/prometheus-example-app

Go app that exposes metrics about its HTTP handlers.

Primary LanguageGoMIT LicenseMIT

Prometheus Example App

This example app serves as an example of how one can easily instrument HTTP handlers with Prometheus metrics. It uses the Prometheus go client to create a new Prometheus registry.

Usage is simple, on any request to / the request will result in a 200 response code. This increments the counter for this response code. Similarly the /err endpoint will result in a 404 response code, therefore increments that respective counter. Duration metrics are also exposed for any request to /.

A Docker image is available at: quay.io/brancz/prometheus-example-app:v0.3.0

Deploying in a Kubernetes cluster

First, deploy one instance of this example application, which listens and exposes metrics on port 8080 using the following Deployment manifest.

Then, in order for Prometheus to be able to discover and scrape these metrics, in this repository we provide Pod Monitor custom resource as an example. For that, Prometheus Operator needs to be running in the cluster. To install it, follow this guide: quickstart.

For this example application, PodMonitor manifest describes how the metrics can be discovered and scrapped by Prometheus.

Exposed Prometheus metrics

The following metrics are exposed:

  • version - of type gauge - containing the app version - as a constant metric value 1 and label version, representing this app version
  • http_requests_total - of type counter - representing the total numbere of incoming HTTP requests
  • http_request_duration_seconds - of type histogram, representing duration of all HTTP requests
  • http_request_duration_seconds_count- total count of all incoming HTTP requeests
  • http_request_duration_seconds_sum - total duration in seconds of all incoming HTTP requests
  • http_request_duration_seconds_bucket - a histogram representation of the duration of the incoming HTTP requests

The sample output of the /metric endpoint after 5 incoming HTTP requests shown below.

Note: with no initial incoming request, only version metric is reported.

# HELP http_request_duration_seconds Duration of all HTTP requests
# TYPE http_request_duration_seconds histogram
http_request_duration_seconds_bucket{code="200",handler="found",method="get",le="0.005"} 5
http_request_duration_seconds_bucket{code="200",handler="found",method="get",le="0.01"} 5
http_request_duration_seconds_bucket{code="200",handler="found",method="get",le="0.025"} 5
http_request_duration_seconds_bucket{code="200",handler="found",method="get",le="0.05"} 5
http_request_duration_seconds_bucket{code="200",handler="found",method="get",le="0.1"} 5
http_request_duration_seconds_bucket{code="200",handler="found",method="get",le="0.25"} 5
http_request_duration_seconds_bucket{code="200",handler="found",method="get",le="0.5"} 5
http_request_duration_seconds_bucket{code="200",handler="found",method="get",le="1"} 5
http_request_duration_seconds_bucket{code="200",handler="found",method="get",le="2.5"} 5
http_request_duration_seconds_bucket{code="200",handler="found",method="get",le="5"} 5
http_request_duration_seconds_bucket{code="200",handler="found",method="get",le="10"} 5
http_request_duration_seconds_bucket{code="200",handler="found",method="get",le="+Inf"} 5
http_request_duration_seconds_sum{code="200",handler="found",method="get"} 0.00047495999999999997
http_request_duration_seconds_count{code="200",handler="found",method="get"} 5
# HELP http_requests_total Count of all HTTP requests
# TYPE http_requests_total counter
http_requests_total{code="200",method="get"} 5
# HELP version Version information about this binary
# TYPE version gauge
version{version="v0.3.0"} 1