Microsoft Azure Interview Questions – Part 2: Azure VMs, Compute, Disks & Availability

Networking concepts such as VNets, NSGs, routing, Load Balancer, Application Gateway, VPN Gateway and Private Endpoints were covered in Part 1 in Microsoft Azure Interview Questions & Answers – Complete Series , so they are not unnecessarily repeated here.

Continue the Microsoft Azure Interview Series

← Previous Part: [Part 1: Azure Networking Fundamentals, VNets, Subnets, NSGs & Connectivity] | Complete Series: [Microsoft Azure Interview Questions & Answers – Complete Series] | Next Part →: [Part 3: Azure Storage, Blob Storage, Azure Files, Redundancy & Troubleshooting]

Azure Virtual Machines – Interview Questions & Answers

1. What is an Azure Virtual Machine?

An Azure Virtual Machine is an IaaS compute resource that provides a virtualized server running Windows or Linux in Microsoft Azure.

A VM gives you control over:

Operating system
Applications
Installed software
Disks
Network interfaces
Security configuration
VM size
Extensions
Monitoring and management

Azure manages the underlying physical infrastructure, while you manage the guest operating system and workload according to the service model.

Example

You can deploy:

Windows Server 2022
Windows Server 2025
Ubuntu
Red Hat Enterprise Linux
SUSE Linux
Other supported Marketplace or custom images

2. What are the main components of an Azure VM?

An Azure VM normally consists of:

VM compute resource
OS disk
Optional data disks
Network Interface Card (NIC)
Virtual Network and subnet
Public IP, if required
Network Security Group, if configured
VM Agent
VM extensions, if used
Monitoring and diagnostics components

The VM itself is the compute resource, while storage and networking are separate Azure resources that can be associated with it.

3. What is an Azure VM size?

A VM size determines the compute capabilities allocated to the VM.

Depending on the VM family and SKU, it determines characteristics such as:

Number of vCPUs
Memory
Maximum data disks
Maximum network interfaces
Network bandwidth
Temporary storage
Supported disk capabilities
Specialized hardware capabilities

Example

A general-purpose VM might have:

4 vCPUs
16 GiB RAM
Multiple data disks
Premium SSD support

The exact capabilities depend on the selected VM SKU.

4. What are Azure VM size families?

Azure provides different VM families optimized for different workloads.

Examples include:

General purpose

Examples:

D-series
B-series

Suitable for:

Application servers
Domain controllers
Web servers
Development environments
Small-to-medium workloads

Compute optimized

Examples:

F-series

Suitable for workloads requiring more CPU relative to memory.

Memory optimized

Examples:

E-series
M-series

Suitable for:

Large databases
In-memory workloads
Analytics

Storage optimized

Examples:

L-series

Designed for workloads requiring high local storage performance.

GPU optimized

Examples include GPU-enabled VM families for:

AI
Machine learning
Rendering
Visualization
High-performance computing

Always verify the exact SKU capabilities for the target Azure region before deployment.

5. What is the difference between vCPU and physical CPU?

A vCPU is a virtual CPU presented to the guest operating system.

Azure abstracts the underlying physical processor infrastructure.

For example:

Physical Azure Host
       |
       +--- VM 1 → 4 vCPU
       |
       +--- VM 2 → 8 vCPU
       |
       +--- VM 3 → 2 vCPU

The number of vCPUs available to a VM depends on its selected VM size.

6. How do you select the right Azure VM size?

Do not select a VM based only on CPU or RAM.

Evaluate:

CPU utilization
Memory utilization
Disk IOPS
Disk throughput
Network requirements
Number of data disks
Application architecture
Availability requirements
Region/SKU availability
Cost
Licensing requirements

Example

For a database server:

CPU → workload dependent
Memory → potentially high
Disk → high IOPS/throughput
Network → potentially high
Availability → multiple instances/zones where appropriate

The correct VM size is determined by the workload’s actual resource requirements.

7. What is the difference between VM size and managed disk size?

They are completely different.

VM size

Determines compute resources such as:

vCPU
Memory
Network capability
Maximum disks

Managed disk size

Determines storage capacity and, depending on disk type and configuration, storage performance characteristics.

For example:

VM:
Standard_D4s_v5
4 vCPU
16 GiB RAM

OS Disk:
Premium SSD
128 GiB

Changing the VM size does not automatically change the disk type.

8. What is a Generation 1 and Generation 2 Azure VM?

Azure supports different VM generations.

Generation 1

Uses the traditional BIOS-based boot architecture.

Generation 2

Uses modern UEFI-based boot architecture.

Generation 2 supports features such as:

UEFI
Secure Boot when supported/configured
vTPM
Trusted Launch

Gen2 also supports larger OS disks beyond the traditional MBR 2-TiB partition limitation when the operating system and configuration use GPT appropriately.

Generation compatibility depends on the image and VM size.

9. What is Trusted Launch for Azure VMs?

Trusted Launch is a security configuration for supported Generation 2 VMs.

It provides security features including:

Secure Boot
vTPM
Measured boot capabilities

The purpose is to help protect the VM against boot-level threats and unauthorized boot components.

Trusted Launch should be considered when security requirements justify it and the selected OS/image and VM configuration support it.

10. What is Secure Boot?

Secure Boot is a UEFI security feature that allows the system firmware to verify that boot components are trusted before executing them.

It helps prevent unauthorized or malicious boot software from loading.

Secure Boot is associated with Generation 2 VMs and Trusted Launch configurations.

11. What is vTPM in Azure?

vTPM stands for virtual Trusted Platform Module.

It provides a virtualized TPM for the VM and can be used by supported operating systems and security features.

It is one of the security capabilities associated with Trusted Launch.

12. What is an Azure managed disk?

An Azure managed disk is a block-level storage volume managed by Azure and used with Azure VMs.

You specify characteristics such as:

Disk type
Disk size
Performance configuration

Azure manages the underlying storage infrastructure.

Current Azure managed disk types include:

Ultra Disk
Premium SSD v2
Premium SSD
Standard SSD
Standard HDD

13. What is the difference between an OS disk and a data disk?

OS disk

Contains the operating system and boot volume.

For a Windows VM, this is normally the C: drive.

Data disk

Used for:

Application data
Databases
Logs
File shares
Other persistent workload data

Recommended architecture

OS Disk
   ↓
Operating System

Data Disk 1
   ↓
Application

Data Disk 2
   ↓
Database

Data Disk 3
   ↓
Logs

Separating workload data from the OS can improve manageability, backup design and performance isolation.

14. What is the temporary disk in an Azure VM?

Most Azure VM sizes provide a temporary disk.

It is intended for temporary or short-lived data such as:

Page files
Swap
Temporary files
SQL Server TempDB
Application cache

It should not be used for important persistent application data.

Data on the temporary disk can be lost during events such as VM maintenance, redeployment or stopping the VM.

A normal restart can preserve temporary-disk data, but applications should never depend on that behavior for durable storage.

15. Where is the temporary disk located on Windows and Linux?

Common examples are:

Windows

D:

Linux

Often exposed through a device such as:

/dev/disk/azure/resource

The exact device/path can vary with the VM size and operating system configuration.

16. What is an ephemeral OS disk?

An ephemeral OS disk stores the VM’s operating system on local VM storage rather than using a conventional persistent managed OS disk.

It is designed primarily for workloads where the VM can be reprovisioned or reimaged and the OS state does not need to be permanently retained.

Advantages can include:

Faster VM deployment
Faster reimage
Lower OS-disk latency
Reduced dependency on remote OS-disk storage

It is particularly useful for stateless workloads.

17. What is the difference between a temporary disk and an ephemeral OS disk?

They are different concepts.

Feature	Temporary Disk	Ephemeral OS Disk
Purpose	Temporary workload storage	Operating system
Persistent	No	No
Managed disk	No	No
Used for OS	No	Yes
Suitable for permanent data	No	No
Typical use	Cache, pagefile, TempDB	Stateless VM OS
Data survives all VM lifecycle events	No	No

Do not confuse:

Temporary disk ≠ Ephemeral OS disk

18. When should you use an ephemeral OS disk?

Consider it when:

The VM is stateless.
The OS can be reprovisioned.
Application state is stored elsewhere.
Fast reimage is valuable.
Local storage performance is beneficial.

Avoid using it when the VM requires the OS disk itself to retain important state across lifecycle operations that can reprovision the OS.

19. What is Premium SSD?

Premium SSD is a high-performance managed disk option designed for workloads requiring lower latency and higher IOPS/throughput than Standard SSD or Standard HDD.

Typical workloads include:

Production application servers
Databases
Transactional applications
Enterprise workloads

The exact performance depends on disk size, VM limits and supported performance features.

20. What is Standard SSD?

Standard SSD provides SSD-based storage at a lower cost than Premium SSD.

It is suitable for workloads with moderate performance requirements where Premium SSD is not necessary.

Examples:

General application servers
Development/test environments
Moderate I/O workloads
File/application workloads with moderate performance requirements

21. What is Standard HDD?

Standard HDD is a lower-cost disk option based on HDD technology.

It is generally suitable for workloads where:

Cost is more important than performance.
I/O requirements are relatively low.
High latency is acceptable.

For new production workloads requiring better performance, SSD-based options should be evaluated.

22. What is Premium SSD v2?

Premium SSD v2 is a newer premium managed disk offering designed for high-performance workloads.

One of its important characteristics is that performance can be configured independently of capacity within supported limits.

This can be useful when an application needs high IOPS or throughput without simply increasing disk capacity.

23. What is Ultra Disk?

Ultra Disk is designed for extremely demanding I/O workloads.

It is appropriate for workloads such as:

High-performance databases
Transaction-heavy applications
High IOPS workloads
High-throughput workloads

Ultra Disk allows independent configuration of performance characteristics within supported limits.

Always verify that both the VM SKU and target region support Ultra Disk.

24. What are IOPS and throughput?

IOPS

IOPS means:

Input/Output Operations Per Second

It measures the number of I/O operations a storage system can process per second.

Throughput

Throughput measures the amount of data transferred per second.

Usually expressed as:

MB/s
GB/s

Example

A database may require:

High IOPS
High throughput
Low latency

Increasing disk capacity alone does not necessarily solve every performance problem.

25. What limits Azure VM disk performance?

Disk performance can be constrained by multiple layers.

For example:

Application
    ↓
Operating System
    ↓
Disk
    ↓
VM disk limits
    ↓
Azure storage

You need to consider both:

Disk limits
VM limits

A very high-performance disk attached to a VM that cannot provide the required aggregate I/O may still be bottlenecked by the VM.

26. What is disk caching in Azure?

Azure VM disk caching uses host-side caching technology to improve storage performance for supported disk types and VM configurations.

Common cache modes include:

None
ReadOnly
ReadWrite

For supported configurations, the default cache setting is commonly:

OS disk → ReadWrite
Data disk → ReadOnly

However, caching should be selected based on the workload rather than blindly accepting defaults.

27. When would you use ReadOnly caching?

ReadOnly caching can be useful when the workload benefits significantly from repeated reads.

For example:

Database/application
      ↓
Frequently read data
      ↓
ReadOnly host cache

The cache can reduce read latency for cached data.

However, workload behavior must be evaluated before changing disk cache settings.

28. When would you use None caching?

None can be appropriate for workloads where host caching is undesirable, including certain write-heavy or write-sensitive workloads.

For example:

Heavy write workload
       ↓
Data disk
       ↓
Host caching = None

The correct setting depends on the application and workload.

29. Can disk caching be changed on a running Azure VM?

Changing disk caching can cause the disk to be detached and reattached.

For an OS disk, changing the cache setting can restart the VM.

Therefore, treat cache changes as an operational change and plan accordingly.

30. What is the Azure VM Agent?

The Azure VM Agent is a lightweight process running inside the guest operating system that enables interaction between Azure and the VM.

One of its major responsibilities is enabling VM extensions.

Without a functioning VM Agent, VM extensions cannot operate correctly.

31. What are Azure VM extensions?

VM extensions are small components that perform post-deployment configuration or management tasks inside a VM.

Examples include extensions for:

Running scripts
Monitoring
Configuration management
Security
Password reset
Custom software installation

Example

You deploy 100 Windows VMs and need to install a monitoring agent.

Instead of manually logging into each VM, you can use a VM extension or another supported automation mechanism.

32. What happens if the Azure VM Agent is not ready?

Several management functions may fail.

For example:

VM extensions may fail.
Run Command may not work.
Guest-level management operations may fail.

The first step is to check the VM Agent status.

For Windows, verify the Azure VM Agent services inside the guest and also check the Azure portal’s VM agent status.

33. What is Run Command in Azure?

Azure Run Command allows you to execute scripts inside a VM through Azure management infrastructure.

For Windows VMs, it can execute PowerShell scripts.

It is particularly useful when:

RDP is unavailable.
You need to make a quick configuration change.
A firewall rule is blocking remote access.
A service must be restarted.
You need to collect diagnostic information.

Run Command relies on the VM Agent.

34. What is Boot Diagnostics?

Boot Diagnostics helps troubleshoot VM startup and boot problems.

It can provide information such as:

Console output
Screenshot of the VM console

This is useful when the VM does not successfully reach the normal operating-system login screen.

35. A VM is running but RDP is unavailable. What would you do?

Do not immediately restart the VM.

Follow a structured process:

Step 1 – Check Azure VM status

Verify:

VM is running.
No platform health issue is reported.

Step 2 – Check Boot Diagnostics

Look for:

Boot errors
Login screen
OS-level errors

Step 3 – Check networking

Verify:

NIC
NSG
Route
Public/private connectivity
RDP port
VPN/Bastion path

Step 4 – Use Run Command

If the VM Agent is healthy, use Run Command to check:

Get-Service TermService
Get-NetTCPConnection -LocalPort 3389

Step 5 – Check Windows Firewall

Verify that RDP is allowed.

Step 6 – Check Event Viewer

Investigate:

System
TerminalServices
Remote Desktop Services

36. What is Azure Serial Console?

Azure Serial Console provides a text-based console connection to supported Azure VMs.

It can be useful when normal network access such as RDP or SSH is unavailable.

For example:

VM running
   ↓
RDP unavailable
   ↓
SSH unavailable
   ↓
Serial Console
   ↓
Guest-level troubleshooting

It is particularly useful for certain boot, network and OS-level troubleshooting scenarios.

37. What is the difference between Restart, Stop and Deallocate?

This is a very common interview question.

Restart

The VM reboots.

The compute allocation remains.

Stop

The VM is stopped, but compute resources may remain allocated depending on the operation/state.

Deallocate

Azure releases the VM’s compute allocation.

For normal managed VMs, deallocation generally stops compute billing, although attached resources such as disks and certain networking resources can continue to incur charges.

Interview summary

Restart     → Reboot VM
Stop        → Stop VM
Deallocate  → Stop + release compute allocation

38. Why is Deallocate important for cost management?

Consider a development VM:

Monday-Friday → Required
Saturday-Sunday → Not required

If the VM remains allocated while not being used, compute charges can continue.

Deallocating it when appropriate can reduce compute costs.

However, disks, snapshots, public IPs and other associated resources can still incur charges.

39. What is Azure VM Redeploy?

Redeploy moves the VM to another Azure host within the Azure infrastructure.

It can be useful when troubleshooting suspected underlying host issues.

A redeploy can help when:

VM host appears unhealthy.
VM is experiencing persistent platform-related problems.
Azure support recommends it.
Guest/network behavior suggests an underlying host issue.

Important:

Data on the temporary disk can be lost during redeployment.

40. What happens when you resize an Azure VM?

Resizing changes the VM’s compute SKU.

For example:

Standard_D2s_v5
       ↓
Standard_D4s_v5

This can increase:

vCPU
Memory
Network capacity
Disk attachment limits

depending on the target SKU.

The target size must be available and compatible with the VM’s configuration and region.

Some resize operations require the VM to restart or be deallocated.

41. Why might a desired VM size not appear in Azure?

Possible reasons include:

SKU unavailable in the region.
SKU unavailable in the current availability zone.
VM generation incompatibility.
Disk compatibility limitations.
Capacity constraints.
VM size is not supported for the current configuration.
Subscription/SKU restrictions.

A senior administrator should verify the available SKUs rather than assuming the portal is malfunctioning.

42. How do you check available VM sizes using Azure CLI?

az vm list-sizes -l eastus -o table

For a real deployment, replace eastus with the target region.

43. How do you check VM sizes using Azure PowerShell?

Get-AzVMSize -Location "EastUS"

You can use this to review available VM sizes in a region.

44. How do you check the current Azure VM size?

Azure CLI:

az vm show \
  --resource-group <ResourceGroupName> \
  --name <VMName> \
  --query hardwareProfile.vmSize \
  -o tsv

PowerShell:

Get-AzVM `
  -ResourceGroupName "<ResourceGroupName>" `
  -Name "<VMName>" |
Select-Object -ExpandProperty HardwareProfile

45. How do you resize an Azure VM using Azure CLI?

Example:

az vm resize \
  --resource-group <ResourceGroupName> \
  --name <VMName> \
  --size Standard_D4s_v5

Before changing production infrastructure, verify:

SKU availability
Application compatibility
Disk compatibility
Maintenance impact
Cost

46. What is Accelerated Networking?

Accelerated Networking uses Azure-supported hardware acceleration capabilities to provide higher network performance and lower latency by reducing part of the networking overhead in the host/guest path.

It is supported only on compatible VM sizes and operating systems.

For demanding workloads, it can improve network performance significantly.

Do not assume every VM size supports it.

47. What is an Availability Set?

An Availability Set is a logical grouping of Azure VMs designed to reduce the probability that related VMs fail simultaneously.

Azure distributes VMs across:

Fault domains
Update domains

Availability Sets are still relevant, particularly for architectures and regions where Availability Zones are unavailable or where their characteristics fit the workload.

48. What is a fault domain?

A fault domain represents a group of infrastructure that shares a common potential point of failure.

Examples can include shared:

Power
Network
Physical infrastructure

By distributing VMs across fault domains, Azure reduces the probability that a single infrastructure failure affects every VM.

49. What is an update domain?

An update domain groups VMs that can be restarted together during planned platform maintenance.

The purpose is to prevent all VMs in an availability set from being restarted simultaneously.

Important current terminology:

Update domains are still relevant to Availability Sets, but Flexible VM Scale Sets use fault-domain-based maintenance behavior rather than traditional update domains.

50. What is an Availability Zone?

An Availability Zone is a physically separate location within an Azure region.

Supported Azure regions have multiple zones with independent infrastructure such as:

Power
Cooling
Networking

Deploying application instances across multiple zones can protect against a failure affecting one zone.

51. What is the difference between Availability Set and Availability Zone?

Feature	Availability Set	Availability Zone
Scope	Fault/update domains	Physically separate zones
Primary protection	Host/infrastructure failures	Zone/datacenter-level failures
Physical separation	Lower	Higher
Network latency	Generally lower between VMs	Generally higher than same-host/fault-domain placement
Zone-level failure protection	No	Yes
Typical modern design	Useful where zones aren’t available	Preferred when zone resilience is required

Availability Zones provide stronger isolation than Availability Sets, while Availability Sets can still be useful where zones are unavailable or where their lower VM-to-VM latency is valuable.

52. What is an Azure Virtual Machine Scale Set?

Azure Virtual Machine Scale Sets allow you to manage a group of VM instances as a scalable application platform.

They provide capabilities such as:

Multiple VM instances
Centralized management
Autoscaling
Load balancing
Availability across fault domains
Availability Zone deployment
Automated VM configuration and updates

53. What is Flexible orchestration mode in VM Scale Sets?

Flexible orchestration provides a more unified VM management model and is the currently recommended Scale Set orchestration mode for many scenarios.

It allows Scale Sets to manage standard Azure VM resources while supporting:

High availability
Fault-domain spreading
Availability Zones
Autoscaling
Different VM configurations
Stateful and stateless workloads

Current Microsoft guidance recommends Flexible orchestration when deploying Scale Sets.

54. What is Uniform orchestration in VM Scale Sets?

Uniform orchestration is designed around a common VM profile where instances are substantially identical.

It is useful for workloads where you need:

Same VM configuration
       +
Large number of instances
       +
Centralized scale management

Flexible and Uniform orchestration have different management and capability characteristics, so the correct choice depends on the workload.

55. What is the difference between Availability Set and VM Scale Set?

Availability Set

Used to distribute individual VMs across fault and update domains.

VM Scale Set

Used to manage a group of VM instances with capabilities such as:

Scaling
Centralized management
Automated deployment
Health management
Load balancing

Modern Azure architectures should evaluate Flexible orchestration Scale Sets when both high availability and scalability are required.

56. How does VM Scale Set autoscaling work?

Autoscaling changes the number of VM instances based on defined conditions.

Example:

CPU > 70%
     ↓
Scale out
     ↓
2 VMs → 4 VMs

When demand decreases:

CPU < 30%
     ↓
Scale in
     ↓
4 VMs → 2 VMs

Autoscaling should be based on meaningful workload metrics rather than simply increasing VM count without understanding the bottleneck.

57. What is zone-spanning VM Scale Set deployment?

A zone-spanning Scale Set distributes VM instances across multiple Availability Zones.

Example:

Azure Region
│
├── Zone 1 → VM 1
│           VM 2
│
├── Zone 2 → VM 3
│           VM 4
│
└── Zone 3 → VM 5
            VM 6

If one zone experiences a failure, instances in other zones can continue serving the application, assuming the application is designed for this architecture.

58. How would you design a highly available web application using Azure VMs?

A basic architecture could be:

                Internet
                   |
            Azure Load Balancer
                   |
        +----------+----------+
        |                     |
     Zone 1                Zone 2
        |                     |
      VM 1                  VM 2
        |                     |
        +----------+----------+
                   |
             Application Data

For higher resilience, the design should also consider:

Multiple Availability Zones
Managed disks
Load balancing
Autoscaling
Application health checks
Backup
Disaster recovery
Database redundancy

The application itself must also be designed so that one VM or zone failure does not destroy service availability.

59. A production VM has very high CPU. How do you troubleshoot it?

Do not immediately resize the VM.

First determine what is consuming CPU.

Windows

Check:

Get-Counter '\Processor(_Total)\% Processor Time'

Then investigate:

Task Manager
Resource Monitor
Application logs
Windows Event Viewer
Azure Monitor

Linux

Use:

top

or:

htop

Then determine whether the issue is:

Application process
Malware/security scan
Database workload
Scheduled job
Backup
OS process
Traffic spike

Only after identifying the bottleneck should you decide whether to optimize or resize.

60. A VM has high memory usage. What would you check?

Check:

Which process consumes memory?
Is memory usage increasing continuously?
Is there a memory leak?
Is the application designed for the current memory size?
Is paging occurring?
Is the VM size appropriate?

Windows

Get-Process |
Sort-Object WorkingSet -Descending |
Select-Object -First 10 Name,Id,WorkingSet

Linux

free -h
top

Do not assume high memory utilization is automatically a problem. Some applications intentionally use available memory for caching.

61. A VM has poor disk performance. What would you check?

Investigate the complete storage path:

Application
    ↓
OS
    ↓
Disk
    ↓
VM SKU
    ↓
Azure Managed Disk

Check:

Disk IOPS
Disk throughput
Latency
Disk type
Disk size
VM disk limits
Disk caching
Application I/O pattern
Read/write ratio
Azure Monitor metrics

A common mistake is upgrading the disk without checking whether the VM itself is the bottleneck.

62. A database VM has high disk latency. What is your approach?

Check:

Step 1

Determine whether the workload is read-heavy or write-heavy.

Step 2

Check disk IOPS and throughput.

Step 3

Check VM-level storage limits.

Step 4

Review disk caching.

Step 5

Check whether the disk type is appropriate.

Step 6

Review database configuration.

Step 7

Check Azure Monitor metrics.

Possible remediation:

Standard HDD
    ↓
Standard SSD
    ↓
Premium SSD

Or, for very demanding workloads:

Premium SSD v2 / Ultra Disk

The final choice should be based on measured requirements.

63. A VM’s disk is full. What would you do?

For Windows:

Get-Volume

For Linux:

df -h

Then identify the consuming files/directories.

For example:

C:\Logs
C:\Windows\Temp
Application logs
Database files

or Linux:

/var/log
/tmp
/var/lib
/application

After confirming the data is safe to modify:

Clean unnecessary files.
Rotate logs.
Move application data.
Expand the disk if necessary.
Extend the partition/filesystem inside the guest OS.

Remember:

Increasing Azure disk capacity does not automatically mean the guest OS filesystem is already expanded.

64. Can you increase the size of an Azure managed disk?

Yes, supported managed disks can generally be expanded.

However, after increasing the Azure disk size, you may still need to extend:

Windows partition/volume
Linux partition
Linux filesystem

Example:

Azure disk:
256 GiB → 512 GiB

Guest OS:
Partition/filesystem may still show 256 GiB

The guest OS must be checked after the Azure-side resize.

65. A VM extension is stuck in provisioning. What would you check?

Check:

VM Agent status.
Extension status.
Extension configuration.
Guest OS compatibility.
Network connectivity to required Azure endpoints.
Extension logs.
Whether another extension operation is currently running.
Whether the extension configuration is valid.

Remember:

VM Agent
    ↓
VM Extension
    ↓
Extension operation

If the VM Agent is unhealthy, extensions can also fail.

66. A VM Agent is not ready. What would you investigate?

Inside the VM, check:

Azure VM Agent service/process
Guest OS health
Disk space
System resources
Network connectivity
Agent logs
Recent OS changes

Also check the Azure portal for:

VM Agent status
VM health
Extension status

Do not repeatedly reinstall extensions without first confirming that the underlying VM Agent is healthy.

67. A VM does not boot. What is your troubleshooting process?

Use a layered approach.

Step 1 – Azure platform

Check:

VM status
Azure Resource Health
Activity Log

Step 2 – Boot Diagnostics

Check:

Screenshot
Console output

Step 3 – Identify failure stage

For example:

Azure infrastructure
       ↓
Bootloader
       ↓
Operating System
       ↓
Services
       ↓
Login

Step 4

Determine whether the issue is caused by:

OS corruption
Boot configuration
Driver
Disk problem
Recent patch
Security software
Filesystem problem

Step 5

Use appropriate recovery methods.

68. A Windows VM shows a blue screen after a recent change. What would you do?

First identify the change:

Driver
Windows Update
Security software
Application
Registry
Startup configuration

Use:

Boot Diagnostics
Azure Serial Console where applicable
Recovery options
Azure support documentation
Windows crash/event information

If the problem started immediately after a known change, investigate that change before randomly modifying unrelated components.

69. A Linux VM cannot be reached over SSH. What should you check?

Follow the same layered methodology.

Azure side

Check:

VM running state
NIC
NSG
Route
Public/private connectivity
Azure platform health

Guest OS

Use available management methods such as Run Command or Serial Console where supported.

Check:

systemctl status ssh

or on distributions using the service name:

systemctl status sshd

Then check:

ss -lntp

Verify that SSH is listening on the expected port.

70. How would you troubleshoot a VM that suddenly became unavailable?

Use this sequence:

1. Azure Resource Health
        ↓
2. VM status
        ↓
3. Activity Log
        ↓
4. Boot Diagnostics
        ↓
5. Network path
        ↓
6. Guest OS
        ↓
7. Application

Do not start by rebooting the VM blindly.

A reboot may remove valuable diagnostic evidence.

71. How do you check the current VM status using Azure CLI?

az vm get-instance-view \
  --resource-group <ResourceGroupName> \
  --name <VMName> \
  -o table

You can also use:

az vm show \
  --resource-group <ResourceGroupName> \
  --name <VMName> \
  --show-details

72. How do you list Azure VMs?

az vm list -o table

For a specific resource group:

az vm list \
  --resource-group <ResourceGroupName> \
  -o table

73. How do you check Azure VMs using PowerShell?

Get-AzVM

For status:

Get-AzVM -Status

For a specific VM:

Get-AzVM `
  -ResourceGroupName "<ResourceGroupName>" `
  -Name "<VMName>" `
  -Status

74. How do you list managed disks using Azure PowerShell?

Get-AzDisk

For a particular resource group:

Get-AzDisk -ResourceGroupName "<ResourceGroupName>"

75. How do you check VM extensions?

Azure CLI:

az vm extension list \
  --resource-group <ResourceGroupName> \
  --vm-name <VMName> \
  -o table

PowerShell:

Get-AzVMExtension `
  -ResourceGroupName "<ResourceGroupName>" `
  -VMName "<VMName>"

76. What is Azure Backup for VMs?

Azure Backup provides backup capabilities for supported Azure VMs.

It can protect:

OS disks
Data disks
VM configuration/state required for recovery

A common architecture is:

Azure VM
   ↓
Recovery Services Vault
   ↓
Recovery Points

Backup should be designed according to:

Recovery Point Objective (RPO)
Recovery Time Objective (RTO)
Retention requirements
Application consistency requirements
Disaster recovery requirements

77. Is an Azure VM snapshot the same as Azure Backup?

No.

A snapshot is a point-in-time copy of a managed disk.

Azure Backup is a broader backup service designed for backup and recovery operations with policy and retention capabilities.

Interview answer

Snapshot → point-in-time disk copy

Azure Backup → managed backup/recovery service

Do not treat a snapshot as a complete enterprise backup strategy.

78. What is the difference between backup and disaster recovery?

Backup

Designed primarily to recover data or systems from a previous recovery point.

Disaster Recovery

Designed to maintain or restore service after a major failure.

Example:

Backup:
Recover yesterday's VM

DR:
Recover application after region-level disaster

Azure Site Recovery is commonly used for VM disaster recovery scenarios.

79. How would you design a highly available two-tier Azure application?

Example:

                  Internet
                     |
              Azure Load Balancer
                     |
          +----------+----------+
          |                     |
       Zone 1                Zone 2
          |                     |
       Web VM                 Web VM
          |                     |
          +----------+----------+
                     |
              Application Tier
                     |
             Database Tier

The database layer requires its own HA strategy.

Simply placing two web VMs in different zones does not automatically make the entire application highly available.

80. What is your approach to an Azure VM production incident?

Use a structured incident process:

1. Determine scope

Is it:

One VM?
Multiple VMs?
One application?
One region?
Multiple services?

2. Check Azure health

Review:

Resource Health
Service Health
Activity Log

3. Check VM state

Determine:

Running
Stopped
Deallocated
Failed
Unresponsive

4. Check boot and guest OS

Use:

Boot Diagnostics
Serial Console where appropriate
Run Command
Guest logs

5. Check networking

Review:

NIC
NSG
Routes
Connectivity
Load balancer health

6. Check compute and storage

Review:

CPU
Memory
Disk latency
IOPS
Throughput
VM limits

7. Mitigate

Depending on the issue:

Restart service
Correct configuration
Resize VM
Change disk configuration
Redeploy VM
Fail over application
Restore from backup

8. Document

Record:

Root cause
Timeline
Actions
Impact
Resolution
Preventive actions

This is the difference between simply administering Azure and performing senior-level infrastructure operations.

Azure VM Commands – Quick Reference

Azure CLI

List VMs

az vm list -o table

Show VM

az vm show \ --resource-group <ResourceGroupName> \ --name <VMName>

Show VM instance status

az vm get-instance-view \ --resource-group <ResourceGroupName> \ --name <VMName>

Start VM

az vm start \ --resource-group <ResourceGroupName> \ --name <VMName>

Restart VM

az vm restart \ --resource-group <ResourceGroupName> \ --name <VMName>

Stop VM

az vm stop \ --resource-group <ResourceGroupName> \ --name <VMName>

Deallocate VM

az vm deallocate \ --resource-group <ResourceGroupName> \ --name <VMName>

Resize VM

az vm resize \ --resource-group <ResourceGroupName> \ --name <VMName> \ --size Standard_D4s_v5

List VM sizes

az vm list-sizes \
  --location eastus \
  -o table

List VM extensions

az vm extension list \
  --resource-group <ResourceGroupName> \
  --vm-name <VMName> \
  -o table

Azure PowerShell – Quick Reference

List VMs

Get-AzVM

Get VM status

Get-AzVM -Status

Get specific VM

Get-AzVM ` -ResourceGroupName "<ResourceGroupName>" ` -Name "<VMName>"

Get VM sizes

Get-AzVMSize -Location "EastUS"

List managed disks

Get-AzDisk

List VM extensions

Get-AzVMExtension ` -ResourceGroupName "<ResourceGroupName>" ` -VMName "<VMName>"

Real-World Azure VM Scenarios

Scenario 1 – Production VM is not reachable

Check:

Resource Health
      ↓
VM status
      ↓
Boot Diagnostics
      ↓
Network
      ↓
Guest OS

Do not immediately resize or restart.

Scenario 2 – RDP fails but VM is running

Use:

Azure network checks
Boot Diagnostics
Run Command
Windows service checks
Firewall checks
Serial Console where applicable

Scenario 3 – VM CPU reaches 100%

Identify the process first.

Do not immediately increase the VM size.

Scenario 4 – Database has high disk latency

Check:

Disk type
IOPS
Throughput
Latency
Caching
VM storage limits
Database workload

Then select the appropriate remediation.

Scenario 5 – VM needs protection from a datacenter-level failure

Evaluate:

Availability Zones

rather than relying only on an Availability Set.

Scenario 6 – Application requires automatic scale-out

Evaluate:

Virtual Machine Scale Sets
+
Autoscaling
+
Load balancing

Scenario 7 – VM Agent is not ready

Investigate:

Guest OS
   ↓
Azure VM Agent
   ↓
Network connectivity
   ↓
VM Extensions

Do not troubleshoot the extension alone if the underlying Agent is unhealthy.

Scenario 8 – VM has a huge temporary disk

Do not assume it is persistent.

Remember:

Temporary disk = temporary

Important application data must be stored on appropriate persistent storage.

Scenario 9 – VM needs very high storage performance

Compare:

Premium SSD
Premium SSD v2
Ultra Disk

Then verify:

IOPS
Throughput
Latency
VM limits
Region availability
Application requirements

Scenario 10 – VM OS can be recreated easily

Consider:

Ephemeral OS Disk

if the workload is stateless and the VM SKU/configuration supports it.

Quick Revision

Remember these Azure VM concepts:

VM Size
→ CPU + Memory + VM capabilities

OS Disk
→ Operating system

Data Disk
→ Persistent application/data storage

Temporary Disk
→ Temporary local storage

Ephemeral OS Disk
→ Stateless/local OS storage

Managed Disk
→ Azure-managed block storage

Premium SSD
→ High-performance general production storage

Premium SSD v2
→ High-performance configurable performance

Ultra Disk
→ Extremely demanding I/O workloads

Availability Set
→ Fault + update domain based availability

Availability Zone
→ Physically separated zone-level resilience

VM Scale Set
→ Group of scalable VMs

Flexible Orchestration
→ Modern recommended Scale Set orchestration

VM Agent
→ Enables guest management and extensions

VM Extension
→ Post-deployment guest configuration

Boot Diagnostics
→ Boot troubleshooting

Run Command
→ Remote guest script execution

Deallocate
→ Release compute allocation

Redeploy
→ Move VM to another Azure host

Exam Answer Summary

1. What is Azure VM?

An Azure VM is an IaaS compute resource that provides a virtualized Windows or Linux server.

2. What is a managed disk?

An Azure-managed block storage volume used by Azure VMs.

3. What are the current managed disk types?

Ultra Disk
Premium SSD v2
Premium SSD
Standard SSD
Standard HDD

4. What is the temporary disk?

Non-persistent local storage intended for temporary workloads.

5. What is an ephemeral OS disk?

An OS disk stored on local VM storage, designed primarily for stateless workloads and fast reprovisioning.

6. Availability Set vs Availability Zone?

Availability Sets distribute VMs across fault and update domains, while Availability Zones provide physically separate zone-level isolation within a region.

What is VM Scale Set?

A service for managing groups of VMs with scaling, centralized management and availability capabilities.

What is VM Agent?

The guest agent that enables Azure VM extensions and guest management functionality.

What is Boot Diagnostics?

A troubleshooting capability providing VM console information such as screenshots and console output.

What is Run Command?

A mechanism for executing scripts inside supported Azure VMs through Azure management infrastructure.

Senior Interview Tip

When an interviewer asks:

“How would you troubleshoot a production Azure VM issue?”

Do not answer:

“I will restart the VM.”

A senior answer should be structured:

First determine the scope and impact.

Then check:
Azure Service Health
Resource Health
Activity Log
VM state
Boot Diagnostics
VM Agent
Networking
CPU
Memory
Disk IOPS
Disk latency
Guest OS
Application

Then identify the root cause.

Apply the least disruptive remediation.

Finally document the incident and preventive action.

That answer demonstrates structured infrastructure troubleshooting, rather than simply knowing Azure portal options.

Progress

Completed:

Part 1 – Azure Networking Fundamentals, VNets, Subnets, NSGs & Connectivity
Part 2 – Azure Virtual Machines, Compute, Disks & Availability

The next section should move into Azure Storage, Azure Files, Blob Storage, Storage Accounts, Redundancy, SAS, Access Keys, Private Endpoints and Storage Troubleshooting without repeating the VM storage concepts already covered here.

Next Part

Continue the Microsoft Azure Interview Series

Part 3 – Azure Storage Interview Questions: Storage Accounts, Blob Storage, Azure Files, Redundancy, SAS & Troubleshooting

For official Microsoft 365 documentation and additional technical information, visit: Microsoft Learn