Grafana Documentation
The official docs for Grafana, the open source dashboard and alerting tool most teams put in front of Prometheus, Loki and Tempo. Learn data sources, panels, dashboard variables, Explore, alerting and provisioning dashboards as code.
77k (grafana/grafana)
GitHub · Sept 2026
v13.2.2
GitHub · Sept 2026
AGPL-3.0
GitHub · Sept 2026
Guide
Other
Grafana Labs
A weekend to get started, reference after that
Beginner
Free
What Grafana is
Grafana describes itself as an open and composable observability and data visualisation platform. In practice it is the screen most engineering teams look at during an incident: dashboards of metrics from Prometheus, logs from Loki, traces from Tempo or Jaeger, and data from SQL databases and cloud monitoring services, all side by side. It does not store the data itself; it queries the systems that do.
What the docs cover
- Getting started: installing Grafana locally or in Docker, adding a first data source and building a first dashboard.
- Data sources: connecting Prometheus, Loki, Tempo, PostgreSQL, MySQL, Elasticsearch, CloudWatch and many more.
- Dashboards and panels: time series, stat, gauge, table, bar chart and heatmap visualisations, units, thresholds, transformations and dashboard variables.
- Explore: querying without building a dashboard, and jumping between metrics, logs and traces.
- Alerting: alert rules, contact points and notification policies.
- Administration: users, teams, roles and permissions, and provisioning data sources and dashboards from files.
Is it worth learning?
Yes, if you run or support anything in production. Knowing how to read and build a good dashboard is expected of backend, DevOps and SRE engineers, and Grafana is one of the most common tools for it. The skills transfer: once you understand panels, variables and units here, most other dashboard tools feel familiar.
Prerequisites and time
You need a data source to point it at. The easiest start is Docker Compose with Prometheus, node_exporter and Grafana on your laptop. Basic PromQL helps a lot, since most panels are Prometheus queries. The getting started guide takes an evening; building useful dashboards for a real service with variables and alerts takes a weekend or two.
Free or paid?
Grafana open source is free to self-host under the AGPL-3.0 licence. Grafana Labs also sells Grafana Cloud, a hosted version that includes an always-free plan with limited usage, and an Enterprise edition with extra features. The documentation for the open source edition is free and covers everything you need to learn.
Common questions
Do I need Prometheus to use Grafana? No, Grafana works with many data sources, but Prometheus is the most common pairing for metrics and the one most tutorials use.
Grafana alerting or Prometheus Alertmanager? Both work. Teams already running Prometheus often keep alert rules in Prometheus and Alertmanager; Grafana alerting is convenient when alerts span several data sources.
What can I do after it? Build RED and USE dashboards for a service, use variables so one dashboard serves every environment, set up alerts with sensible routing, and keep dashboards in Git through provisioning.
Best for: backend, DevOps, platform and SRE engineers, and students building a monitoring stack for a final year project.
Ready to start?
Opens on Grafana Labs in a new tab.
Stuck on something specific?
Writing only gets you so far. If you want an answer to your situation rather than the general case, book a session and we will work through it together. Sessions are free for approved Sefism members, and a few slots open each week.
Follow along
New writing, resources and project ideas land here first.