CKAD field manual

k8s requests limits YAML example

requests is what the scheduler reserves. limits is the runtime cap. Set both on the container.

Set on a Deployment

kubectl set resources deploy web --requests=cpu=100m,memory=128Mi --limits=cpu=200m,memory=256Mi

YAML

apiVersion: v1
kind: Pod
metadata:
  name: web
spec:
  containers:
    - name: web
      image: nginx:1.27
      resources:
        requests:
          cpu: "100m"
          memory: 128Mi
        limits:
          cpu: "200m"
          memory: 256Mi

Inspect

kubectl describe po web | grep -A6 Limits
kubectl top po web

Fields

requests: What the scheduler guarantees. The Pod stays Pending if no node has that much free.
limits: Cap. Memory over → OOMKilled. CPU over → throttle, no kill.
cpu: "100m" is 0.1 core. Quote it. 1 is one full core.
memory: 128Mi, not 128m. m is millicores. Mi is mebibytes.
kubectl set resources: Patches a live Deployment. Triggers a rollout.

Watch

requests = scheduling. limits = cap. Memory over limit → OOMKilled. CPU over limit → throttle.
128Mi is memory. 128m is 0.128 CPU. Mixing them is a common exam miss.
request == limit on every container → QoS Guaranteed. HPA CPU targets the request, not the limit.

Official docs Resource Management

Practice these objects on a live cluster →