Nagios Support Forum

Hi all,

As I expand my Nagios setup, I’m starting to face a challenge with setting alert thresholds particularly for metrics like CPU load, memory usage, and network traffic.

I’m wondering if there are any plugins or tools that help with threshold tuning maybe something that suggests sensible default values or even learns from historical data to reduce false positives.

Has anyone used a plugin or method that helps fine-tune thresholds more intelligently rather than relying on static values?

I don't know of any. I usually set values at a percentage and use templates to set what I can to standard values. For example linux hosts will all have CPU usage values set to 80% and 90% (or whatever).
If you have defined things so you have many cpu (or whatever) checks for each linux host something is wrong.
As far as false positives go, it a value is over/under/whatever what's been defined and it alerts, it isn't a false positive--the system is doing what it was told to do.