Question 190A storage policy requires encrypted volumes but an engineer creates a StorageClass without the provider's encryption parameters. Which Kubernetes administration action most directly addresses the observed failure? Select a StorageClass with an installed valid provisioner or install the required CSI driver. Define the correct CSI StorageClass parameters and verify the provisioned volume security properties. Install and verify the CSI node plugin on every eligible attachment node. Set exactly the intended default class annotation and specify storageClassName explicitly for critical claims. ExplanationCorrect Answer B:Define the correct CSI StorageClass parameters and verify the provisioned volume security properties. This corrects the specific failure because storageClass parameters are driver-specific and determine backend provisioning behavior. Verify it by checking: A test PV is created with the required encryption setting reported by the storage provider.Incorrect Answers A:Dynamic provisioning depends on a controller that recognizes the StorageClass provisioner identifier. That is relevant to a different intervention. For the stated situation, the required action is: Define the correct CSI StorageClass parameters and verify the provisioned volume security properties. The governing reason is: StorageClass parameters are driver-specific and determine backend provisioning behavior. C:A controller may provision a volume while the node-side CSI plugin is still required to mount it. That is relevant to a different intervention. For the stated situation, the required action is: Define the correct CSI StorageClass parameters and verify the provisioned volume security properties. The governing reason is: StorageClass parameters are driver-specific and determine backend provisioning behavior. D:PVCs without a specified class can inherit the cluster's default storage class. That is relevant to a different intervention. For the stated situation, the required action is: Define the correct CSI StorageClass parameters and verify the provisioned volume security properties. The governing reason is: StorageClass parameters are driver-specific and determine backend provisioning behavior.
Question 191A namespace maintainer can delete Deployments in every team namespace after a broad ClusterRoleBinding was applied. Following the proposed change—Replace the cluster-wide binding with a namespaced RoleBinding granting only needed resource verbs—which observation best demonstrates that the particular fault is resolved? The viewer can list only the approved custom resources in the bound namespace after reconciliation. kubectl auth can-i list configmaps -n billing --as=<identity> changes to yes without granting delete rights. Impersonated requests reach the approved subresource and reject unrelated secret reads. An impersonated authorization check denies deletion in another team's namespace but permits the required local update. ExplanationCorrect Answer D:An impersonated authorization check denies deletion in another team's namespace but permits the required local update. This validates the stated intervention—replace the cluster-wide binding with a namespaced RoleBinding granting only needed resource verbs.—rather than an unrelated cluster change. The relevant mechanism is: A RoleBinding restricts a Role or referenced ClusterRole to its own namespace; a ClusterRoleBinding grants across namespaces.Incorrect Answers A:Aggregated ClusterRoles rely on matching labels on source ClusterRoles; ordinary ClusterRoles do not aggregate automatically. This relates to a separate check and does not verify the specified repair. The relevant acceptance evidence is: An impersonated authorization check denies deletion in another team's namespace but permits the required local update. B:Authentication establishes identity while authorization separately decides permitted API verbs and resources. This relates to a separate check and does not verify the specified repair. The relevant acceptance evidence is: An impersonated authorization check denies deletion in another team's namespace but permits the required local update. C:RBAC treats interactive Pod subresources separately from operations on the parent Pod resource. This relates to a separate check and does not verify the specified repair. The relevant acceptance evidence is: An impersonated authorization check denies deletion in another team's namespace but permits the required local update.
Question 192A Secret update is expected to replace a running process's environment value immediately without restarting the container. An engineer recommends: Roll out the workload or make the application reread supported projected files as required. Which Kubernetes behavior best explains why this correction fits? Environment variables sourced from Secrets are fixed for the lifetime of a container process. Secret key references are exact and missing mandatory keys prevent valid container configuration. ConfigMaps separate nonsecret configuration from the application container image. ConfigMap volumes can update eventually, whereas a subPath-mounted key does not receive automatic updates. ExplanationCorrect Answer A:Environment variables sourced from Secrets are fixed for the lifetime of a container process. This explains why the proposed action—roll out the workload or make the application reread supported projected files as required.—is appropriate. Direct confirmation is: New Pods use the updated Secret value and old processes terminate in a controlled rollout.Incorrect Answers B:Secret key references are exact and missing mandatory keys prevent valid container configuration. This does not establish the proposed intervention. The governing principle here is: Environment variables sourced from Secrets are fixed for the lifetime of a container process. Apply it by: Roll out the workload or make the application reread supported projected files as required. C:ConfigMaps separate nonsecret configuration from the application container image. This does not establish the proposed intervention. The governing principle here is: Environment variables sourced from Secrets are fixed for the lifetime of a container process. Apply it by: Roll out the workload or make the application reread supported projected files as required. D:ConfigMap volumes can update eventually, whereas a subPath-mounted key does not receive automatic updates. This does not establish the proposed intervention. The governing principle here is: Environment variables sourced from Secrets are fixed for the lifetime of a container process. Apply it by: Roll out the workload or make the application reread supported projected files as required.
Question 193A Service sends requests to a Pod that reports Running but fails readiness checks. Which Kubernetes administration action most directly addresses the observed failure? Correct the Service selector to match the intended ready Pod labels. Make the Pod's readiness probe meaningful and confirm only ready endpoints receive traffic. Return to ClusterIP and remove the unnecessary external frontend. Document and use headless DNS/endpoint discovery for per-Pod identities instead of VIP load balancing. ExplanationCorrect Answer B:Make the Pod's readiness probe meaningful and confirm only ready endpoints receive traffic. This corrects the specific failure because readiness controls endpoint eligibility; Pod phase Running alone does not guarantee serving capacity. Verify it by checking: EndpointSlices exclude the unready Pod and traffic flows only to healthy replicas.Incorrect Answers A:Selector-backed Services route through EndpointSlices populated from matching eligible Pods. That is relevant to a different intervention. For the stated situation, the required action is: Make the Pod's readiness probe meaningful and confirm only ready endpoints receive traffic. The governing reason is: Readiness controls endpoint eligibility; Pod phase Running alone does not guarantee serving capacity. C:ClusterIP is intended for internal service discovery; LoadBalancer requests external integration where supported. That is relevant to a different intervention. For the stated situation, the required action is: Make the Pod's readiness probe meaningful and confirm only ready endpoints receive traffic. The governing reason is: Readiness controls endpoint eligibility; Pod phase Running alone does not guarantee serving capacity. D:A headless Service (clusterIP None) publishes endpoint records rather than a normal ClusterIP proxy target. That is relevant to a different intervention. For the stated situation, the required action is: Make the Pod's readiness probe meaningful and confirm only ready endpoints receive traffic. The governing reason is: Readiness controls endpoint eligibility; Pod phase Running alone does not guarantee serving capacity.
Question 194A release of a payment API introduces an invalid image tag and the Deployment cannot complete its rolling update. Which Kubernetes administration action most directly addresses the observed failure? Configure rolling-update maxUnavailable and maxSurge according to available capacity and readiness requirements. Apply the intended image and observe rollout status and updated ReplicaSets before declaring success. Define a Job with suitable completions, parallelism and retry policy instead of a long-lived Deployment. Use rollout history to identify the last healthy revision and roll back the Deployment to it. ExplanationCorrect Answer D:Use rollout history to identify the last healthy revision and roll back the Deployment to it. This corrects the specific failure because deployment revisions track supported rollout history and a rollback restores a previous Pod template. Verify it by checking: kubectl rollout status shows the prior revision fully available and application probes succeed.Incorrect Answers A:The Deployment rollout strategy controls the allowed temporary excess and unavailable replicas. That is relevant to a different intervention. For the stated situation, the required action is: Use rollout history to identify the last healthy revision and roll back the Deployment to it. The governing reason is: Deployment revisions track supported rollout history and a rollback restores a previous Pod template. B:Desired-state changes are asynchronous and a successful API update is not proof of healthy Pods. That is relevant to a different intervention. For the stated situation, the required action is: Use rollout history to identify the last healthy revision and roll back the Deployment to it. The governing reason is: Deployment revisions track supported rollout history and a rollback restores a previous Pod template. C:Jobs manage successful completion of finite work; Deployments maintain continuous desired replicas. That is relevant to a different intervention. For the stated situation, the required action is: Use rollout history to identify the last healthy revision and roll back the Deployment to it. The governing reason is: Deployment revisions track supported rollout history and a rollback restores a previous Pod template.
Question 195A team adds an HPA to a Deployment whose Pods lack CPU requests, producing unknown CPU utilization. An engineer recommends: Set defensible CPU requests and confirm metrics-server data before relying on percentage-based HPA. Which Kubernetes behavior best explains why this correction fits? HPA CPU utilization calculations depend on resource requests and relevant metrics API data. Different workload bottlenecks require appropriately supported HPA metrics. HPA may delay downscaling to avoid flapping even after metrics fall. CPU percentage targets use utilization relative to requested CPU, not just absolute node usage. ExplanationCorrect Answer D:CPU percentage targets use utilization relative to requested CPU, not just absolute node usage. This explains why the proposed action—set defensible CPU requests and confirm metrics-server data before relying on percentage-based HPA.—is appropriate. Direct confirmation is: The HPA stops reporting unknown metrics and scales within configured bounds.Incorrect Answers A:HPA CPU utilization calculations depend on resource requests and relevant metrics API data. This does not establish the proposed intervention. The governing principle here is: CPU percentage targets use utilization relative to requested CPU, not just absolute node usage. Apply it by: Set defensible CPU requests and confirm metrics-server data before relying on percentage-based HPA. B:Different workload bottlenecks require appropriately supported HPA metrics. This does not establish the proposed intervention. The governing principle here is: CPU percentage targets use utilization relative to requested CPU, not just absolute node usage. Apply it by: Set defensible CPU requests and confirm metrics-server data before relying on percentage-based HPA. C:HPA may delay downscaling to avoid flapping even after metrics fall. This does not establish the proposed intervention. The governing principle here is: CPU percentage targets use utilization relative to requested CPU, not just absolute node usage. Apply it by: Set defensible CPU requests and confirm metrics-server data before relying on percentage-based HPA.
Question 196A planned kubeadm control-plane upgrade is attempted without first taking a recoverable etcd snapshot. Which Kubernetes administration action most directly addresses the observed failure? Resolve compatibility and required preflight conditions before applying the approved upgrade target. Take, securely store, and verify an etcd backup as part of the documented upgrade rollback plan. Follow Kubernetes version-skew rules and upgrade the control plane and workers in the supported order. Inspect the PodDisruptionBudget and safely increase available replicas or adjust policy before retrying drain. ExplanationCorrect Answer B:Take, securely store, and verify an etcd backup as part of the documented upgrade rollback plan. This corrects the specific failure because cluster state durability requires more than rerunning kubeadm; an etcd snapshot protects API objects. Verify it by checking: The backup has the expected revision and an isolated restore rehearsal can list known objects.Incorrect Answers A:kubeadm upgrade plan surfaces supported upgrade paths and potential blockers, not just an install command. That is relevant to a different intervention. For the stated situation, the required action is: Take, securely store, and verify an etcd backup as part of the documented upgrade rollback plan. The governing reason is: Cluster state durability requires more than rerunning kubeadm; an etcd snapshot protects API objects. C:Component version skew is constrained; a newer kubelet must not be assumed compatible with an older API server. That is relevant to a different intervention. For the stated situation, the required action is: Take, securely store, and verify an etcd backup as part of the documented upgrade rollback plan. The governing reason is: Cluster state durability requires more than rerunning kubeadm; an etcd snapshot protects API objects. D:The eviction API respects PodDisruptionBudget limits for voluntary disruption. That is relevant to a different intervention. For the stated situation, the required action is: Take, securely store, and verify an etcd backup as part of the documented upgrade rollback plan. The governing reason is: Cluster state durability requires more than rerunning kubeadm; an etcd snapshot protects API objects.
Question 197An incident responder has permission to create Pods but not to attach interactively to running containers. Which Kubernetes administration action most directly addresses the observed failure? Bind Secret get permissions only to that service account for the required namespace and named resources when feasible. Replace the cluster-wide binding with a namespaced RoleBinding granting only needed resource verbs. Grant the narrowly required pods/attach or pods/exec subresource verbs according to the approved response workflow. Replace wildcard rules with explicit get list watch permissions on the exact resources the collector reads. ExplanationCorrect Answer C:Grant the narrowly required pods/attach or pods/exec subresource verbs according to the approved response workflow. This corrects the specific failure because rBAC treats interactive Pod subresources separately from operations on the parent Pod resource. Verify it by checking: Impersonated requests reach the approved subresource and reject unrelated secret reads.Incorrect Answers A:Kubernetes service accounts do not automatically inherit administrator permissions from the Pod owner. That is relevant to a different intervention. For the stated situation, the required action is: Grant the narrowly required pods/attach or pods/exec subresource verbs according to the approved response workflow. The governing reason is: RBAC treats interactive Pod subresources separately from operations on the parent Pod resource. B:A RoleBinding restricts a Role or referenced ClusterRole to its own namespace; a ClusterRoleBinding grants across namespaces. That is relevant to a different intervention. For the stated situation, the required action is: Grant the narrowly required pods/attach or pods/exec subresource verbs according to the approved response workflow. The governing reason is: RBAC treats interactive Pod subresources separately from operations on the parent Pod resource. D:Least-privilege RBAC narrows blast radius by limiting the allowed verbs, resources, and API groups. That is relevant to a different intervention. For the stated situation, the required action is: Grant the narrowly required pods/attach or pods/exec subresource verbs according to the approved response workflow. The governing reason is: RBAC treats interactive Pod subresources separately from operations on the parent Pod resource.
Question 198An operator applies the development Kustomize base directly into production, bypassing a production-only resource patch. Following the proposed change—Apply the intended production overlay and validate the build output before kubectl apply—which observation best demonstrates that the particular fault is resolved? kustomize build displays the expected name and workload references after an overlay change. The generated manifest includes production image, replicas, namespace and labels without development-only settings. helm get values and rendered manifests show the change, which survives the next upgrade. The reviewed rendered difference contains only intended resource changes before promotion. ExplanationCorrect Answer B:The generated manifest includes production image, replicas, namespace and labels without development-only settings. This validates the stated intervention—apply the intended production overlay and validate the build output before kubectl apply.—rather than an unrelated cluster change. The relevant mechanism is: Bases compose reusable manifests while overlays encode environment-specific differences.Incorrect Answers A:Kustomize may hash generated ConfigMap names to make content changes trigger referential updates. This relates to a separate check and does not verify the specified repair. The relevant acceptance evidence is: The generated manifest includes production image, replicas, namespace and labels without development-only settings. C:Helm's release history and desired manifests should remain the authoritative source for chart-managed resources. This relates to a separate check and does not verify the specified repair. The relevant acceptance evidence is: The generated manifest includes production image, replicas, namespace and labels without development-only settings. D:Template rendering plus a values diff exposes chart-side effects before they affect cluster resources. This relates to a separate check and does not verify the specified repair. The relevant acceptance evidence is: The generated manifest includes production image, replicas, namespace and labels without development-only settings.
Question 199A PVC remains Pending although a statically created PV exists with the same capacity because access modes differ. An engineer recommends: Match the claim's accessModes, storage class and capacity to the intended static PV. Which Kubernetes behavior best explains why this correction fits? PVCs are namespace-scoped and Pods cannot mount another namespace's claim by name. A Released PV may preserve its previous claim reference until an administrator makes it available. A PVC must satisfy PV binding predicates, not merely request a compatible size. A Bound PVC proves allocation but not that attach, stage and publish operations have succeeded. ExplanationCorrect Answer C:A PVC must satisfy PV binding predicates, not merely request a compatible size. This explains why the proposed action—match the claim's accessModes, storage class and capacity to the intended static PV.—is appropriate. Direct confirmation is: The PVC binds the expected PV and a consumer Pod mounts it successfully.Incorrect Answers A:PVCs are namespace-scoped and Pods cannot mount another namespace's claim by name. This does not establish the proposed intervention. The governing principle here is: A PVC must satisfy PV binding predicates, not merely request a compatible size. Apply it by: Match the claim's accessModes, storage class and capacity to the intended static PV. B:A Released PV may preserve its previous claim reference until an administrator makes it available. This does not establish the proposed intervention. The governing principle here is: A PVC must satisfy PV binding predicates, not merely request a compatible size. Apply it by: Match the claim's accessModes, storage class and capacity to the intended static PV. D:A Bound PVC proves allocation but not that attach, stage and publish operations have succeeded. This does not establish the proposed intervention. The governing principle here is: A PVC must satisfy PV binding predicates, not merely request a compatible size. Apply it by: Match the claim's accessModes, storage class and capacity to the intended static PV.
Question 200An operator reconciles a custom resource but never creates the expected Deployment after its controller Pod crashes. Which Kubernetes administration action most directly addresses the observed failure? Grant the controller the minimum required update or patch permission for the status subresource. Review finalizer ownership and perform a supported cleanup sequence before removing the responsible controller. Restore the controller/operator and check reconciliation logs and permissions before changing the custom object. Install the CRD and wait for its Established condition before applying dependent custom resources. ExplanationCorrect Answer C:Restore the controller/operator and check reconciliation logs and permissions before changing the custom object. This corrects the specific failure because a CRD defines the API schema while a running controller implements the desired behavior. Verify it by checking: The operator returns healthy and the managed Deployment appears with the expected owner references.Incorrect Answers A:Custom resource status updates are separately authorized from spec updates when status is enabled. That is relevant to a different intervention. For the stated situation, the required action is: Restore the controller/operator and check reconciliation logs and permissions before changing the custom object. The governing reason is: A CRD defines the API schema while a running controller implements the desired behavior. B:Finalizers block object deletion until the associated controller completes the required cleanup. That is relevant to a different intervention. For the stated situation, the required action is: Restore the controller/operator and check reconciliation logs and permissions before changing the custom object. The governing reason is: A CRD defines the API schema while a running controller implements the desired behavior. D:The API server must register an extension type before it can accept objects of that kind. That is relevant to a different intervention. For the stated situation, the required action is: Restore the controller/operator and check reconciliation logs and permissions before changing the custom object. The governing reason is: A CRD defines the API schema while a running controller implements the desired behavior.