metrics.md

Reading metrics from Helm sidecars

NOTE: You can look at all the metrics definitions and what they mean on our metrics reference page

Metric collection on the helm sidecar can be configured in numerous ways, depending on the Prometheus configuration for your Kubernetes cluster. You can set the metrics port by adding the following to your values.yaml file:

containerPorts:
  metrics: 9000 # this is the default value

By default, this port will not be exposed on the Service object created by the helm chart. To enable its exposure, you can add the following to your values.yaml file:

service:
  ports:
    metrics: 9000
  targetPort:
    metrics: metrics

Prometheus configuration

Service Monitor discovery configuration

The sidecar helm chart packages a ServiceMonitor object which can be used in conjunction with a prometheus operator to monitor all pods in the sidecar's Deployment. To enable the service monitor, you can add the following to your values.yaml file:

metrics:
  serviceMonitor:
    enabled: true

NOTE: There are many other configuration options for the ServiceMonitor object, you can look at the default values.yaml file to know all the options.

Annotation based Prometheus discovery configuration

You can add common Prometheus annotations by adding the following to your values.yaml file:

podAnnotations:
  "prometheus.io/scrape": "true"
  "prometheus.io/port": "9000"

NOTE: You can look at configuring Prometheus service discovery for Kubernetes on Prometheus' documentation

Datadog configuration

Datadog metric scraping on Kubernetes can be done in several different ways, and refer to their documentation for a more in depth explanation on Datadog metrics collection on Kubernetes.

Metrics are exposed through the metrics-aggregator container, on the metrics.port port, and are on OpenMetrics format, so a sample annotation you can create by changing your values.yaml file is the following:

podAnnotations:
  ad.datadoghq.com/metrics-aggregator.checks: |
      {
        "openmetrics": {
          "init_config": {},
          "instances": [
            {
              "openmetrics_endpoint": "http://%%host%%:9000/metrics ",
              "namespace": "cyral",
              "metrics": ["cyral*", "up"]
            }
          ]
        }
      }

This example would expose any metrics starting with cyral and the up metric to Datadog, on the cyral namespace.