nodata trigger

[xs-]

First of all, the nodata trigger is great, really, but it has one downside.
When a zabbix_server has been down for more than the nodata(<time>) all triggers, and related action, fire and our server sends out tons of email and sms alerts . . . which kinda sucks.
The reason for this is that this trigger works independantly of the actual received data. It just compares the timestamp of the last received item value, not the value.

So my question is; does anyone now of a way (or is using a method) to not have the nodata triggers fire on zabbix_server start for at least 1x or 2x the poll interval of the item its for?
(or another way to prevent the nodata triggers from going off at the wrong time).

[bbrendon]

I just manually disable all actions.

You could add a mysql command to your init script to do this for you automatically as well. I've thought about doing this, but haven't...

[Alexei]

The nodata triggers are not supposed to be activated after ZABBIX server restart, at least in 1.4.4 and later releases. Otherwise it is a bug.

[xs-]

For example, the following will cause actions to be sent when i stop the zabbix server for 600+ secs and start it again:
({template:agent.ping.last(0)}#1)|({template:agent .ping.nodata(600)}=1)
the item is an active check.

[bbrendon]

The nodata triggers are not supposed to be activated after ZABBIX server restart, at least in 1.4.4 and later releases. Otherwise it is a bug.

Then I'm not sure. I'm so use to disabling them, its automatic. I don't want 100 txt messages when I start the server.
...I'll wait for someone else to test this. I'm not brave enough.