This chapter was packed with concepts concerning reacting to events that happen in your monitored environment. We learned how to describe conditions that should be reacted to as trigger expressions. Triggers themselves have useful functionality with dependencies, and we can make them depend on each other. We also explored several ways of reducing trigger flapping right in the trigger expression, including using functions such as min(), max(), and avg(), as well as trigger hysteresis.

Among other trigger tricks, we looked at the following:

• Using the nodata() function to detect missing data
• Using the same nodata() function to make a trigger time out
• Creating triggers that have different used disk space threshold values based on the total disk space

• Creating triggers that only work during a specific time period
• Having a relative threshold, where recent data is compared with the situation some time ago

Trigger configuration has a lot of things that can both make life easier and introduce hard-to-spot problems. Hopefully, the coverage of the basics here will help you leverage the former and avoid the latter.

With the trigger knowledge available to us, we will take the time in the next chapter to see where we can go after a trigger has fired. We will explore actions that will allow us to send emails, or even run commands, in response to a trigger firing.

We have seen the different sorts of events and we have seen how we can configure global event correlation in Zabbix.