Tidy up our Prometheus metrics!
In this article I’ll explain my process of analyzing and setting up Prometheus, where I maybe effectively increase query performance and addressed challenges related to cardinality. FYI, I use this stack on my monitoring system: Grafana, Promethues, Thanos. The metrics itself gethered from several source: node exporter, kube-state-metrics, etc.
1. List unused metrics Firs, we need to list metrcis that used in our Grafana. To do this task, we can use mimirtool.
