Grafana Cheatsheet
Top dashboard tips.
Overview
The Grafana cheatsheet captures the dashboard patterns operators actually use during investigation. Panel choice is the easy part; variables, repeating panels, and provisioning are what make dashboards useful at team scale.
- Variables, repeating panels, drill-down. The three patterns that turn one dashboard into a per-service, per-tenant view.
- Variables. Per-dashboard filters (
$service,$namespace,$env); one dashboard serves many contexts. - Repeating panels. One panel per service, generated automatically from a variable; the dashboard scales with the inventory.
- Annotations plus provisioning. Mark deploys and incidents on the chart; ship dashboards as code so they survive moves and audits.
The approach
The practical approach: variable-driven, repeating panels, dashboards as code, annotations on deploys. The team’s discipline produces dashboards engineers actually open during incidents.
- Variables for filtering.
$service,$namespace,$env; one dashboard reuse across services and environments. - Repeating panels. One row per service generated from a variable; the dashboard grows with the service catalog automatically.
- Annotations on deploys. CI publishes deploy annotations; the chart shows "what changed" at the moment behaviour shifted.
- Provisioning via Helm or Terraform. Dashboards live in git; reviewable, recreatable, audit-friendly.
- Document the dashboard. Per-dashboard purpose committed to the repo; supports operational reviews and onboarding.
Why this compounds
Grafana fluency compounds across services. Each captured pattern grows the team’s investigation capability; cost-per-question falls as the dashboard library matures.
- Faster investigation. Right dashboard supports fast root cause; queries return in milliseconds, decisions in seconds.
- Better reuse. Variable-driven dashboards scale across services; one dashboard serves the whole inventory.
- Better correlation. Annotations link deploys to behaviour; "what changed" becomes visible at the moment of regression.
- Institutional knowledge. Each dashboard teaches monitoring patterns; the team’s observability muscle grows.
Grafana fluency is an operational discipline that pays off across years. Nova AI Ops integrates with Grafana telemetry, surfaces patterns, and supports the team’s observability discipline.