Monitoring and Observability: Prometheus, Alertmanager and Grafana
Exit

Lesson 8 of 8 · Production

On-call and incident management: runbooks and post-mortems

40 min Debian 12

Objective

Make sure that an alert received at 3 a.m. quickly leads to the right action: an organized on-call rotation, a written procedure for each alert, clear roles during the incident, then a post-mortem that prevents the problem from coming back.

This lesson is part of the full course

Unlock Monitoring and Observability: Prometheus, Alertmanager and Grafana to read every lesson, do the exercises, access the simulations and earn your certificate.

Already enrolled? Log in.