Free Kubernetes and Cloud Native Associate (KCNA) Practice Questions

Sample Practice Questions

Try these 10 sample questions from the Kubernetes and Cloud Native Associate (KCNA) practice bank. Each includes the answer and explanation. Continue with the interactive practice above to review the full question set.

1. 1Which Kubernetes component exposes the HTTP API used to manage cluster resources?

   1. A. kube-controller-manager
   2. B. kube-apiserver
   3. C. kubelet
   4. D. kube-proxy

   Answer: B

   Explanation: The documentation describes kube-apiserver as "The core component server that exposes the Kubernetes HTTP API." Controllers reconcile resources, kubelet manages Pods on nodes, and kube-proxy implements Service networking.

   Reference: Kubernetes Components — Control Plane Components — kube-apiserver

2. 2In the standard Kubernetes network model, what receives its own unique cluster-wide IP address?

   1. A. Every container image
   2. B. Every Pod
   3. C. Every container inside a Pod
   4. D. Every namespace

   Answer: B

   Explanation: The network model states that each Pod gets its "own unique cluster-wide IP address." Containers in a Pod share its network namespace; images and namespaces do not each receive a Pod IP.

   Reference: Services, Load Balancing, and Networking — The Kubernetes network model

3. 3Which change to an existing Kubernetes Deployment triggers a new rollout?

   1. A. Increasing only the replica count
   2. B. Changing only the Deployment's top-level annotation
   3. C. Changing a container image in the Pod template
   4. D. Reading the Deployment's status subresource

   Answer: C

   Explanation: The guide states, "A Deployment's rollout is triggered if and only if the Deployment's Pod template (that is, .spec.template)" is changed. Scaling alone or metadata outside the template does not trigger it.

   Reference: Deployments — Updating a Deployment — note

4. 4What does observability enable operators to understand from a system's external outputs?

   1. A. Only whether an externally tested endpoint is up or down
   2. B. Only internal state obtained by directly inspecting each process
   3. C. Its internal state and actionable conditions
   4. D. The absence of any possible future failure

   Answer: C

   Explanation: The glossary states, "It allows users to understand a system’s state from these external outputs and take (corrective) action." Observability is broader than an up/down test, does not require direct process inspection for every insight, and does not prove future faultlessness.

   Reference: Observability — Observability — definition

5. 5Where does the Kubernetes control plane store API server data?

   1. A. In the kubelet's Pod cache
   2. B. In the container image registry
   3. C. In the kube-proxy rule table
   4. D. In the etcd key-value store

   Answer: D

   Explanation: The etcd component is a "Consistent and highly-available key value store for all API server data." The other locations serve node execution, image distribution, or networking rather than API data persistence.

   Reference: Kubernetes Components — Control Plane Components — etcd

6. 6Two application containers share a Pod. Which address can they use to communicate with each other?

   1. A. Only the node's external address
   2. B. Only a separate Service address
   3. C. Only a public load balancer address
   4. D. localhost within the shared Pod network

   Answer: D

   Explanation: The network model says containers in the same Pod can communicate with each "other over localhost." A Service or external address is not required because the containers share a network namespace.

   Reference: Services, Load Balancing, and Networking — The Kubernetes network model — shared network namespace

7. 7A Deployment update is unstable and the previous revision remains in history. Which command rolls it back?

   1. A. kubectl rollout undo deployment/web
   2. B. kubectl rollout status deployment/web
   3. C. kubectl rollout restart deployment/web
   4. D. kubectl rollout history deployment/web

   Answer: A

   Explanation: The guide demonstrates "kubectl rollout undo deployment/nginx-deployment". undo restores a prior revision; status and history inspect, while restart initiates a restart rather than restoring the prior template.

   Reference: Deployments — Rolling Back to a Previous Revision

8. 8Which telemetry signal describes the path of a request through an application?

   1. A. A trace
   2. B. A metric
   3. C. A log record
   4. D. Baggage

   Answer: A

   Explanation: OpenTelemetry defines traces as "The path of a request through your application." Metrics measure runtime values, logs record events, and baggage carries contextual information.

   Reference: OpenTelemetry Signals — Traces

9. 9What is the role of kube-controller-manager in the Kubernetes control plane?

   1. A. Run controllers that implement Kubernetes API behavior
   2. B. Maintain Service networking rules on every worker node
   3. C. Run application containers through the container runtime
   4. D. Store all API server data in a persistent key-value database

   Answer: A

   Explanation: kube-controller-manager "Runs controllers to implement Kubernetes API behavior." Networking rules belong to kube-proxy, container execution to the runtime, and API data storage to etcd.

   Reference: Kubernetes Components — Control Plane Components — kube-controller-manager

10. 10For ordinary Pods without intentional network segmentation, what connectivity does the Kubernetes Pod network provide across nodes?

    1. A. Direct Pod communication without requiring NAT
    2. B. Communication only through public load balancers
    3. C. Communication only for Pods on the same node
    4. D. Communication only after assigning each Pod a NodePort

    Answer: A

    Explanation: The model specifies communication "directly, without the use of proxies or address translation (NAT)." The question concerns ordinary Pod networking and excludes segmentation and host-network exceptions.

    Reference: Services, Load Balancing, and Networking — The Kubernetes network model — Pod network