← Docs
Recipe

Prometheus + Grafana

Metrics that matter. Dashboards that don't suck.

Prometheus scrapes time-series metrics from your apps. Grafana turns those numbers into panels you actually want to look at. Together they form the backbone of every serious observability stack.

The stack

  • ▸Prometheus — pull-based TSDB, PromQL, alerting rules
  • ▸Node Exporter — host-level CPU, memory, disk, network
  • ▸Grafana — dashboards, panels, variables, alerting UI

Why this combo

Prometheus stores metrics efficiently with automatic compaction and retention. Its query language PromQL lets you slice by label, compute rates, and join series. Grafana connects to Prometheus as a data source and gives you drag-and-drop panel builders, template variables, and shareable dashboard JSON.

Quick PromQL

rate(http_requests_total[5m])
histogram_quantile(0.99, rate(request_duration_bucket[5m]))
up == 0

Nimbus integration

Expose a /metrics endpoint from your loader or dashboard backend. Track license validations, injection latency, and heartbeat failures. Grafana dashboards ship with every Nimbus enterprise plan.

Back to docsLoki + Tempo →