Payment by bank transfer · Moroccoتعلَّم · مارِس · أتقِن

Catalogue / DevOps

Monitoring and Observability: Prometheus, Alertmanager and Grafana

Measure, alert on and troubleshoot your servers and applications with Prometheus, PromQL, Alertmanager, Grafana and centralized logs, all the way to on-call duty.

  • Beginner to advanced
  • 8 lessons
  • 5 h 30 min

A MasterLinux course · Original text

What you will learn

  • Tell metrics, logs and traces apart, and choose the right signal
  • Define SLIs and SLOs and calculate an error budget
  • Install Prometheus and node_exporter, and write prometheus.yml
  • Query metrics with PromQL (rate, sum by, histograms)
  • Write alerting rules and route notifications with Alertmanager
  • Build provisioned, parameterized Grafana dashboards
  • Centralize logs with Loki and query them with LogQL
  • Monitor Docker containers and a Kubernetes cluster
  • Organize on-call duty, write runbooks and post-mortems

Course content

8 lessons · 5 h 30 min
Section 1 — Fundamentals3 lessons · 1 h 55 min
  1. Observability: metrics, logs, traces, SLIs and SLOs Preview30 min
  2. Prometheus: architecture, scraping and node_exporter 40 min
  3. PromQL: querying metrics 45 min
Section 2 — Alerts and dashboards3 lessons · 2 h 5 min
  1. Alerting: alerting rules and Alertmanager 45 min
  2. Grafana: data sources, dashboards and variables 40 min
  3. Centralized logs: Loki, Promtail and the ELK stack 40 min
Section 3 — Production2 lessons · 1 h 30 min
  1. Monitoring Docker and Kubernetes: cAdvisor, kube-state-metrics, Prometheus Operator 50 min
  2. On-call and incident management: runbooks and post-mortems 40 min

Requirements

  • The “Fundamentals” course (included) or solid hands-on experience with the Linux command line
  • Able to read and write a YAML file

Who is this course for?

  • System administrators and operations technicians
  • Junior DevOps engineers and SREs
  • Developers who want to measure their applications in production