Deployment as higher-level abstraction in Kubernetes - Practice Problems & Coding Challenges

Challenge - 5 Problems

🧠 Conceptual

intermediate

What is the primary purpose of a Kubernetes Deployment?

Choose the main function of a Deployment in Kubernetes.

A To expose Pods to external traffic via a stable IP address

B To manage the lifecycle of Pods and ensure the desired number of replicas are running

C To store configuration data for Pods and containers

D To schedule Pods onto specific nodes based on resource availability

💡 Hint

Think about what keeps your app running smoothly even if some parts fail.

A Deployment manages Pods by creating, updating, and scaling them to maintain the desired state.

💻 Command Output

intermediate

Output of 'kubectl get deployment myapp' after scaling

What is the output of the following command after scaling the Deployment to 3 replicas?

kubectl get deployment myapp

A

NAME   READY   UP-TO-DATE   AVAILABLE   AGE
myapp  0/3     0            0           10m

B

NAME   READY   UP-TO-DATE   AVAILABLE   AGE
myapp  1/3     3            1           10m

C

NAME   READY   UP-TO-DATE   AVAILABLE   AGE
myapp  3/3     3            3           10m

D Error from server (NotFound): deployments.apps "myapp" not found

💡 Hint

Scaling to 3 replicas means all 3 should be ready and available.

The READY column shows how many Pods are running and ready out of the desired replicas. After scaling to 3, all 3 should be ready.

❓ Configuration

advanced

Correct YAML snippet for a Deployment with rolling update strategy

Which YAML snippet correctly configures a Deployment to use a rolling update strategy with max 1 unavailable Pod?

A

strategy:
  type: RollingUpdate
  rollingUpdate:
    maxUnavailable: 1

B

strategy:
  type: Recreate
  rollingUpdate:
    maxUnavailable: 1

C

strategy:
  type: RollingUpdate
  rollingUpdate:
    maxSurge: 0

D

strategy:
  type: RollingUpdate
  rollingUpdate:
    maxUnavailable: 0
    maxSurge: 2

💡 Hint

RollingUpdate with maxUnavailable controls how many Pods can be down during update.

Option A correctly sets the strategy type and maxUnavailable to 1, allowing one Pod to be unavailable during updates.

❓ Troubleshoot

advanced

Why does a Deployment keep creating new Pods endlessly?

You notice your Deployment keeps creating new Pods and old ones are terminating repeatedly. What is the most likely cause?

A The Deployment has no replicas specified, so it keeps creating Pods

B The nodes have insufficient resources causing Pods to crash

C The Service selector does not match the Pod labels

D The Pod template spec has changed causing a new ReplicaSet to be created each time

💡 Hint

Think about what triggers a new ReplicaSet creation in a Deployment.

Changing the Pod template causes Kubernetes to create a new ReplicaSet and terminate old Pods to match the new spec.

🔀 Workflow

expert

Order the steps to perform a zero-downtime Deployment update

Arrange the steps in the correct order to update a Deployment without downtime.

A 1,2,3,4

B 2,1,3,4

C 1,3,2,4

D 3,1,2,4

💡 Hint

Think about the natural flow from editing to applying and then Kubernetes handling the rollout.

First update the manifest, then apply it. Kubernetes then creates new Pods and gradually removes old ones to avoid downtime.