CKAD Application Deployment Practice Question
A developer needs to run a one-off batch job that processes a queue and must complete exactly once before the pod is terminated. The container image is 'registry.example.com/worker:1.4'. Which resource should be created to meet this requirement?
⚠ Common exam trap
The trap here is treating a one-off batch task like a long-running service and choosing a Deployment, which would restart the container after it exits successfully.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
A Job with a pod template that sets restartPolicy to Never or OnFailure and a backoffLimit controlling retries.
A Job is purpose-built for run-to-completion workloads. Setting restartPolicy to Never or OnFailure and configuring backoffLimit gives controlled retries on failure, while a successful exit marks the Job complete without restarting the pod. Deployments restart containers indefinitely, CronJobs repeat on a schedule, and bare Pods lack controller-managed completion and rescheduling.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
A CronJob with a schedule of '* * * * *' so the worker runs every minute and eventually processes the queue.
Why it's wrong here
A CronJob creates Jobs on a recurring schedule, which is intended for periodic work, not a single run-to-completion task. Scheduling it every minute would launch overlapping workers, potentially processing queue items multiple times. The requirement is a one-off execution, so a scheduled resource adds unnecessary complexity and violates the exactly-once intent.
- ✗
A Deployment with replicas set to 1 and a restartPolicy of Always so the worker restarts if it exits.
Why it's wrong here
A Deployment manages long-running services, and its pod template must use restartPolicy Always. That means when the batch container exits successfully, the pod is restarted instead of being considered complete, causing the job to run repeatedly. Deployments also lack completion semantics, so there is no way to signal that the one-off task finished.
- ✗
A bare Pod manifest with no controller, relying on the kubelet to restart the container if it fails.
Why it's wrong here
A bare Pod has no controller to track completion or retry policy, and if the node fails the pod is not rescheduled. It also cannot express a backoffLimit or completion state. While it avoids the Deployment restart loop, it does not provide the reliable one-off execution semantics that a Job offers for batch workloads.
- ✓
A Job with a pod template that sets restartPolicy to Never or OnFailure and a backoffLimit controlling retries.
Why this is correct
A Job is the Kubernetes workload designed for run-to-completion tasks. Its pod template must specify restartPolicy Never or OnFailure, and backoffLimit governs retry attempts on failure. Once the container exits successfully, the Job marks completion and does not restart the pod, which matches the requirement that the batch process complete exactly once.
Go deeper
Related to this question
About these practice questions
This CKAD question is part of Courseiva's 826-question bank — original exam-style content with full explanations and wrong-answer analysis, never real exam questions or exam dumps. Learn why practice questions differ from exam dumps →
JA
Written and reviewed by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
Last reviewed September 2026 · checked against the official CNCF exam blueprint
This CKAD practice question is part of Courseiva's free CNCF certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the CKAD exam.