Skip to main content
CertRampLearning

Microsoft · Free practice questions

AZ-104 Practice Questions

15 original AZ-104 practice questions across the five skill areas of the Microsoft study guide (skills measured as of 17 April 2026), weighted roughly like the real exam. Each one comes with the answer, a short explanation and the reason every other option is wrong.

Written for the skills measured as of 17 April 2026 · Updated October 27, 2026 · Übungsfragen auf Deutsch

These are original practice questions written by CertRamp to the official Microsoft exam outline. They are not real exam questions and do not come from any exam. Try each question first, then open the answer to see the explanation for every option.

Domain 1

Manage Azure identities and governance

Manage access to Azure resourcesIntermediateDomain: Manage Azure identities and governance

Question 1. A developer must be able to create and manage all resources in the resource group RG-Dev, but must not be able to grant access to other users. The access must not apply to any other resource group. What should you do?

  1. Option A: Assign the Owner role at the subscription scope
  2. Option B: Assign the Contributor role at the RG-Dev resource group scope
  3. Option C: Assign the Reader role at the RG-Dev resource group scope
  4. Option D: Assign the User Access Administrator role at the RG-Dev scope
Show answer and explanation

Correct answer: B. Assign the Contributor role at the RG-Dev resource group scope

Pick the least-privileged built-in role and the narrowest scope that meets the need. Role assignments are inherited downwards, so a subscription-level assignment would reach every resource group.

  • A (incorrect)Owner can grant access to others, and the subscription scope covers every resource group.
  • B (correct)Correct. Contributor can manage all resources but cannot assign roles, and the resource group scope limits it to RG-Dev.
  • C (incorrect)Reader can only view resources; the developer could not create or change anything.
  • D (incorrect)This role manages access assignments only, which is exactly what the developer must not do.

Reference: Microsoft Learn: AZ-104 study guide (opens in a new tab)

Manage Azure subscriptions and governanceIntermediateDomain: Manage Azure identities and governance

Question 2. Your organisation requires that new resources can only be created in the West Europe and North Europe regions. Requests for other regions must be blocked. What should you use?

  1. Option A: A ReadOnly resource lock on the subscription
  2. Option B: A tag named Region on every resource group
  3. Option C: An Azure Policy assignment that uses the allowed locations definition
  4. Option D: A budget with an alert at 100%
Show answer and explanation

Correct answer: C. An Azure Policy assignment that uses the allowed locations definition

Azure Policy evaluates resource properties such as location, SKU or tags and can deny non-compliant requests. Resource locks protect existing resources from deletion or modification but cannot apply conditions.

  • A (incorrect)A ReadOnly lock blocks changes to all resources regardless of region; it cannot allow some regions and deny others.
  • B (incorrect)Tags are metadata; on their own they do not block deployments.
  • C (correct)Correct. Azure Policy can deny deployments whose location is not in an approved list.
  • D (incorrect)Budgets track spending and raise alerts; they do not control where resources are deployed.

Reference: Microsoft Learn: AZ-104 study guide (opens in a new tab)

Manage Azure subscriptions and governanceFoundationDomain: Manage Azure identities and governance

Question 3. A production resource group contains critical resources. Administrators must still be able to change settings on these resources, but nobody should be able to delete them by accident. What should you configure?

  1. Option A: A CanNotDelete (Delete) lock on the resource group
  2. Option B: A ReadOnly lock on the resource group
  3. Option C: The Reader role for all administrators
  4. Option D: A tag named DoNotDelete
Show answer and explanation

Correct answer: A. A CanNotDelete (Delete) lock on the resource group

Azure has two lock levels: CanNotDelete (shown as Delete in the portal) and ReadOnly. Locks apply to all users, including Owners, until the lock is removed.

  • A (correct)Correct. A delete lock allows reading and modifying resources but prevents deleting them.
  • B (incorrect)A ReadOnly lock would also block the required configuration changes.
  • C (incorrect)Reader would stop administrators from changing settings at all.
  • D (incorrect)Tags do not enforce anything by themselves.

Reference: Microsoft Learn: AZ-104 study guide (opens in a new tab)

Manage Microsoft Entra users and groupsFoundationDomain: Manage Azure identities and governance

Question 4. The help desk spends a lot of time resetting forgotten passwords. You want a pilot group of users to be able to reset their own passwords after verifying their identity. What should you configure in Microsoft Entra ID?

  1. Option A: Azure Policy
  2. Option B: A resource lock on the directory
  3. Option C: Guest user access
  4. Option D: Self-service password reset (SSPR) scoped to the pilot group
Show answer and explanation

Correct answer: D. Self-service password reset (SSPR) scoped to the pilot group

SSPR can be enabled for none, selected or all users. Starting with a selected group is a common way to pilot it before rolling it out to everyone.

  • A (incorrect)Azure Policy governs Azure resources, not user password resets.
  • B (incorrect)Resource locks apply to Azure resources and have nothing to do with passwords.
  • C (incorrect)Guest access concerns external users, not password resets for employees.
  • D (correct)Correct. SSPR can be enabled for a selected group and requires users to verify their identity with the configured methods.

Reference: Microsoft Learn: AZ-104 study guide (opens in a new tab)

Domain 2

Implement and manage storage

Configure access to storageAdvancedDomain: Implement and manage storage

Question 5. You issue several service SAS tokens for a blob container to a partner. You need the ability to revoke all of them at once later without regenerating the storage account keys. What should you do when creating the tokens?

  1. Option A: Create them as account SAS tokens
  2. Option B: Associate them with a stored access policy on the container
  3. Option C: Set a very long expiry time
  4. Option D: Enable blob versioning
Show answer and explanation

Correct answer: B. Associate them with a stored access policy on the container

A stored access policy defines start, expiry and permissions on the container side. Service SAS tokens that reference it can be revoked centrally, which is not possible for ad hoc tokens without key rotation.

  • A (incorrect)An account SAS cannot be linked to a stored access policy; revoking it requires rotating the key that signed it.
  • B (correct)Correct. Deleting the stored access policy, renaming it, or setting its expiry time in the past revokes every service SAS linked to it.
  • C (incorrect)A long expiry makes the problem worse and does not provide revocation.
  • D (incorrect)Versioning keeps previous versions of blobs; it does not control SAS validity.

Reference: Microsoft Learn: AZ-104 study guide (opens in a new tab)

Configure and manage storage accountsIntermediateDomain: Implement and manage storage

Question 6. A storage account must stay available if a single datacentre in the primary region fails, and data must also be copied to a secondary region in case the whole primary region is lost. Which redundancy option meets both requirements?

  1. Option A: Locally redundant storage (LRS)
  2. Option B: Zone-redundant storage (ZRS)
  3. Option C: Geo-zone-redundant storage (GZRS)
  4. Option D: Geo-redundant storage (GRS)
Show answer and explanation

Correct answer: C. Geo-zone-redundant storage (GZRS)

Remember the two dimensions: zones within the primary region (ZRS) and a copy in a secondary region (GRS). GZRS combines both; RA-GZRS adds read access to the secondary.

  • A (incorrect)LRS keeps copies within one datacentre only, so it protects against neither a datacentre nor a regional outage.
  • B (incorrect)ZRS spreads data across availability zones in one region but does not copy it to a secondary region.
  • C (correct)Correct. GZRS replicates across availability zones in the primary region and asynchronously to a secondary region.
  • D (incorrect)GRS copies to a secondary region, but within the primary region it uses LRS, so a single datacentre failure is not covered by zones.

Reference: Microsoft Learn: AZ-104 study guide (opens in a new tab)

Configure Azure Files and Azure Blob StorageIntermediateDomain: Implement and manage storage

Question 7. Log files in a blob container are read often in their first month, rarely afterwards, and must be deleted after one year. You want this handled automatically with minimal cost. What should you configure?

  1. Option A: A blob lifecycle management policy that moves blobs to cooler tiers by age and deletes them after 365 days
  2. Option B: Soft delete for blobs with a 365-day retention period
  3. Option C: Object replication to a second storage account
  4. Option D: Blob versioning
Show answer and explanation

Correct answer: A. A blob lifecycle management policy that moves blobs to cooler tiers by age and deletes them after 365 days

Lifecycle management policies use rules based on conditions such as days since modification to move blobs between access tiers or delete them. This is the standard way to cut storage costs for ageing data.

  • A (correct)Correct. Lifecycle management rules can tier and delete blobs automatically based on age.
  • B (incorrect)Soft delete only keeps deleted blobs recoverable; it neither tiers nor deletes data automatically.
  • C (incorrect)Object replication copies blobs to another account, which adds cost rather than reducing it.
  • D (incorrect)Versioning keeps previous versions of blobs and does not tier or expire the current ones.

Reference: Microsoft Learn: AZ-104 study guide (opens in a new tab)

Domain 3

Deploy and manage Azure compute resources

Create and configure virtual machinesIntermediateDomain: Deploy and manage Azure compute resources

Question 8. Two virtual machines run the same web application. The solution must keep running if an entire datacentre in the region becomes unavailable. How should you deploy the VMs?

  1. Option A: In the same availability set
  2. Option B: In different resource groups
  3. Option C: On the same dedicated host
  4. Option D: In different availability zones
Show answer and explanation

Correct answer: D. In different availability zones

Availability sets protect against rack-level hardware failures and maintenance; availability zones protect against the loss of a whole datacentre. Pick the option that matches the failure the scenario describes.

  • A (incorrect)Availability sets spread VMs across fault and update domains within one datacentre, so a full datacentre outage would affect both.
  • B (incorrect)Resource groups are a management boundary, not a physical one.
  • C (incorrect)A single host is a single point of failure.
  • D (correct)Correct. Availability zones are physically separate locations within a region, so one zone can fail while the other keeps running.

Reference: Microsoft Learn: AZ-104 study guide (opens in a new tab)

Automate deployment of resources by using Azure Resource Manager (ARM) templates or Bicep filesIntermediateDomain: Deploy and manage Azure compute resources

Question 9. Your team has an existing ARM template in JSON and wants to maintain it as a Bicep file from now on. What is the most efficient way to start?

  1. Option A: Rewrite the template by hand in PowerShell
  2. Option B: Decompile the ARM template into a Bicep file with the Bicep tooling, then review the result
  3. Option C: Rename the .json file to .bicep
  4. Option D: Deploy the template and then delete it
Show answer and explanation

Correct answer: B. Decompile the ARM template into a Bicep file with the Bicep tooling, then review the result

The outline includes exporting a deployment as an ARM template and converting an ARM template to a Bicep file. Decompiling gives a starting point, but you should still check names, parameters and any warnings.

  • A (incorrect)PowerShell scripts are not Bicep, and rewriting by hand is slow and error-prone.
  • B (correct)Correct. The Bicep CLI can decompile JSON templates into Bicep; the output should be reviewed and tidied.
  • C (incorrect)Bicep has its own syntax; renaming the file does not convert it.
  • D (incorrect)Deploying does not produce a Bicep file to maintain.

Reference: Microsoft Learn: AZ-104 study guide (opens in a new tab)

Create and configure Azure App ServiceIntermediateDomain: Deploy and manage Azure compute resources

Question 10. You need to test a new version of an App Service web app in Azure and then release it to production with no downtime, with the option to switch back quickly if problems appear. What should you use?

  1. Option A: A second App Service plan in another region
  2. Option B: Scale out the App Service plan to more instances
  3. Option C: A staging deployment slot and a slot swap
  4. Option D: An App Service backup
Show answer and explanation

Correct answer: C. A staging deployment slot and a slot swap

Deployment slots are live apps with their own host names. Swapping exchanges them without downtime, and swapping again is a fast rollback. Slots require a pricing tier that supports them.

  • A (incorrect)This creates a separate deployment but provides no built-in swap or quick rollback.
  • B (incorrect)Scaling adds capacity; it does not separate test and production versions.
  • C (correct)Correct. You deploy to a staging slot, warm it up, swap it with production and can swap back if needed.
  • D (incorrect)Backups support restoring an app, not zero-downtime releases.

Reference: Microsoft Learn: AZ-104 study guide (opens in a new tab)

Domain 4

Implement and manage virtual networking

Configure and manage virtual networks in AzureIntermediateDomain: Implement and manage virtual networking

Question 11. VNet-A (10.1.0.0/16) is peered with VNet-B, and VNet-B is peered with VNet-C. There is no network virtual appliance or gateway transit. Virtual machines in VNet-A cannot reach VNet-C. What is the simplest fix?

  1. Option A: Create a peering between VNet-A and VNet-C
  2. Option B: Add a second address space to VNet-B
  3. Option C: Create a public IP address for each VM in VNet-C
  4. Option D: Add a DNS zone for VNet-C
Show answer and explanation

Correct answer: A. Create a peering between VNet-A and VNet-C

Peering connects exactly two virtual networks. To route through a hub you need a network virtual appliance or gateway with user-defined routes; otherwise peer the networks directly.

  • A (correct)Correct. Virtual network peering is not transitive, so VNet-A and VNet-C need their own peering.
  • B (incorrect)Changing VNet-B's address space does not make peering transitive.
  • C (incorrect)This exposes VMs to the internet instead of fixing private connectivity.
  • D (incorrect)Name resolution does not create a network path.

Reference: Microsoft Learn: AZ-104 study guide (opens in a new tab)

Configure secure access to virtual networksAdvancedDomain: Implement and manage virtual networking

Question 12. An Azure SQL database must be reachable from your virtual network and from on-premises through a VPN, using a private IP address, and public network access must be disabled. What should you configure?

  1. Option A: A service endpoint for Microsoft.Sql on the subnet
  2. Option B: An NSG rule that allows port 1433
  3. Option C: Azure Bastion
  4. Option D: A private endpoint for the database in the virtual network
Show answer and explanation

Correct answer: D. A private endpoint for the database in the virtual network

Private endpoints bring the PaaS service into your address space, so public access can be turned off. Plan DNS so that clients resolve the service name to the private IP address.

  • A (incorrect)Service endpoints keep traffic on the Azure backbone, but the service still uses its public endpoint and there is no private IP address for on-premises clients.
  • B (incorrect)An NSG filters traffic but does not give the database a private address.
  • C (incorrect)Bastion provides RDP and SSH access to VMs, not private access to PaaS services.
  • D (correct)Correct. A private endpoint gives the service a private IP address in your VNet, reachable from peered networks and on-premises.

Reference: Microsoft Learn: AZ-104 study guide (opens in a new tab)

Configure secure access to virtual networksAdvancedDomain: Implement and manage virtual networking

Question 13. An NSG on a subnet has an inbound rule allowing TCP 443 from any source with priority 200, and an inbound rule denying all traffic from the internet with priority 100. Will HTTPS traffic from the internet reach the VMs in the subnet?

  1. Option A: Yes, because allow rules always take precedence over deny rules
  2. Option B: Yes, because port 443 is always open in Azure
  3. Option C: No, because the deny rule has a lower number and is processed first
  4. Option D: No, because NSGs cannot filter traffic on subnets
Show answer and explanation

Correct answer: C. No, because the deny rule has a lower number and is processed first

Evaluating effective security rules means following priority order: the lowest number is checked first and the first matching rule wins. To let HTTPS through, give the allow rule a lower number than the deny rule (for example, allow 443 at 100 and move the deny rule to 110).

  • A (incorrect)NSGs do not favour allow rules; they process rules by priority.
  • B (incorrect)No port is open by default for inbound internet traffic unless a rule allows it.
  • C (correct)Correct. Lower numbers mean higher priority. Processing stops at the first match, so the deny rule at 100 blocks the traffic.
  • D (incorrect)NSGs can be associated with subnets and network interfaces.

Reference: Microsoft Learn: AZ-104 study guide (opens in a new tab)

Domain 5

Monitor and maintain Azure resources

Monitor resources in AzureFoundationDomain: Monitor and maintain Azure resources

Question 14. When CPU usage on a production VM stays above 90% for 15 minutes, the operations team must receive an email and an SMS. What do you configure in Azure Monitor?

  1. Option A: A diagnostic setting that sends metrics to a storage account
  2. Option B: A metric alert rule with an action group that contains email and SMS notifications
  3. Option C: A resource lock on the VM
  4. Option D: An Azure Advisor recommendation
Show answer and explanation

Correct answer: B. A metric alert rule with an action group that contains email and SMS notifications

Alert rules, action groups and alert processing rules work together: the rule detects, the action group notifies or acts, and processing rules can suppress or add actions, for example during maintenance.

  • A (incorrect)Diagnostic settings route data; they do not notify anyone.
  • B (correct)Correct. The alert rule defines the condition; the action group defines who is notified and how.
  • C (incorrect)Locks protect resources from changes and have nothing to do with alerts.
  • D (incorrect)Advisor gives best-practice recommendations; it does not alert on live metrics.

Reference: Microsoft Learn: AZ-104 study guide (opens in a new tab)

Implement backup and recoveryIntermediateDomain: Monitor and maintain Azure resources

Question 15. Business-critical Azure VMs must be replicated continuously to a secondary region so that you can fail over within minutes if the primary region becomes unavailable. Which service should you configure?

  1. Option A: Azure Backup with a daily backup policy
  2. Option B: Blob soft delete
  3. Option C: Azure Monitor VM insights
  4. Option D: Azure Site Recovery
Show answer and explanation

Correct answer: D. Azure Site Recovery

Azure Backup is for point-in-time restore; Site Recovery is for disaster recovery with replication and failover. Many solutions use both.

  • A (incorrect)Daily backups allow restore, but not continuous replication and fast regional failover.
  • B (incorrect)Soft delete protects blobs against accidental deletion, not VMs against regional outages.
  • C (incorrect)VM insights monitor performance and dependencies; they do not replicate VMs.
  • D (correct)Correct. Site Recovery replicates Azure VMs to another region and orchestrates failover and failback.

Reference: Microsoft Learn: AZ-104 study guide (opens in a new tab)

Want more practice?

15 questions are a sample. The exam is much longer.

The full CertRamp AZ-104 practice exams have 6 tests that grow with you, from a diagnostic start to an exam-level Test 4 and a challenge set beyond the exam, with an explanation for every option.

Who writes the exams

Written by a practitioner, not a question mill

Joshua Ravnjak, founder of CertRamp Learning

Joshua Ravnjak

Senior Consultant, digital transformation and AI

Founder, CertRamp Learning

921 learners and 61 reviews on Udemy (as of November 3, 2026)

I have more than 12 years of professional experience and work as a senior consultant in an IT management consultancy, focusing on digital transformation, AI and requirements engineering in the insurance industry. I have earned many certifications myself, so I know what exam preparation feels like. That is why every CertRamp practice exam follows the same progression: an easy start to see where you stand, then real exam level, then beyond it. You walk into the exam with a buffer and a good feeling.

Selected certifications

  • IREB® CPRE Foundation Level
  • ITIL® 4 Foundation
  • Professional Scrum Master I (PSM I)
  • Professional Scrum Product Owner I (PSPO I)
  • Certified SAFe® 6 Scrum Master
  • Certified AI-Empowered SAFe® 6 Product Owner/Product Manager

Plus further certifications from Microsoft and Anthropic (Claude).

ITIL® is a registered trademark of PeopleCert. IREB® is a registered trademark of IREB e.V. SAFe® is a registered trademark of Scaled Agile, Inc. PSM and PSPO are trademarks of Scrum.org.

Also in print

Cover: CPRE Foundation Level Prüfungsvorbereitung 2026 by Joshua Ravnjak

CPRE Foundation Level Prüfungsvorbereitung 2026

6 Übungsprüfungen mit 270 Fragen und ausführlichen Erklärungen

Paperback, German, October 2026

My CPRE Foundation Level exam-prep book (in German) follows the same progressive approach.

View on Amazon (opens in a new tab)