

[tinus]

Hi,

We are Using zabbix (2.0.7) to monitor a large group of hosts (200+) and amongst those are svereal servers which have bad uplinks.

We get a LOT of emails about the host beeing unreachable and a few seconds after that the OK because the host is reachable again... and so on.

I have been playing around a bit with the triggers, conditions and examples on the forum, it pretty much drove me insane.

The standard trigger definition is :

{Template App Zabbix Agent:agent.ping.nodata(300)}=1

I toyed around with a lot of suggestions but the system keeps mailing down and up notfies like a raving madman (even when i clear the escalation table).

What can i do about the flapping of the agent alive?

[Colttt]

use min() instead!

[tinus]

Correct

Really have a hard time contructing it :S


{Template App Zabbix Agent:agent.ping.nodata.min(300)}=1 would be ok?


Thanks!

[tinus]

Can't seem to get the trigger definition right.....

Want to extend it so it polls and only triggers when 5 minutes down, and only give an OK if its back for 10 minutes....

And there goes my sleep

[vlam]

Hi

Did you get this sorted and if so how because I have a similar problem

Thanks

[steveboyson]

You may want to have a look at Zabbix' hysteresis definition.

https://www.zabbix.com/documentation/2.0/manual/config/triggers/expression

[alp]

Really have a hard time contructing it :S
{Template App Zabbix Agent:agent.ping.nodata.min(300)}=1 would be ok?
Thanks!

When i set it, zabbix get error:
Incorrect item key "agent.ping.nodata" provided for trigger expression on "Template App Zabbix Agent".

How can i fix unreacheble flapping?