Host Agent to monitor Kubernetes.

Cette page n'est pas encore disponible en français, sa traduction est en cours.
Si vous avez des questions ou des retours sur notre projet de traduction actuel, n'hésitez pas à nous contacter.

Setup the Kubernetes state integration

If you decide to run your Agent on your host to monitor kubernetes, in order to start collecting your Kubernetes metrics you should:

  1. Install the Agent on your host

  2. Download the Kube-State manifests folder.

  3. Apply them to your Kubernetes cluster:

    kubectl apply -f <NAME_OF_THE_KUBE_STATE_MANIFESTS_FOLDER>

Once done, the Kubernetes State integration automatically collects kube-state metrics.

Setup the kube-dns integration

Configuration

Since Agent v6, Kubernetes DNS integration works automatically with the Autodiscovery.

Note: these metrics are unavailable for Azure Kubernetes Service (AKS).

Collect container logs

Available for Agent >6.0

There are two ways to collect logs from containers running in Kubernetes:

  • Through the Docker socket.
  • Through the Kubernetes log files.

Datadog recommends using the Kubernetes log files approach when you are either not using Docker, or are using more than 10 containers per pod.

Datadog also recommends that you take advantage of DaemonSets to automatically deploy the Datadog Agent on all your nodes. Otherwise, to manually enable log collection from one specific node, add the following parameters in the datadog.yaml:

logs_enabled: true
listeners:
  - name: kubelet
config_providers:
  - name: kubelet
    polling: true
logs_config:
  container_collect_all: true

Restart the Agent.

Use Autodiscovery with Pod Annotations to configure log collection to add multiline processing rules, or to customize the source and service attributes.

Further Reading

To get a better idea of how (or why) to integrate your Kubernetes service, see the related series of Datadog blog posts.