Runai Container Restarting

Meaning¶

A Run:ai container has restarted more than twice in the last 10 minutes.

Impact¶

The container may be unavailable and affect Run:ai system feature functionalities.

Severity¶

Warning

Diagnosis¶

To diagnose the issue, and the pods with issues, run:

kubectl get pods -n runai
kubectl get pods -n runai-backend

The expected result should be one or more pods where the restart count >= 2.

Mitigation¶

Run kubectl logs -n NAMESPACE POD_NAME on the relevant pod. Check to see if there is something in the logs that stands out. Then, check that the container has enough resources.

Contact Run:ai for more assistance.