Vertical Pod Autoscaler concept in Kubernetes - Practice Problems & Coding Challenges

Challenge - 5 Problems

🧠 Conceptual

intermediate

What is the primary function of the Vertical Pod Autoscaler (VPA) in Kubernetes?

Choose the best description of what the Vertical Pod Autoscaler does in a Kubernetes cluster.

A It scales the number of pod replicas up or down based on incoming traffic.

B It automatically adjusts the CPU and memory requests of pods based on usage patterns.

C It manages persistent storage volumes for pods dynamically.

D It schedules pods to nodes based on node labels and taints.

💡 Hint

Think about resource allocation inside a pod rather than the number of pods.

The Vertical Pod Autoscaler changes the CPU and memory requests of pods to better fit their actual usage, helping pods run efficiently without manual tuning.

💻 Command Output

intermediate

What is the output of this command related to VPA?

Given the command kubectl get vpa in a cluster with one VPA resource named my-app-vpa, what is the expected output?

Kubernetes

kubectl get vpa

A

NAME        READY
my-app-vpa  True

B No resources found in default namespace.

C

NAME        RECOMMENDATION
my-app-vpa  cpu: 500m, memory: 256Mi

D Error from server (NotFound): verticalpodautoscalers.autoscaling.k8s.io "my-app-vpa" not found

💡 Hint

The command lists VPA resources and their current recommendations.

The kubectl get vpa command shows the VPA resources and their recommended CPU and memory settings for pods.

❓ Configuration

advanced

Which VPA configuration snippet correctly sets the update mode to 'Auto'?

Identify the correct YAML snippet that configures a Vertical Pod Autoscaler with update mode set to Auto.

A

apiVersion: autoscaling.k8s.io/v1
kind: VerticalPodAutoscaler
metadata:
  name: example-vpa
spec:
  targetRef:
    apiVersion: apps/v1
    kind: Deployment
    name: example-deployment
  updatePolicy:
    update_mode: Auto

B

apiVersion: autoscaling.k8s.io/v1
kind: VerticalPodAutoscaler
metadata:
  name: example-vpa
spec:
  targetRef:
    apiVersion: apps/v1
    kind: Deployment
    name: example-deployment
  updatePolicy:
    updateMode: Manual

C

apiVersion: autoscaling.k8s.io/v1
kind: VerticalPodAutoscaler
metadata:
  name: example-vpa
spec:
  targetRef:
    apiVersion: apps/v1
    kind: Deployment
    name: example-deployment
  updatePolicy:
    mode: Auto

D

apiVersion: autoscaling.k8s.io/v1
kind: VerticalPodAutoscaler
metadata:
  name: example-vpa
spec:
  targetRef:
    apiVersion: apps/v1
    kind: Deployment
    name: example-deployment
  updatePolicy:
    updateMode: Auto

💡 Hint

Check the exact field name and value for update mode in the VPA spec.

The correct field is updateMode under updatePolicy with the value Auto. Other variations are invalid keys or values.

❓ Troubleshoot

advanced

Why might a VPA not update pod resources even when usage is high?

Given a VPA configured with updateMode: Auto, pods still run with old resource requests despite high CPU and memory usage. What is a likely cause?

A The VPA cannot update running pods; it only updates resource requests for new pods after restart.

B The pods are managed by a Deployment with <code>replicaSet</code> and VPA does not support Deployments.

C The VPA is in <code>Auto</code> mode but the pods have <code>resourceLimits</code> set too low.

D The VPA is configured with <code>updateMode: Off</code> instead of <code>Auto</code>.

💡 Hint

Think about how resource changes affect running pods.

VPA in Auto mode updates resource requests but cannot change resources of running pods immediately. Pods must be restarted to apply new requests.

✅ Best Practice

expert

What is the recommended approach to safely use VPA with Horizontal Pod Autoscaler (HPA)?

When using both Vertical Pod Autoscaler and Horizontal Pod Autoscaler together, which practice helps avoid conflicts and ensures stable scaling?

A Set VPA update mode to <code>Initial</code> so it only sets resources at pod start, while HPA manages pod count.

B Configure VPA in <code>Off</code> mode and let HPA handle all scaling decisions.

C Set VPA update mode to <code>Auto</code> and HPA to scale based on CPU only.

D Run VPA and HPA with no coordination; Kubernetes will handle conflicts automatically.

💡 Hint

Consider how resource changes and pod count scaling interact.

Using VPA in Initial mode sets resource requests only when pods start, avoiding conflicts with HPA which scales pod numbers dynamically.