Verify operator functionality: Practice Questions — Kubernetes Integration (NVIDIA-Certified Professional: AI Networking)

Practice Questions: Verifying NVIDIA Network Operator Functionality in Kubernetes

This set of multiple-choice questions is designed to help candidates prepare for the Verify operator functionality task within the Kubernetes Integration topic of the NVIDIA-Certified Professional: AI Networking exam. The focus is on deploying and verifying the NVIDIA Network Operator for RDMA and InfiniBand environments.

1. Which command is used to check the status of the NVIDIA Network Operator deployment in a Kubernetes cluster?

   • A. kubectl get pods -n nvidia-operator
   • B. kubectl describe node
   • C. kubectl get svc -n kube-system
   • D. kubectl logs -f nvidia-operator

   Correct answer: A

   Explanation: The command kubectl get pods -n nvidia-operator lists the pods running in the namespace where the NVIDIA Network Operator is deployed, allowing verification of its status.

2. After deploying the NVIDIA Network Operator, which resource should you inspect to confirm that RDMA devices are correctly exposed to the Kubernetes nodes?

   • A. nvidia.com/gpu device plugin logs
   • B. nvidia.com/rdma device plugin status
   • C. nvidia.com/infiniBand config map
   • D. nvidia.com/network-operator pod events

   Correct answer: B

   Explanation: The nvidia.com/rdma device plugin status confirms that RDMA devices are correctly registered and available on the nodes.

3. Which Kubernetes Custom Resource Definition (CRD) is primarily used by the NVIDIA Network Operator to manage InfiniBand and RDMA configurations?

   • A. NetworkNodePolicy
   • B. NetworkOperatorConfig
   • C. NetworkNodeState
   • D. NetworkDevicePlugin

   Correct answer: C

   Explanation: The NetworkNodeState CRD is used by the operator to represent and manage the state of network devices such as InfiniBand and RDMA on each node.

4. Which command helps verify that the NVIDIA Network Operator has successfully configured InfiniBand devices on the nodes?

   • A. kubectl get networknodestates -A
   • B. kubectl get pods -n kube-system
   • C. kubectl get nodes -o wide
   • D. kubectl describe daemonset nvidia-network-operator

   Correct answer: A

   Explanation: Listing networknodestates across all namespaces shows the current state of InfiniBand and RDMA devices as managed by the operator.

5. What is the best way to confirm that the NVIDIA Network Operator is correctly managing the lifecycle of RDMA device plugins?

   • A. Check that the rdma-device-plugin pods are running and ready in the operator namespace
   • B. Verify that GPU metrics are visible in Prometheus
   • C. Confirm that the Kubernetes scheduler is assigning GPU workloads
   • D. Inspect the logs of the Kubernetes API server

   Correct answer: A

   Explanation: The rdma-device-plugin pods are responsible for exposing RDMA devices to Kubernetes. Their running and ready status indicates proper operator management.

6. Which of the following is NOT a valid method to verify NVIDIA Network Operator functionality?

   • A. Checking pod status in the operator namespace
   • B. Inspecting networknodestates CRD resources
   • C. Running nvidia-smi to check GPU status
   • D. Reviewing device plugin logs for errors

   Correct answer: C

   Explanation: While nvidia-smi is useful for GPUs, it does not verify RDMA or InfiniBand device configuration managed by the NVIDIA Network Operator.

7. When verifying the operator functionality, which log source is most useful for troubleshooting issues with RDMA device plugin deployment?

   • A. NVIDIA Network Operator pod logs
   • B. Kubernetes API server logs
   • C. kubelet logs on worker nodes
   • D. RDMA device plugin pod logs

   Correct answer: D

   Explanation: Logs from the RDMA device plugin pods provide detailed information about device discovery and registration, essential for troubleshooting.

8. What Kubernetes namespace is typically used for deploying the NVIDIA Network Operator and its components?

   • A. default
   • B. nvidia-operator
   • C. kube-system
   • D. rdma-system

   Correct answer: B

   Explanation: The NVIDIA Network Operator is commonly deployed in the nvidia-operator namespace to isolate its components.

More in this topic

Verify operator functionality: Common Mistakes — Kubernetes Integration (NVIDIA-Certified Professional: AI Networking)Deploy the NVIDIA Network Operator for RDMA and InfiniBand: Quick Reference — Kubernetes Integration (NVIDIA-Certified Professional: AI Networking)Deploy the NVIDIA Network Operator for RDMA and InfiniBand: Worked Example — Kubernetes Integration (NVIDIA-Certified Professional: AI Networking)Verify operator functionality: Quick Reference — Kubernetes Integration (NVIDIA-Certified Professional: AI Networking)Verify operator functionality: Worked Example — Kubernetes Integration (NVIDIA-Certified Professional: AI Networking)Deploy the NVIDIA Network Operator for RDMA and InfiniBand — Kubernetes Integration (NVIDIA-Certified Professional: AI Networking)Kubernetes Integration — NVIDIA-Certified Professional: AI NetworkingVerify operator functionality — Kubernetes Integration (NVIDIA-Certified Professional: AI Networking)Deploy the NVIDIA Network Operator for RDMA and InfiniBand: Common Mistakes — Kubernetes Integration (NVIDIA-Certified Professional: AI Networking)Deploy the NVIDIA Network Operator for RDMA and InfiniBand: Practice Questions — Kubernetes Integration (NVIDIA-Certified Professional: AI Networking)