Same item, multiple triggers w/ different thresholds

[ibtanhe]

Hi all!

We're currently migrating our Nagios-setup to Zabbix, and I'm trying to find a clean way of implementing our diskcapacity-checks.

I want to have a lower threshold which only alerts via email, when capacity usage is >= 90% (representing WARNING in Nagios).

I want to have a higher threshold which alerts via both email and SMS when capacity usage is >= 95% (representing CRITICAL in Nagios).

Using an example of constant increasing capacity-usage on a filesystem, it reaches 90% and starts sending emails. When reaching 95%, I don't want any kind of recovery from the 90%-trigger as it's misleading (if it decreases below 90%, recovery is wanted). At 95%, I just want the 95%-trigger to start alerting via email/SMS. Recovery from the 95%-trigger (going below 95%) is wanted.

Is this possible somehow? I was thinking of trying out something with dependencies, but it was greatly complicated by the fact that these triggers are created during discovery, and I found no way of setting up deps there.

Cheers,
-- Andy

[pc99096]

it is possible, check this:

https://www.zabbix.com/documentation/2.0/manual/config/notifications/action

[Boghdan]

1. It's better to monitor actual disk space available than value in percentage. Depending on total disk capacity the "Critical" <5% (in your case) left on disk may mean 1GB or these can be 15GB, which would not be that critical.

If you need to monitor only system disks (C: on windows or / on linux if not divided), you may create OS specific templates with an item monitoring disk space left and two triggers (Warning and High or Critical) with different thresholds. Less critical would have a dependency configured towards the more critical one. Actions (email, sms) will be configured either for these triggers specifically or for all triggers from a group with specific status)

[ibtanhe]

1. It's better to monitor actual disk space available than value in percentage. Depending on total disk capacity the "Critical" <5% (in your case) left on disk may mean 1GB or these can be 15GB, which would not be that critical.

For us, percentages is better. But that's beside the point.

If you need to monitor only system disks (C: on windows or / on linux if not divided), you may create OS specific templates with an item monitoring disk space left and two triggers (Warning and High or Critical) with different thresholds. Less critical would have a dependency configured towards the more critical one. Actions (email, sms) will be configured either for these triggers specifically or for all triggers from a group with specific status)

Well, that was my idea as well, but as I wrote earlier, I found no way of setting up trigger-dependencies in discovery. And bypassing discovery it not really an option for us.

Cheers,
-- A.

[sudo092]

I know this is an old thread but for the benefit of anyone else that comes across this thread like i did looking for a way to specify thresholds for the same item in a low level discovery setup....

You can specify more than one condition in a single trigger prototype expression using the "and" operator.

Something that i've done in the past is have a single trigger prototype in a template for a range of values between two thresholds.

{Template Name:item.key[{#MACRO}].last(0)}>=5 and {Template Name:item.key[{#MACRO}].last(0)}<=9

Then a second trigger prototype will have an expression for another range of values greater than the range above, like this...

{Template Name:item.key[{#MACRO}].last(0)}>=10 and {Template Name:item.key[{#MACRO}].last(0)}<=14

and so on... with this, you can specify as many trigger thresholds as you want.

Then you can specify which of the triggers above create alerts in Actions.