Why would you add a nodeSelector to a Pod specification in Kubernetes?

Why would you add a nodeSelector to a Pod specification in Kubernetes?

• A. To ensure the pod runs on a specific node due to storage limitations
• B. To set environment variables for the container
• C. To control resource requests and limits for the pod
• D. To ensure image pulls happen from a specific node

Answer: (D)

NodeSelector constrains where a Pod can be scheduled by requiring the target node to have specific labels. You label nodes with properties you care about (for example, high-storage, GPU-enabled, or a particular zone), and then reference those labels in the Pod specification. The scheduler will only place the Pod on nodes whose labels match, ensuring the workload runs on hardware or in an environment that fits its needs. This is handy when storage capabilities, hardware, or locality matter for the Pod. It’s not about environment variables, resource limits, or image pulling. Resource requests/limits are configured in the resources field, and environment variables are set in the container spec. Image pulls happen on the node where the Pod runs and are determined by the image name and imagePullPolicy, not by NodeSelector. If you need to ensure a specific image exists on a node, you’d pre-pull it or use a mechanism like a DaemonSet to prepare nodes.

NodeSelector constrains where a Pod can be scheduled by requiring the target node to have specific labels. You label nodes with properties you care about (for example, high-storage, GPU-enabled, or a particular zone), and then reference those labels in the Pod specification. The scheduler will only place the Pod on nodes whose labels match, ensuring the workload runs on hardware or in an environment that fits its needs. This is handy when storage capabilities, hardware, or locality matter for the Pod.

It’s not about environment variables, resource limits, or image pulling. Resource requests/limits are configured in the resources field, and environment variables are set in the container spec. Image pulls happen on the node where the Pod runs and are determined by the image name and imagePullPolicy, not by NodeSelector. If you need to ensure a specific image exists on a node, you’d pre-pull it or use a mechanism like a DaemonSet to prepare nodes.