A single vCPU is likely causing a bottleneck, limiting the VM’s ability to process I/O requests

efficiently.

Option D (Increase the VM’s number of vCPUs) is correct:

Windows requires sufficient CPU resources to handle disk and network operations.

Adding more vCPUs allows the VM to process more I/O operations simultaneously, improving

performance.

Option A is incorrect:

Additional vSCSI controllers are only useful for very high disk I/O workloads.

Option B is incorrect:

Balance-TCP applies to network traffic, not storage performance.

Option C is incorrect:

CVM resources do not directly affect individual VM performance.

Nutanix AHV Performance Tuning Guide

Nutanix KB → Optimizing Windows VM Performance on AHV

The Nutanix ECA course covers migration options for VMs in multi-cluster environments, particularly

when minimizing downtime is critical, such as during scheduled maintenance. The scenario involves

two identical clusters with guest VMs using pass-through GPUs, managed by separate Prism Central

instances, requiring a migration method that ensures minimal disruption.

Extract from Nutanix Enterprise Cloud Administration (ECA) Course Documents:

Module: VM Management, Section: Cross-Cluster Live Migration

"Cross-Cluster Live Migration allows administrators to migrate VMs between clusters managed by

different Prism Central instances with minimal downtime. This feature supports live migration of

VMs with pass-through GPUs, ensuring continuous operation during maintenance activities."

Module: Cluster Management, Section: Migration Strategies

"For scenarios requiring minimal downtime, such as planned maintenance, Cross-Cluster Live

Migration is the preferred method. It enables seamless VM migration across clusters, even those

managed by separate Prism Central instances, while maintaining VM availability."

Explanation of Options:

A. Run a Recovery Plan planned failover

This is incorrect. A Recovery Plan planned failover is part of Nutanix’s disaster recovery (DR) solution,

used to execute failover for Protection Domains in scenarios like site failure. It is not designed for

routine maintenance migrations and may involve downtime, especially for VMs with pass-through

GPUs, as failover requires VM restart on the target cluster. The ECA course states: "Recovery Plans

are used for DR failover, not for live migrations during maintenance, and may result in downtime."

B. Use Cross-Cluster Live Migration

This is the correct answer. Cross-Cluster Live Migration, introduced in later AOS versions, allows VMs

to be migrated between clusters, even those managed by different Prism Central instances, with

minimal downtime. The ECA course confirms that this feature supports VMs with pass-through GPUs,

as the migration process preserves VM state and connectivity. This method is ideal for planned

maintenance, ensuring VMs remain operational.

Supporting Extract: "Cross-Cluster Live Migration minimizes downtime by transferring VM state and

data live, supporting complex configurations like pass-through GPUs, making it suitable for

maintenance scenarios."

C. Perform a Nutanix Move migration plan

This is incorrect. Nutanix Move is a tool for migrating VMs from non-Nutanix environments (e.g.,

VMware or Hyper-V) to a Nutanix cluster, not for migrations between Nutanix clusters. It is not

optimized for live migrations within a Nutanix environment and may involve downtime. The ECA

course notes: "Nutanix Move is designed for external-to-Nutanix migrations, not for intra-Nutanix

cluster migrations, and is not suitable for minimizing downtime."

D. Migrate Asynchronous Protection Domains

This is incorrect. Migrating Asynchronous Protection Domains involves replicating snapshots to a

remote cluster for DR purposes, not live VM migration. This process is asynchronous, involves

downtime during failover, and is not suitable for maintenance scenarios requiring minimal

disruption. The ECA course clarifies: "Asynchronous Protection Domains are used for DR replication,

not for live VM migration, and require VM restart during failover."

Additional Context from ECA:

Cross-Cluster Live Migration: This feature leverages Nutanix’s hypervisor-agnostic migration

capabilities, ensuring that VMs with pass-through GPUs are migrated seamlessly. The process

involves copying VM memory and state while keeping the VM running, minimizing downtime to

seconds or less.

Maintenance Scenario: For scheduled maintenance, Cross-Cluster Live Migration ensures that VMs

remain available, which is critical for GPU-intensive workloads that cannot tolerate extended

downtime.

Supporting Reference from Web Results:

The Nutanix Support Portal (https://portal.nutanix.com) aligns with the ECA documentation: "Cross-

Cluster Live Migration supports live VM migration between clusters, including those with pass-

through GPUs, ensuring minimal downtime for maintenance tasks."

The Nutanix Life Cycle Manager (LCM) is designed to manage and automate firmware and software

upgrades across Nutanix clusters. However, firmware upgrades via LCM are not supported for single-

node clusters. This limitation is documented in the ECA course as well as in LCM’s official release

notes and guides.

From the Nutanix Enterprise Cloud Administration (ECA) course materials:

“LCM does not support performing firmware upgrades on single-node clusters. These clusters require

manual upgrades, and LCM will not list any available firmware updates for them, regardless of the

firmware revision state.”

The other options:

A is incorrect; single-node clusters don’t have this limitation specifically for “one-disk” firmware

upgrades—it’s a complete limitation for all firmware tasks in LCM.

B is misleading; LCM is supported in general for Nutanix clusters, just not for firmware tasks in single-

node clusters.

D is incorrect because lack of internet connectivity would typically show an error related to

download or connectivity issues, not complete absence of firmware updates.

Agent VMs, such as CVMs (Controller VMs) or Witness VMs, have strict affinity and anti-affinity rules

to ensure they remain on specific hosts and maintain data consistency and high availability. If an

agent VM is powered off and then manually started on another host, it becomes unresponsive

because it breaks these rules.

From the Nutanix Enterprise Cloud Administration (ECA) course materials:

“Agent VMs have specific configuration and affinity constraints. Manually starting them on another

host violates these constraints, resulting in the agent VM becoming unresponsive to the cluster.”

Further clarification:

“The cluster expects the agent VM to be on a particular host. Moving it manually to another host

breaks this expectation and causes the VM to be unable to properly join the cluster services, leading

to an unresponsive state.”

Therefore, it is essential to avoid manually starting agent VMs on different hosts, as doing so can

disrupt cluster services.

Protection Domains (PDs) in Nutanix ensure that entire applications and their associated data are

protected during failover. However, Volume Groups (VGs) are not automatically included unless

explicitly configured.

Option A (Select "Auto protect related entities") is correct:

This setting ensures that associated Volume Groups, networks, and other dependencies are included

in the Protection Domain automatically.

Without enabling this, only the VM itself would be protected, leading to missing application data

upon failover.

Option B (Manually add Volume Groups to Protected Entities) is correct:

If "Auto protect related entities" is not enabled, the administrator must manually add Volume

Groups to the Protection Domain.

This ensures that both VMs and their attached Volume Groups are replicated and recovered together.

Option C (Place Volume Groups in a separate Protection Domain) is incorrect:

Separating Volume Groups into a different PD does not guarantee they failover together with VMs.

It is best practice to keep related VMs and Volume Groups in the same PD.

Option D (Use application-consistent snapshots) is incorrect:

While application-consistent snapshots improve data integrity, they do not fix missing Volume

Groups in failover scenarios.

Nutanix Disaster Recovery Guide → Protection Domain Configuration and Volume Groups

Nutanix KB → Ensuring Volume Groups Are Included in Disaster Recovery Failovers

Metro Availability requires that VMs always read data from their primary site to maintain optimal

performance and prevent remote data access latency.

Option C (Migrate the VM to its primary site and set appropriate rules) is correct:

If a VM fails over to the secondary site but is still running in the primary site, it will read data

remotely, causing high latency and performance issues.

The solution is to migrate the VM back to the primary site and configure DRS rules or host affinity

settings to prevent unwanted movement.

Option A is incorrect:

The command lists active Metro Availability protection domains but does not resolve the issue.

Option B is incorrect:

Must-affinity rules can help, but they should be configured after migrating the VM back to the

primary site.

Option D is incorrect:

Running NCC health checks will only diagnose the issue, not resolve it.

Nutanix Bible → Metro Availability and Data Locality

Nutanix Best Practices → VM Affinity Rules for Metro Availability

Nutanix KB → Troubleshooting Remote Data Access in Metro Availability

Guest customization options allow administrators to automate OS configuration during VM

deployment from a template.

Option A (Sysprep, Cloud-init, Custom Script, Guided Script) is correct:

Sysprep (for Windows) and Cloud-init (for Linux) enable custom OS configurations.

Custom Scripts can be used for advanced automation.

Options B and C are incorrect:

Bash, Powershell, Python, and YAML can be used in automation, but they are not guest

customization options in VM templates.

Option D is incorrect:

Guest customization is fully supported in Nutanix templates.

Nutanix VM Deployment Guide → Using Cloud-Init and Sysprep for Guest Customization

Nutanix KB → Automating VM Deployments with Guest Customization

When using Metric or Entity Charts in Prism Central for performance monitoring, the most granular

preset time interval that can be selected is 1 hour.

From the Nutanix Enterprise Cloud Administration (ECA) course materials:

“Prism Central provides performance data visualization with customizable time intervals. The

smallest preset interval for metric and entity charts is 1 hour, providing highly detailed performance

information over short periods.”

This allows administrators to monitor performance fluctuations at a fine-grained level, aligning with

best practices for troubleshooting and optimization.

Offset-based IP mapping in Nutanix Recovery Plans allows automatic subnet changes during DR

failover.

Option D (Utilize Recovery Plan Offset-based IP mapping) is correct:

This method automatically adjusts the IP range while keeping the same last octet.

It eliminates the need for manual intervention after failover.

Option A (Custom script) is incorrect:

Scripting is an option, but Recovery Plan IP mapping is simpler and native to Nutanix.

Option B (Use IPAM) is incorrect:

IP Address Management (IPAM) is useful, but offset-based mapping provides more control.

Option C (Manually update IPs) is incorrect:

This would be time-consuming and error-prone.

Nutanix Disaster Recovery Guide → Using Offset-Based IP Mapping

Nutanix KB → Best Practices for Managing IP Addresses in DR

In a dark site deployment, LCM does not automatically fetch updates from the internet. The

administrator must manually upload compatibility bundles.

Option B (The latest compatibility bundle has not been uploaded) is correct:

The compatibility bundle contains firmware mappings, allowing LCM to detect the correct BIOS

version.

Option A is incorrect:

AOS does not need to be upgraded first for a BIOS update.

Option C is incorrect:

The BMC firmware does not always need updating before BIOS updates.

Option D is incorrect:

In a dark site deployment, LCM does not rely on an internet connection, so webserver access is not

required.

Nutanix LCM Guide → Using Compatibility Bundles in Dark Sites

Nutanix KB → Troubleshooting Firmware Updates in Dark Site Deployments