GPU nodes are tainted, and a pod will not schedule there. What must the pod have?
Why D is the answer
A matching toleration
Taints repel pods that do not tolerate them.
In detail
- Taints repel pods that do not tolerate them.
- Taints repel pods without tolerations.
Use case
An illustrative scenario, not a real incident.
GPU nodes are tainted to keep other work off them, so GPU pods need a toleration.
See it in different ways
flowchart TD
A["Pod Pending"] --> B["kubectl describe pod"]
B --> C{"Does the node report GPUs?"}
C -->|No| D["Check operator and device plugin"]
C -->|Yes| E["Check quotas, taints and selectors"]stateDiagram-v2 [*] --> Pending Pending --> NoGpuReported: node shows no GPUs Pending --> Taint: no matching toleration Pending --> WrongRequest: no nvidia.com/gpu limit NoGpuReported --> CheckOperator Taint --> AddToleration WrongRequest --> FixPodSpec
flowchart TD
Q{"Does the pod tolerate the taint?"}
Q -->|"Yes"| A0["It can schedule"]
Q -->|"No"| A1["It stays away"]Key point: Taints need matching tolerations.
Why the other options are wrong
- A. A shorter name
- A name is unrelated.
- B. A larger image
- Image size is unrelated.
- C. A longer command
- A command is unrelated.