Check-Manual von kube_pod_status

Kubernetes: Pod Status

Distribution	Official part of Checkmk
Lizenz	GPLv2
Unterstützte Agenten	Kubernetes

This check provides the pod status. The status of a pod consist of a single word summarizing the individual statuses of containers as well as the pod phase. Specifically, if a container is faulty, then the status corresponds to the reason given by the container. If there is no such container, then the status corresponds to the pod phase. If multiple containers meet the criterion, only one container is selected for the status message. A container is considered faulty, if it is waiting for a reason other than ContainerCreating or if it has terminated with a non-zero exit code.

If a pod does not achieve the status Running or Succeeced for longer than 5/10 minutes the service goes WARN/CRIT. This can be configured via the rule Kubernetes pod status. Note, the timer begins with the first time a status is seen by checkmk, not on data provided by the Kubernetes API.

The check requires that the special agent kube is configured.

Discovery

One service is created for each pod.