Which component of Kubernetes uses the API to track the state of pods and nodes, and to schedule pods onto nodes?

Which component of Kubernetes uses the API to track the state of pods and nodes, and to schedule pods onto nodes?

• A. kube-proxy
• B. etcd
• C. kubelet
• D. kube-scheduler

Answer: (D)

Scheduling is the component that uses the API to watch pods and nodes, decide where a pod should run, and then bind the pod to a chosen node. It continually monitors for pods that aren’t assigned to a node, evaluates which nodes can host them based on resources, constraints (taints/tolerations, node selectors, affinity), and priorities, and then updates the pod’s binding to the selected node. Once bound, the kubelet on that node takes over and starts the containers. Other components don’t perform scheduling: kube-proxy manages network rules for services, etcd stores cluster state in the backing store, and kubelet runs containers on each node.

Scheduling is the component that uses the API to watch pods and nodes, decide where a pod should run, and then bind the pod to a chosen node. It continually monitors for pods that aren’t assigned to a node, evaluates which nodes can host them based on resources, constraints (taints/tolerations, node selectors, affinity), and priorities, and then updates the pod’s binding to the selected node. Once bound, the kubelet on that node takes over and starts the containers.

Other components don’t perform scheduling: kube-proxy manages network rules for services, etcd stores cluster state in the backing store, and kubelet runs containers on each node.