QuestionQ98

VerifiedImplementing observability practices and troubleshooting issues

Your company operates applications in Google Kubernetes Engine (GKE). Several applications depend on ephemeral volumes. You observed that some applications were unstable because of the DiskPressure node condition on worker nodes. You must identify which Pods are causing the problem, but you do not have execute access to workloads or nodes. What should you do?

Explanation

The container/ephemeral_storage/used_bytes Cloud Monitoring metric reports local ephemeral-storage usage for each Kubernetes container. In Metrics Explorer, it can be filtered or grouped by resource labels to identify the Pods consuming the storage. The node-level metric provides only aggregate node usage and cannot identify the responsible Pods.

Learn more

Community Discussion

No comments yet. Be the first to start the discussion!