Kubernetes
Hard
Troubleshooting
A container keeps trying to use more CPU than its CPU limit. What happens?
Select your answer:
Show answer and explanation
Answer: The kernel throttles it through the CFS quota and it keeps running
Explanation
CPU is a compressible resource. When a container reaches its CPU limit, the Linux CFS quota throttles it, which shows up as latency, not restarts. Memory is different. A container that goes over its memory limit is killed by the OOM killer and reports the reason OOMKilled.
More hard Kubernetes questions
- A pod was scheduled using a requiredDuringSchedulingIgnoredDuringExecution node affinity rule. Later someone removes the matching label from the node. What happens to the pod?
- A node gets a NoExecute taint. A pod already running on it tolerates the taint with tolerationSeconds: 300. What happens?
- What does a DNS lookup of a headless Service (clusterIP: None) with a selector return?
- What does externalTrafficPolicy: Local do on a NodePort or LoadBalancer Service?
- In a NetworkPolicy ingress rule, what is the difference between putting namespaceSelector and podSelector in the same 'from' entry and putting them in two separate entries?
- Two controllers read the same Deployment and both send an update based on what they read. How does the API server stop the second update from silently overwriting the first?