What does CPU Throttling chart mean in the Dashboard ?

Solution Verified - Updated -

Environment

  • Red Hat OpenShift Service on AWS
    • 4.x
  • Red Hat OpenShift Container Platform
    • 4.x

Issue

  • What does CPU Throttling chart mean at Pod section of the Dashboard in web console ?
  • You can see the above chart in the web console with Administrator mode as follows.
    • Move Observe -> Dashbords, and select the "Kubernetes/Compute Resources/Pod" in the Dashboard drop box there.

Resolution

  • The percent of the CPU time that is in throttling status within all the CPU time for the pods. A high ratio might indicate that the pod could not take the CPU time enough to process the workload.
  • Refer the following PromQL for the CPU Throttling chart for more details. You can replace the TARGET_POD_NAME and TARGET_PROJECT_NAME with each one for your env appropriately.
sum(increase(container_cpu_cfs_throttled_periods_total{job="kubelet", metrics_path="/metrics/cadvisor", namespace="TARGET_PROJECT_NAME", pod="TARGET_POD_NAME", container!="", cluster=""}[5m])) by (container)
/
sum(increase(container_cpu_cfs_periods_total{job="kubelet", metrics_path="/metrics/cadvisor", namespace="TARGET_PROJECT_NAME", pod="TARGET_POD_NAME", container!="", cluster=""}[5m])) by (container)
  • The metrics value means a ratio. For example, each metric value means that 1 is 100% and 0.5 is 50%.

This solution is part of Red Hat’s fast-track publication program, providing a huge library of solutions that Red Hat engineers have created while supporting our customers. To give you the knowledge you need the instant it becomes available, these articles may be presented in a raw and unedited form.

Comments