Practice Exams:

Linux Foundation KCNA Dumps

EXAM: Kubernetes and Cloud Native Associate
Verified By Experts
Premium Bundle
$40.00
$59.97 $99.97

KCNA Premium Bundle


  • Premium File 333 Questions & Answers
  • Training Course 54 Lectures
  • Study Guide 410 Pages

Linux Foundation KCNA Practice Test 62

Question 610

A retail payments app enters CrashLoopBackOff after deploying a malformed startup flag. What pair of supported action and practical verification should be approved?

  1. Inspect Pod container last state, restart count and kubectl logs --previous when a process repeatedly exits after startup. Acceptance evidence: Run a synthetic app that exits with a nonzero code, then confirm describe, previous logs and restart data reveal the failure path.
  2. Give clients a Service DNS name rather than hardcoding Pod IP addresses that can change when replicas are replaced. Acceptance evidence: Delete and replace a Deployment Pod and verify a client continues to reach its Service DNS name despite changed backend Pod addresses.
  3. Resolve an object's apiVersion, kind, metadata namespace and name before applying or deleting it. Acceptance evidence: List Deployments with --all-namespaces and inspect one manifest's apiVersion and kind to target a specific test object.
  4. Inspect FailedMount and PVC events, then verify StorageClass, PVC binding and node driver prerequisites before rewriting application code. Acceptance evidence: Reproduce a missing PVC in a test namespace and confirm the Pod Events directly indicate the unavailable claim.

Explanation

Correct Answer

A:

The situation is that a retail payments app enters CrashLoopBackOff after deploying a malformed startup flag. Inspect Pod container last state, restart count and kubectl logs --previous when a process repeatedly exits after startup. Acceptance evidence: Run a synthetic app that exits with a nonzero code, then confirm describe, previous logs and restart data reveal the failure path. This pairing is justified because crashLoopBackOff reflects kubelet restart backoff after container failures; the underlying cause may be application configuration, exit code or a failed liveness probe.

Incorrect Answers

B:

This action and check address the distinct objective “Distinguish Pod IPs from stable Services during rescheduling” instead of the stated acceptance requirement. The matching action is to inspect Pod container last state, restart count and kubectl logs --previous when a process repeatedly exits after startup. Its confirming test is to run a synthetic app that exits with a nonzero code, then confirm describe, previous logs and restart data reveal the failure path.

C:

This action and check address the distinct objective “Read the API resource model using namespaces and names rather than guessing ownership” instead of the stated acceptance requirement. The matching action is to inspect Pod container last state, restart count and kubectl logs --previous when a process repeatedly exits after startup. Its confirming test is to run a synthetic app that exits with a nonzero code, then confirm describe, previous logs and restart data reveal the failure path.

D:

This action and check address the distinct objective “Use Events and pod describe for failed volume mounting” instead of the stated acceptance requirement. The matching action is to inspect Pod container last state, restart count and kubectl logs --previous when a process repeatedly exits after startup. Its confirming test is to run a synthetic app that exits with a nonzero code, then confirm describe, previous logs and restart data reveal the failure path.

Question 611

A logistics demo Pod never receives a nodeName due to missing accelerator nodes. Which supported Kubernetes practice directly addresses the stated requirement?

  1. Combine RBAC, Pod security policies, NetworkPolicy and resource quotas for teams that share a Kubernetes control plane.
  2. Read Pod Events and node assignment to distinguish a Pod that cannot be scheduled from a scheduled Pod whose image cannot be pulled.
  3. Taint dedicated nodes and grant only intended workloads matching tolerations, combining node affinity when exclusive placement is required.
  4. Run controlled DNS queries and IP connectivity tests before changing service discovery, CNI routes or application endpoints.

Explanation

Correct Answer

B:

Read Pod Events and node assignment to distinguish a Pod that cannot be scheduled from a scheduled Pod whose image cannot be pulled. This resolves the issue because failedScheduling occurs before node binding; ImagePullBackOff occurs when a node's kubelet cannot obtain the container image. To verify, create two controlled faulty workloads, one requiring an impossible node label and one referencing a missing image tag, then compare Events.

Incorrect Answers

A:

This option addresses the distinct objective “Separate multi-tenant namespace governance from actual isolation guarantees”, not the immediate requirement. Namespaces alone partition naming but do not provide complete security isolation, capacity governance or protection from privileged node-level workloads. For the stated case, failedScheduling occurs before node binding; ImagePullBackOff occurs when a node's kubelet cannot obtain the container image.

C:

This option addresses the distinct objective “Understand taints and tolerations as repulsion controls”, not the immediate requirement. A NoSchedule taint repels Pods without a matching toleration; tolerance permits consideration but does not force placement on that node. For the stated case, failedScheduling occurs before node binding; ImagePullBackOff occurs when a node's kubelet cannot obtain the container image.

D:

This option addresses the distinct objective “Diagnose cluster DNS separately from Pod network reachability”, not the immediate requirement. Healthy direct IP connectivity with failed name resolution suggests a DNS path issue; failed Pod-to-Pod IP traffic suggests a lower-level network problem. For the stated case, failedScheduling occurs before node binding; ImagePullBackOff occurs when a node's kubelet cannot obtain the container image.

Question 612

A logistics demo Pod never receives a nodeName due to missing accelerator nodes. Which causal explanation supports the intended design choice?

  1. CrashLoopBackOff reflects kubelet restart backoff after container failures; the underlying cause may be application configuration, exit code or a failed liveness probe.
  2. FailedScheduling occurs before node binding; ImagePullBackOff occurs when a node's kubelet cannot obtain the container image.
  3. EndpointSlices describe endpoint addresses and readiness associated with a Service, allowing its dataplane to route toward matching healthy Pods.
  4. Labels are designed for selectors and filtering, whereas annotations hold descriptive metadata not intended as a resource-matching control.

Explanation

Correct Answer

B:

FailedScheduling occurs before node binding; ImagePullBackOff occurs when a node's kubelet cannot obtain the container image. It supports this action: Read Pod Events and node assignment to distinguish a Pod that cannot be scheduled from a scheduled Pod whose image cannot be pulled. To confirm, create two controlled faulty workloads, one requiring an impossible node label and one referencing a missing image tag, then compare Events.

Incorrect Answers

A:

That statement explains the distinct objective “Diagnose CrashLoopBackOff from restart history and previous logs” rather than the cause in this case. The deciding mechanism here is that failedScheduling occurs before node binding; ImagePullBackOff occurs when a node's kubelet cannot obtain the container image.

C:

That statement explains the distinct objective “Use EndpointSlices to connect Service selectors with actual backend Pods” rather than the cause in this case. The deciding mechanism here is that failedScheduling occurs before node binding; ImagePullBackOff occurs when a node's kubelet cannot obtain the container image.

D:

That statement explains the distinct objective “Use annotations for tool metadata rather than workload selection” rather than the cause in this case. The deciding mechanism here is that failedScheduling occurs before node binding; ImagePullBackOff occurs when a node's kubelet cannot obtain the container image.

Question 613

A logistics demo Pod never receives a nodeName due to missing accelerator nodes. Which test provides the strongest evidence of a correct implementation?

  1. Run a test web Pod without a token mount and verify service operation succeeds while in-cluster API calls are unavailable.
  2. Create two controlled faulty workloads, one requiring an impossible node label and one referencing a missing image tag, then compare Events.
  3. Test the same Service from a diagnostic Pod by fully qualified DNS name and known ClusterIP, then inspect CoreDNS logs if only resolution fails.
  4. Build a test image, run a Pod and recreate it, verifying packaged files remain while an unmounted temporary file disappears.

Explanation

Correct Answer

B:

Create two controlled faulty workloads, one requiring an impossible node label and one referencing a missing image tag, then compare Events. This is the relevant test because failedScheduling occurs before node binding; ImagePullBackOff occurs when a node's kubelet cannot obtain the container image.

Incorrect Answers

A:

This check would help evaluate the distinct objective “Enforce workload service account token boundaries and avoid needless mounts”, but not the required result. For this case, create two controlled faulty workloads, one requiring an impossible node label and one referencing a missing image tag, then compare Events.

C:

This check would help evaluate the distinct objective “Diagnose cluster DNS separately from Pod network reachability”, but not the required result. For this case, create two controlled faulty workloads, one requiring an impossible node label and one referencing a missing image tag, then compare Events.

D:

This check would help evaluate the distinct objective “Distinguish a container image from a running container and its writable layer”, but not the required result. For this case, create two controlled faulty workloads, one requiring an impossible node label and one referencing a missing image tag, then compare Events.

Question 614

A logistics demo Pod never receives a nodeName due to missing accelerator nodes. A shortcut is proposed: “Treat all Pending Pods as registry authentication failures and reset image pull credentials without inspecting scheduling events.” Which technical consequence follows most closely?

  1. The scheduler may evaluate an unintended Pod group or fail to spread the intended replicas.
  2. The actual capacity or node selection problem remains because no node was assigned.
  3. The Pod continues restarting because Service DNS configuration is not the cause of the process exit.
  4. Unexpected data loss or orphaned sensitive volumes can result from misunderstanding reclaim and retention policies.

Explanation

Correct Answer

B:

The actual capacity or node selection problem remains because no node was assigned. The shortcut creates this risk because failedScheduling occurs before node binding; ImagePullBackOff occurs when a node's kubelet cannot obtain the container image.

Incorrect Answers

A:

That outcome concerns the distinct objective “Spread replicas across failure domains with topologySpreadConstraints”; it is not the most direct consequence of this shortcut. The relevant consequence is that the actual capacity or node selection problem remains because no node was assigned.

C:

That outcome concerns the distinct objective “Diagnose CrashLoopBackOff from restart history and previous logs”; it is not the most direct consequence of this shortcut. The relevant consequence is that the actual capacity or node selection problem remains because no node was assigned.

D:

That outcome concerns the distinct objective “Compare PV reclaim policies and explain data retention on PVC deletion”; it is not the most direct consequence of this shortcut. The relevant consequence is that the actual capacity or node selection problem remains because no node was assigned.

Question 615

A logistics demo Pod never receives a nodeName due to missing accelerator nodes. Which combined solution and independent check are most directly justified?

  1. Require verified API server certificates, properly scoped identities and audit policy rather than turning off certificate checks for convenience. Acceptance evidence: With synthetic users verify a trusted API connection, denied privilege escalation and appropriate audit entries for failed requests.
  2. Inspect probe configuration, Pod Ready conditions and restart events to identify whether traffic eligibility or container execution is failing. Acceptance evidence: Induce a temporary readiness failure in staging and verify traffic stops without a restart; then separately test controlled liveness failure.
  3. Choose service boundaries around business capabilities and operational needs, documenting communication cost, ownership and fault tolerance. Acceptance evidence: Model a synthetic two-service checkout dependency and compare rollout independence with failure and latency overhead against a single service.
  4. Read Pod Events and node assignment to distinguish a Pod that cannot be scheduled from a scheduled Pod whose image cannot be pulled. Acceptance evidence: Create two controlled faulty workloads, one requiring an impossible node label and one referencing a missing image tag, then compare Events.

Explanation

Correct Answer

D:

Read Pod Events and node assignment to distinguish a Pod that cannot be scheduled from a scheduled Pod whose image cannot be pulled. Acceptance evidence: Create two controlled faulty workloads, one requiring an impossible node label and one referencing a missing image tag, then compare Events. This pairing is justified because failedScheduling occurs before node binding; ImagePullBackOff occurs when a node's kubelet cannot obtain the container image.

Incorrect Answers

A:

This action and check address the distinct objective “Protect API server access using authenticated TLS clients and audited authorization” instead of the stated acceptance requirement. The matching action is to read Pod Events and node assignment to distinguish a Pod that cannot be scheduled from a scheduled Pod whose image cannot be pulled. Its confirming test is to create two controlled faulty workloads, one requiring an impossible node label and one referencing a missing image tag, then compare Events.

B:

This action and check address the distinct objective “Investigate readiness failures separately from liveness-related restarts” instead of the stated acceptance requirement. The matching action is to read Pod Events and node assignment to distinguish a Pod that cannot be scheduled from a scheduled Pod whose image cannot be pulled. Its confirming test is to create two controlled faulty workloads, one requiring an impossible node label and one referencing a missing image tag, then compare Events.

C:

This action and check address the distinct objective “Contrast cloud-native microservices with a monolith without treating either as inherently superior” instead of the stated acceptance requirement. The matching action is to read Pod Events and node assignment to distinguish a Pod that cannot be scheduled from a scheduled Pod whose image cannot be pulled. Its confirming test is to create two controlled faulty workloads, one requiring an impossible node label and one referencing a missing image tag, then compare Events.

Question 616

A charity's Pod has a nodeName but fails to download a private image with incorrect credentials. Which engineering approach is the most appropriate for this cloud-native case?

  1. Inspect Pod container last state, restart count and kubectl logs --previous when a process repeatedly exits after startup.
  2. Use the scheduler's node-binding decision to explain initial Pod placement, then inspect the chosen node's kubelet for local container lifecycle status.
  3. Read Pod Events and node assignment to distinguish a Pod that cannot be scheduled from a scheduled Pod whose image cannot be pulled.
  4. Define appropriate HTTP routing through an installed Ingress controller or Gateway API controller rather than expecting the API object alone to provision a proxy.

Explanation

Correct Answer

C:

The situation is that a charity's Pod has a nodeName but fails to download a private image with incorrect credentials. Read Pod Events and node assignment to distinguish a Pod that cannot be scheduled from a scheduled Pod whose image cannot be pulled. This resolves the issue because failedScheduling occurs before node binding; ImagePullBackOff occurs when a node's kubelet cannot obtain the container image. To verify, create two controlled faulty workloads, one requiring an impossible node label and one referencing a missing image tag, then compare Events.

Incorrect Answers

A:

This option addresses the distinct objective “Diagnose CrashLoopBackOff from restart history and previous logs”, not the immediate requirement. CrashLoopBackOff reflects kubelet restart backoff after container failures; the underlying cause may be application configuration, exit code or a failed liveness probe. For the stated case, failedScheduling occurs before node binding; ImagePullBackOff occurs when a node's kubelet cannot obtain the container image.

B:

This option addresses the distinct objective “Separate kube-scheduler decisions from kubelet execution”, not the immediate requirement. The scheduler selects a suitable node for an unscheduled Pod; kubelet on that node uses the runtime to start and monitor containers. For the stated case, failedScheduling occurs before node binding; ImagePullBackOff occurs when a node's kubelet cannot obtain the container image.

D:

This option addresses the distinct objective “Choose Ingress or Gateway API with an implementing controller for HTTP exposure”, not the immediate requirement. Ingress and Gateway API express routing intent, while an actual controller/dataplane implements TLS termination, listeners and backend forwarding. For the stated case, failedScheduling occurs before node binding; ImagePullBackOff occurs when a node's kubelet cannot obtain the container image.

Question 617

A charity's Pod has a nodeName but fails to download a private image with incorrect credentials. What infrastructure principle makes the correct treatment effective?

  1. ClusterRole defines permissions while ClusterRoleBinding makes them cluster-wide; RoleBinding limits effective namespaced scope even when referencing a ClusterRole.
  2. FailedScheduling occurs before node binding; ImagePullBackOff occurs when a node's kubelet cannot obtain the container image.
  3. A NotReady or pressure-affected node can stop hosting healthy Pods even when their Deployment and Service definitions remain valid.
  4. Metrics reveal aggregate trends, logs record events and traces connect operations across services, so each signal answers a different debugging question.

Explanation

Correct Answer

B:

The situation is that a charity's Pod has a nodeName but fails to download a private image with incorrect credentials. FailedScheduling occurs before node binding; ImagePullBackOff occurs when a node's kubelet cannot obtain the container image. It supports this action: Read Pod Events and node assignment to distinguish a Pod that cannot be scheduled from a scheduled Pod whose image cannot be pulled. To confirm, create two controlled faulty workloads, one requiring an impossible node label and one referencing a missing image tag, then compare Events.

Incorrect Answers

A:

That statement explains the distinct objective “Distinguish ClusterRoles from Roles and binding scope” rather than the cause in this case. The deciding mechanism here is that failedScheduling occurs before node binding; ImagePullBackOff occurs when a node's kubelet cannot obtain the container image.

C:

That statement explains the distinct objective “Use node conditions and allocatable capacity to diagnose NotReady workloads” rather than the cause in this case. The deciding mechanism here is that failedScheduling occurs before node binding; ImagePullBackOff occurs when a node's kubelet cannot obtain the container image.

D:

That statement explains the distinct objective “Use metrics, logs and traces together when investigating cross-service latency” rather than the cause in this case. The deciding mechanism here is that failedScheduling occurs before node binding; ImagePullBackOff occurs when a node's kubelet cannot obtain the container image.

Question 618

A charity's Pod has a nodeName but fails to download a private image with incorrect credentials. What verification result would most convincingly distinguish a working solution from a claim?

  1. Model a synthetic two-service checkout dependency and compare rollout independence with failure and latency overhead against a single service.
  2. Induce a temporary readiness failure in staging and verify traffic stops without a restart; then separately test controlled liveness failure.
  3. Run nslookup for a Service using short and namespace-qualified names from test Pods in two different namespaces.
  4. Create two controlled faulty workloads, one requiring an impossible node label and one referencing a missing image tag, then compare Events.

Explanation

Correct Answer

D:

The situation is that a charity's Pod has a nodeName but fails to download a private image with incorrect credentials. Create two controlled faulty workloads, one requiring an impossible node label and one referencing a missing image tag, then compare Events. This is the relevant test because failedScheduling occurs before node binding; ImagePullBackOff occurs when a node's kubelet cannot obtain the container image.

Incorrect Answers

A:

This check would help evaluate the distinct objective “Contrast cloud-native microservices with a monolith without treating either as inherently superior”, but not the required result. For this case, create two controlled faulty workloads, one requiring an impossible node label and one referencing a missing image tag, then compare Events.

B:

This check would help evaluate the distinct objective “Investigate readiness failures separately from liveness-related restarts”, but not the required result. For this case, create two controlled faulty workloads, one requiring an impossible node label and one referencing a missing image tag, then compare Events.

C:

This check would help evaluate the distinct objective “Verify CoreDNS resolution and namespace-qualified Service names”, but not the required result. For this case, create two controlled faulty workloads, one requiring an impossible node label and one referencing a missing image tag, then compare Events.

Question 619

A charity's Pod has a nodeName but fails to download a private image with incorrect credentials. The change review contains this proposal: “Treat all Pending Pods as registry authentication failures and reset image pull credentials without inspecting scheduling events.” What risk matters most?

  1. The actual capacity or node selection problem remains because no node was assigned.
  2. Cross-namespace calls can fail or resolve a same-named but unrelated local Service.
  3. The container may enter a restart loop before completing initialization, prolonging unavailability.
  4. A workload remains nonfunctional and the persistent data dependency is never satisfied.

Explanation

Correct Answer

A:

The situation is that a charity's Pod has a nodeName but fails to download a private image with incorrect credentials. The actual capacity or node selection problem remains because no node was assigned. The shortcut creates this risk because failedScheduling occurs before node binding; ImagePullBackOff occurs when a node's kubelet cannot obtain the container image.

Incorrect Answers

B:

That outcome concerns the distinct objective “Verify CoreDNS resolution and namespace-qualified Service names”; it is not the most direct consequence of this shortcut. The relevant consequence is that the actual capacity or node selection problem remains because no node was assigned.

C:

That outcome concerns the distinct objective “Understand readiness, liveness and startup probes as different signals”; it is not the most direct consequence of this shortcut. The relevant consequence is that the actual capacity or node selection problem remains because no node was assigned.

D:

That outcome concerns the distinct objective “Use Events and pod describe for failed volume mounting”; it is not the most direct consequence of this shortcut. The relevant consequence is that the actual capacity or node selection problem remains because no node was assigned.

Question 620

A charity's Pod has a nodeName but fails to download a private image with incorrect credentials. Which implementation together with matching acceptance evidence best meets this case?

  1. Configure an HPA for an eligible scalable controller using available resource or custom metrics and sensible minimum/maximum replicas. Acceptance evidence: Generate controlled CPU load, verify HPA current/target metric reporting and observe replica changes within configured limits.
  2. Give each Pod an appropriate ServiceAccount, disable automountServiceAccountToken where API access is unnecessary and test token audience and role scope. Acceptance evidence: Run a test web Pod without a token mount and verify service operation succeeds while in-cluster API calls are unavailable.
  3. Inspect probe configuration, Pod Ready conditions and restart events to identify whether traffic eligibility or container execution is failing. Acceptance evidence: Induce a temporary readiness failure in staging and verify traffic stops without a restart; then separately test controlled liveness failure.
  4. Read Pod Events and node assignment to distinguish a Pod that cannot be scheduled from a scheduled Pod whose image cannot be pulled. Acceptance evidence: Create two controlled faulty workloads, one requiring an impossible node label and one referencing a missing image tag, then compare Events.

Explanation

Correct Answer

D:

The situation is that a charity's Pod has a nodeName but fails to download a private image with incorrect credentials. Read Pod Events and node assignment to distinguish a Pod that cannot be scheduled from a scheduled Pod whose image cannot be pulled. Acceptance evidence: Create two controlled faulty workloads, one requiring an impossible node label and one referencing a missing image tag, then compare Events. This pairing is justified because failedScheduling occurs before node binding; ImagePullBackOff occurs when a node's kubelet cannot obtain the container image.

Incorrect Answers

A:

This action and check address the distinct objective “Use HorizontalPodAutoscaler metrics rather than ad hoc instance scripts” instead of the stated acceptance requirement. The matching action is to read Pod Events and node assignment to distinguish a Pod that cannot be scheduled from a scheduled Pod whose image cannot be pulled. Its confirming test is to create two controlled faulty workloads, one requiring an impossible node label and one referencing a missing image tag, then compare Events.

B:

This action and check address the distinct objective “Enforce workload service account token boundaries and avoid needless mounts” instead of the stated acceptance requirement. The matching action is to read Pod Events and node assignment to distinguish a Pod that cannot be scheduled from a scheduled Pod whose image cannot be pulled. Its confirming test is to create two controlled faulty workloads, one requiring an impossible node label and one referencing a missing image tag, then compare Events.

C:

The situation is that a charity's Pod has a nodeName but fails to download a private image with incorrect credentials. This action and check address the distinct objective “Investigate readiness failures separately from liveness-related restarts” instead of the stated acceptance requirement. The matching action is to read Pod Events and node assignment to distinguish a Pod that cannot be scheduled from a scheduled Pod whose image cannot be pulled. Its confirming test is to create two controlled faulty workloads, one requiring an impossible node label and one referencing a missing image tag, then compare Events.

Kubernetes and Cloud Native Associate (KCNA) Practice Tests