The alert handler serve as central component for building reactive self-healing systems, based on Prometheus.
It is meant to be a single binary, which will be distribued and installed to all nodes of clusters.
The routing itself is delegated to alertmanager of Prometheus.
It is main functionality is to selfheal and handle Prometheus alerts automatically, which are fired by Prometheus.
The alert-handler share a common declarative API with the alertmanager (based on labels) which need to be respected by users. See API
- build:
go build .
Deploy to the node
systemctl restart prometheus-alertmanager
prometheus-alertmanager
amtool cli for alertmanager:
examples: https://github.com/prometheus/alertmanager#examples
Allocate big file
fallocate -l 4G fake2.img
promtool check rules /path/to/example.rules.yml