Detailed Exam Domain Coverage

The Certified Kubernetes Application Developer (CKAD) is an intermediate-level, performance-based exam offered by the Cloud Native Computing Foundation (CNCF). It assesses your ability to design, build, configure, and expose cloud-native applications on Kubernetes using the command line. The exam is taken online, proctored remotely, and is open-book. I have structured this practice material to comprehensively cover all exam domains so you can test your knowledge effectively.

Application Design and Build (20%)
Application Deployment (20%)
Application Observability and Maintenance (15%)
Application Environment, Configuration and Security (25%)
Services and Networking (20%)

Mastering the command line and understanding how to troubleshoot live clusters is critical. I created these practice tests to simulate the pressure and technical depth of the actual exam, ensuring you are fully prepared to pass on your first attempt.

Sample Practice Questions Preview

Here is a preview of the types of questions you will find in this course.

Question 1: Application Observability and Maintenance

You have a pod named 'web-app' running in the 'production' namespace that is continually crashing. You need to check the logs of the previous instance of the container named 'nginx-container' within this pod to determine the root cause. Which command should you use?

Option A: kubectl logs web-app -c nginx-container -n production
Option B: kubectl logs web-app -c nginx-container -p -n production
Option C: kubectl logs -p web-app -n production
Option D: kubectl describe pod web-app -n production
Option E: kubectl get logs web-app -c nginx-container --previous -n production
Option F: kubectl logs pod/web-app --previous -n production
Correct Answer: Option B
Overall Explanation: When troubleshooting a crashing pod, examining the logs of the previously terminated container instance is essential. The -p or --previous flag in the kubectl logs command allows you to view the logs for the crashed instance. You must also specify the exact container name using -c if there are multiple containers, and pass the correct namespace using -n.
Why Option A is incorrect: It fetches the logs of the currently running container, not the previous crashed instance.
Why Option B is correct: It correctly includes the pod name, the specific container name (-c), the previous flag (-p), and the correct namespace (-n).
Why Option C is incorrect: It omits the container name. If the pod has multiple containers, this command will fail and prompt you to specify a container.
Why Option D is incorrect: The describe command provides events and resource configuration details, not the standard output logs of the application.
Why Option E is incorrect: kubectl get logs is invalid syntax. The correct verb is kubectl logs.
Why Option F is incorrect: It lacks the -c flag to specify the exact container name, which is required if the pod contains sidecars or init containers.

Question 2: Application Design and Build

You are tasked with creating a Kubernetes Job named 'batch-job' that must run a specific container image and guarantee that exactly 5 completions are successful, with no more than 2 pods running in parallel. Which configuration parameters must be set in the Job spec?

Option A: completions: 5, parallelism: 2
Option B: completions: 2, parallelism: 5
Option C: replicas: 5, maxSurge: 2
Option D: parallel: 2, successfulJobs: 5
Option E: completions: 5, activeDeadlineSeconds: 2
Option F: backoffLimit: 2, completions: 5
Correct Answer: Option A
Overall Explanation: In Kubernetes, a Job creates one or more Pods and ensures that a specified number of them successfully terminate. The completions field specifies the target number of successful pod completions. The parallelism field limits how many pods can run concurrently.
Why Option A is correct: It accurately assigns 5 to completions and 2 to parallelism, fulfilling all the requirements of the prompt.
Why Option B is incorrect: It reverses the requirements, setting it to 2 completions and 5 parallel pods.
Why Option C is incorrect: replicas and maxSurge are properties of a Deployment, not a Job.
Why Option D is incorrect: parallel and successfulJobs are not valid fields in the Kubernetes Job API specification.
Why Option E is incorrect: activeDeadlineSeconds determines the maximum duration the Job can be active, which does not address the parallel execution requirement.
Why Option F is incorrect: backoffLimit specifies the number of retries before marking the job as failed, which is irrelevant to the parallelism requirement.

Question 3: Application Deployment

You have updated the image of a Deployment named 'frontend' from version v1 to v2. You notice the new pods are crash-looping, and you want to revert the Deployment to the previous working state immediately. Which command accomplishes this?

Option A: kubectl rollout restart deployment frontend
Option B: kubectl scale deployment frontend --replicas=0
Option C: kubectl set image deployment/frontend nginx=nginx:v1
Option D: kubectl rollout undo deployment frontend
Option E: kubectl deployment rollback frontend
Option F: kubectl revert deployment frontend --to-revision=1
Correct Answer: Option D
Overall Explanation: Kubernetes Deployments keep a history of rollouts. If a new rollout introduces a bug or crash loop, the fastest and most standard way to restore service is to undo the rollout, which restores the Deployment's PodTemplate to its previous revision.
Why Option A is incorrect: This will force a restart of the pods using the current, broken configuration (v2 image), failing to fix the issue.
Why Option B is incorrect: Scaling to zero stops the application entirely, resulting in downtime rather than reverting to a working state.
Why Option C is incorrect: While this manually changes the image back, it creates a brand new rollout revision rather than utilizing the built-in rollback history mechanism.
Why Option D is correct: The kubectl rollout undo command is the standard and most efficient way to revert a Deployment to its immediately previous revision.
Why Option E is incorrect: This is not a valid Kubernetes command.
Why Option F is incorrect: kubectl revert does not exist in the standard Kubernetes CLI toolset.

Welcome to the Mock Exam Practice Tests Academy to help you prepare for your CKAD: Certified Kubernetes Application Developer
You can retake the exams as many times as you want
This is a huge original question bank
You get support from instructors if you have questions
Each question has a detailed explanation
Mobile-compatible with the Udemy app

I hope that by now you're convinced! And there are a lot more questions inside the course.