Resource requests and limits in Kubernetes - Practice Problems & Coding Challenges

Challenge - 5 Problems

💻 Command Output

intermediate

Pod resource requests and limits effect on scheduling

Given the following Pod YAML, what will be the status of the Pod after applying it to a cluster with only 500m CPU available on any node?

apiVersion: v1
kind: Pod
metadata:
  name: test-pod
spec:
  containers:
  - name: app
    image: nginx
    resources:
      requests:
        cpu: "600m"
      limits:
        cpu: "1"

Kubernetes

kubectl apply -f pod.yaml
kubectl get pod test-pod -o jsonpath='{.status.phase}'

A Pending

B Running

C CrashLoopBackOff

D Succeeded

💡 Hint

Think about how Kubernetes schedules Pods based on resource requests.

The Pod requests 600m CPU, but nodes only have 500m available. Kubernetes cannot schedule it, so the Pod stays Pending.

🧠 Conceptual

intermediate

Understanding resource limits impact on container behavior

What happens if a container tries to use more CPU than its limit in Kubernetes?

A The container is throttled and CPU usage is limited to the set limit

B The container is immediately terminated

C The container ignores the limit and uses more CPU

D The Pod is evicted from the node

💡 Hint

Think about how CPU limits control usage but do not kill the container.

CPU limits cause the container to be throttled, meaning it cannot exceed the CPU usage set by the limit but continues running.

❓ Troubleshoot

advanced

Diagnosing OOMKilled due to memory limits

A Pod is repeatedly restarting with status 'OOMKilled'. The Pod spec includes:

resources:
  limits:
    memory: "256Mi"
  requests:
    memory: "128Mi"

What is the most likely cause?

A The container image is corrupted causing crashes

B The container did not request enough CPU and was throttled

C The container tried to use more than 256Mi memory and was killed by the system

D The Pod was evicted due to node disk pressure

💡 Hint

OOMKilled means out of memory kill by the kernel.

When a container exceeds its memory limit, the kernel kills it to protect the node, causing OOMKilled status.

🔀 Workflow

advanced

Setting resource requests and limits in a Deployment YAML

Which snippet correctly sets CPU requests to 200m and limits to 500m for a container in a Deployment?

A

resources:
  requests:
    memory: "200m"
  limits:
    memory: "500m"

B

resources:
  limits:
    cpu: "200m"
  requests:
    cpu: "500m"

C

resources:
  cpu:
    requests: "200m"
    limits: "500m"

D

resources:
  requests:
    cpu: "200m"
  limits:
    cpu: "500m"

💡 Hint

Remember the correct keys under resources are requests and limits.

Option D correctly uses requests and limits keys with cpu values as strings with units.

✅ Best Practice

expert

Choosing resource requests and limits for production workloads

Which approach is best when setting resource requests and limits for a production Kubernetes workload?

A Set requests low and limits very high to allow burst usage without scheduling issues

B Set requests based on average usage and limits slightly above to allow bursts, monitoring and adjusting over time

C Set requests and limits equal to the maximum expected usage to avoid throttling and evictions

D Do not set requests or limits to let Kubernetes manage resources automatically

💡 Hint

Think about balancing resource guarantees and flexibility.

Setting requests based on average usage ensures scheduling fits node capacity, while limits allow bursts. Monitoring helps adjust values.