metrics.md

Metrics

Prometheus is used to scrape metrics from the pod. You have to add Prometheus in your NAIS manifest to enable scrape.

Each app that have scraping enabled can use the default Grafana dashboard nais-app-dashboard, or create theire own.

NAIS manifest config

spec:
  prometheus:
    enabled: true  # default: false. Pod will now be scraped for metrics by Prometheus.
    path: /metrics # Path where prometheus metrics are served.

JVM Application

If you are building an app on the JVM you can use Prometheus' own Java client library. Make sure to enable scraping in the NAIS manifest.

We also recommend to export the default metrics.

DefaultExports.initialize();

Push metrics

If you don't want to just rely on pull metrics, you can push data directly to InfluxDB via Sensu.

This is easily done by writing to the Sensu socket.

sensu.nais:3030

Example message:

{
  "name": "myapp_metrics",
  "type": "metric",
  "handlers": ["events_nano"],
  "output": "myapp.event1,tag1=x,tag2=y value=1,value2=2 1571402276000000000\nmyapp.event2,tag1=xx,tag2=yy value=42,value2=69 1571402276000000000"
}

The format of the data (The output field in the message) should be formatted as Influxdb Line Protocol.

{% hint style="warning" %} Note that each variation of tag values will create a new time series, so avoid using tags for data that varies a lot. Read more about best practices here: InfluxDB schema design and data layout {% endhint %}

Overview