Minor POC to test time based metrics alternatives

Like a cake recipe.

1. Simple Kotlin/Spring Boot app
2. Prometheus server up and running
3. Grafana container (because install in Mac sucks).

After a working rest endpoint just add those dependencies:

dependencies {
  // ...
  implementation("io.micrometer:micrometer-core")
  implementation("io.micrometer:micrometer-registry-prometheus")
  // ...
}

and this parameters to application.properties:

#Metrics related configurations
management.endpoint.metrics.enabled=true
management.endpoints.web.exposure.include=*
management.endpoint.prometheus.enabled=true
management.metrics.export.prometheus.enabled=true

This will add endpoints to actuar and expose metrics through a rest api.

In the code, create objects to collect the metrics like:

@Component
class PrometheusMetrics(@Autowired val meterRegistry: MeterRegistry) {
   private final val callCounter : Counter = Counter
             .builder("echocounter")
             .description("this is an echo counter")
             .register(meterRegistry)

    fun count() {
        println("Count!")
         callCounter.increment()
    }
}

once count() is called, it will add ONE in the prometheus metrics.

Download Prometheus and create a config file as bellow:

#App Config file
#Global configurations
global:
  scrape_interval:     5s # Set the scrape interval to every 5 seconds.
  evaluation_interval: 5s # Evaluate rules every 5 seconds.
scrape_configs:
  - job_name: 'echo-app'
    metrics_path: '/actuator/prometheus'
    static_configs:
      - targets: ['localhost:8080']

now start it ./prometheus --config.file=app-config.yml.

Start with docker-compose and add a new Prometheus Datasource. (simple right?)

Alex Rocha - about.me