Awesome Open Source
Awesome Open Source

Build Status GoDoc Go Report codecov

metrics - lightweight package for exporting metrics in Prometheus format


  • Lightweight. Has minimal number of third-party dependencies and all these deps are small. See this article for details.
  • Easy to use. See the API docs.
  • Fast.
  • Allows exporting distinct metric sets via distinct endpoints. See Set.
  • Supports easy-to-use histograms, which just work without any tuning. Read more about VictoriaMetrics histograms at this article.



import ""

// Register various time series.
// Time series name may contain labels in Prometheus format - see below.
var (
	// Register counter without labels.
	requestsTotal = metrics.NewCounter("requests_total")

	// Register summary with a single label.
	requestDuration = metrics.NewSummary(`requests_duration_seconds{path="/foobar/baz"}`)

	// Register gauge with two labels.
	queueSize = metrics.NewGauge(`queue_size{queue="foobar",topic="baz"}`, func() float64 {
		return float64(foobarQueue.Len())

	// Register histogram with a single label.
	responseSize = metrics.NewHistogram(`response_size{path="/foo/bar"}`)

// ...
func requestHandler() {
	// Increment requestTotal counter.

	startTime := time.Now()
	// Update requestDuration summary.

	// Update responseSize histogram.

// Expose the registered metrics at `/metrics` path.
http.HandleFunc("/metrics", func(w http.ResponseWriter, req *http.Request) {
	metrics.WritePrometheus(w, true)

See docs for more info.



Why the metrics API isn't compatible with

Because the is too complex and is hard to use.

Why the metrics.WritePrometheus doesn't expose documentation for each metric?

Because this documentation is ignored by Prometheus. The documentation is for users. Just give meaningful names to the exported metrics or add comments in the source code or in other suitable place explaining each metric exposed from your application.

How to implement CounterVec in metrics?

Just use GetOrCreateCounter instead of CounterVec.With. See this example for details.

Why Histogram buckets contain vmrange labels instead of le labels like in Prometheus histograms?

Buckets with vmrange labels occupy less disk space compared to Promethes-style buckets with le labels, because vmrange buckets don't include counters for the previous ranges. VictoriaMetrics provides prometheus_buckets function, which converts vmrange buckets to Prometheus-style buckets with le labels. This is useful for building heatmaps in Grafana. Additionally, its' histogram_quantile function transparently handles histogram buckets with vmrange labels.

Get A Weekly Email With Trending Projects For These Topics
No Spam. Unsubscribe easily at any time.
go (15,135
metrics (346
prometheus (323
fast (207
lightweight (197