Nutanix Updated NCP-CN Exam Questions and Answers by lake

The NKPA course explains that NKP’s monitoring stack includes Prometheus for metrics collection and storage, and Fluent Bit for log collection and forwarding. To retain system resource utilization data over several months, the company must focus on Prometheus, as it is responsible for storing time-series metrics data, such as CPU, memory, and network utilization, which are critical for long-term trend analysis.

To handle the increased workload from the online retail application, the Platform Engineer should adjust Prometheus by increasing its container resource limits and memory settings to ensure it can process and store more metrics, and increasing its storage to retain data for several months. The Nutanix Cloud Native (NCP-CN) 6.10 Study Guide states: “For long-term metrics retention in NKP, scale the Prometheus deployment by increasing its resource limits (CPU and memory) and expanding its persistent storage to accommodate larger time-series data.” This involves editing the Prometheus deployment’s resource settings (e.g., via kubectl edit deployment) and updating the PersistentVolumeClaim (PVC) to allocate more storage.

Incorrect Options:

A. Adjust the number of replicas for Fluent Bit: Fluent Bit handles logs, not metrics. Increasing replicas does not address long-term metrics storage.

B. Adjust the number of replicas for Prometheus: Increasing replicas improves availability but does not directly address storage or resource needs for metrics retention.

C. Adjust the resource settings for Fluent Bit: Fluent Bit is for log collection, not metrics storage, and is not relevant for this use case.

The NKPA course explains that upgrading an NKP license to the Ultimate tier unlocks advanced features such as fleet management, air-gapped deployments, and additional platform services (e.g., monitoring, logging, backup/restore). These services are deployed on the Management cluster and require sufficient resources to operate effectively. A key prerequisite is to size the Control Plane nodes appropriately to support the installation of default platform services (Option C).

The Control Plane nodes in the NKP Management cluster run critical components like the API server, controller manager, and Kommander, as well as platform services such as Prometheus, Grafana Loki, and Velero. The Ultimate license increases resource demands due to the expanded functionality, and undersized Control Plane nodes can lead to performance issues or failed deployments. The Nutanix Cloud Native (NCP-CN) 6.10 Study Guide states: “Before upgrading to an NKP Ultimate license, ensure the Control Plane nodes of the Management cluster are sized appropriately (e.g., sufficient CPU, memory, and storage) to support the additional default platform services enabled by the license.” For example, Control Plane nodes might need to be scaled to at least 4 vCPUs and 16 GB RAM, depending on the workload.

Incorrect Options:

A. Size the Sidecar containers appropriately: Sidecar containers are typically used for specific workloads (e.g., logging agents), not for platform services deployment, and are not a licensing prerequisite.

B. Size the ETCD nodes appropriately: While etcd is part of the Control Plane, the course refers to the Control Plane nodes as a whole, not etcd specifically, as the sizing requirement.

D. Size the Worker nodes appropriately: Worker nodes run user workloads, not the platform services, which are deployed on the Control Plane in the Management cluster.

As per the NKPA 6.10 documentation under “Application Backup and Recovery with Velero,” Velero is the out-of-the-box backup and restore solution integrated with NKP. To schedule application backups at a specific frequency (in this case, every six hours, equivalent to four times daily), the recommended approach is to use the Velero CLI to create a backup schedule.

The procedure involves:

Using the Velero CLI to create a scheduled backup targeting the specific Kubernetes namespace (victory-finance) rather than the entire workspace.

Specifying the frequency as --schedule="0 */6 * * *" (every six hours).

Providing the cluster’s kubeconfig for authentication.

Exact extract from the documentation:

“Velero’s CLI can be used to create backup schedules for application-specific workloads. Use the namespace parameter to target the namespace where the applications are deployed, and set the cron-like schedule to the desired frequency.”

The NKPA 6.10 documentation clarifies that when creating a workload cluster using the nkp create cluster command, specifying the target workspace (namespace) is critical for properly associating the workload cluster with that workspace in the NKP UI. If the --namespace parameter is omitted, the cluster is provisioned in the default workspace, not in the intended workspace (in this case, code-green).

Key documentation excerpt:

“If you do not specify the workspace (namespace) using the --namespace parameter when creating a cluster, the cluster will be created in the default workspace. It will not appear in custom workspaces until manually assigned.”