OpenShift Lightspeed Demo 14: Troubleshooting worker node's container runtime CRI-O error
In this video, we simulate a severe platform incident by intentionally paralyzing the CRI-O container runtime on a worker node to trigger a KubeNodeNotReady alert. Pods get stuck in Terminating and Pending states, effectively blinding the kubelet.
Then we turn to OpenShift Lightspeed as our Senior SRE. Watch as it translates the alert, pinpoints the unresponsive container runtime as the root cause, and provides the exact node-level commands to clear the deadlock and restore the node to a healthy Ready state.