Azure AZ-104 Practice Test

Exam AZ-104: Microsoft Azure Administrator

Azure AZ-104 overview

  • 200 practice questions
  • 32 flashcards
  • 5 study guide topics
  • 40 key terms

What you'll study

  • Manage Azure identities and governance23%
  • Implement and manage storage18%
  • Deploy and manage Azure compute resources23%
  • Implement and manage virtual networking18%
  • Monitor and maintain Azure resources13%

Microsoft Certified: Azure Administrator Associate is an intermediate-level Microsoft certification for administrators who implement, manage, and monitor an organization's Azure environment, including virtual networks, storage, compute, identity, security, and governance. Candidates usually sit within a larger cloud infrastructure team and coordinate with networking, security, database, application development, and DevOps roles, so the credential assumes familiarity with operating systems, networking, servers, and virtualization, plus working experience with PowerShell, Azure CLI, the Azure portal, Azure Resource Manager templates or Bicep files, and Microsoft Entra ID.

The qualifying exam, AZ-104, is proctored, allows 100 minutes, may include interactive components, and requires a score of 700 or greater. It measures five domains: manage Azure identities and governance (20–25%), implement and manage storage (15–20%), deploy and manage Azure compute resources (20–25%), implement and manage virtual networking (15–20%), and monitor and maintain Azure resources (10–15%). It is delivered in English, Chinese (Simplified), Korean, Japanese, French, Spanish, German, Portuguese (Brazil), Chinese (Traditional), and Italian.

Preparation is best built around the free Microsoft Learn study guide and learning paths, hands-on practice in a personal Azure subscription, the exam sandbox to rehearse the interface, and the practice assessment. The certification is renewed annually by passing a free online assessment on Microsoft Learn.

Common mistakes

Assigning an Entra ID directory role such as Global Administrator when the requirement is to let an administrator manage all resources in a subscription.

Directory roles and Azure RBAC roles look interchangeable in the portal search box, so candidates pick the role whose name sounds most powerful instead of the role that scopes to resources.

What to do instead

Map the requirement to Azure RBAC: Owner for full resource control, Contributor for manage-without-grant, User Access Administrator for role assignments only. Reserve Entra roles (Global Administrator, User Administrator) for directory objects like users, groups, and app registrations.

Designing least privilege by assigning roles on every individual subscription while a management group above them already grants the same or broader access.

Role assignments are inherited and additive; candidates mentally model each subscription as a standalone boundary and forget the tree above it.

What to do instead

Walk the scope hierarchy top-down before answering: list assignments at management group, subscription, resource group, and resource levels, remember that the effective permission set is the union of all of them, and note that narrowing an inherited permission requires a deny assignment (Blueprints/managed apps), not an extra role assignment.

Placing data in the Archive access tier when the scenario says the data may need to be read on short notice, and ignoring early-deletion charges.

Archive looks like the cheapest option, and the tier table is memorized as a price ladder rather than as a state machine with latency and minimum-retention constraints.

What to do instead

Before choosing a tier, check the access requirement against rehydration latency (Standard priority can take up to about 15 hours; High priority is faster and costs more) and against minimum retention before early deletion fees apply (30 days Cool, 90 days Cold, 180 days Archive).

Granting access with a SAS built from the storage account key, or with an ad-hoc service SAS, in a scenario that requires the ability to revoke access for one consumer later.

Portal-generated SAS tokens are quick to create, and candidates do not distinguish a SAS signed by a stored access policy from one signed directly by a key.

What to do instead

If revocation is required, create a stored access policy on the container or share, issue a service SAS that references the policy, and revoke by editing or deleting the policy. Rotating the account key is the only revocation path for account-key-signed SAS and it invalidates every token using that key.

Selecting Stop in the portal for a VM and stating that compute charges stop.

The portal shows the VM as Stopped, which sounds like the same thing as deallocated, and the separate Deallocate action is easy to overlook. Guest-level shutdown inside the OS also leaves the VM allocated.

What to do instead

Only the deallocated state releases the compute allocation and stops compute billing; disks and static public IPs still bill. Recommend auto-shutdown configured to deallocate, or Stop-AzVM (which deallocates) rather than shutdown from inside the guest.

Answering with an availability set when the requirement is a 99.99% SLA or protection against a whole datacenter failing, or sizing a single VM instead of a Virtual Machine Scale Set for autoscaling.

Availability sets and availability zones are both described as 'high availability', and candidates latch onto the first HA construct they learned.

What to do instead

Anchor on the SLA and failure domain: a single VM with Premium SSD is 99.9%, two or more VMs in an availability set is 99.95%, and two or more VMs spread across availability zones is 99.99%. Use zones for datacenter-level failure, availability sets only inside a single datacenter, and VMSS when the scenario mentions scaling or hundreds of identical instances.

Designing a hub-and-spoke topology and assuming spoke VNets can reach each other because each spoke is peered to the hub.

Peering is often drawn as a network cable in diagrams, so candidate assume routed transit that the platform does not provide by default.

What to do instead

State explicitly that VNet peering is non-transitive. To get spoke-to-spoke traffic, deploy an NVA or Azure Firewall in the hub and add user-defined routes on the spoke subnets pointing to it, or use Azure Virtual WAN; enable gateway transit and use remote gateways only for on-premises connectivity sharing.

Adding an outbound NSG rule to allow the return traffic for a permitted inbound connection, or applying the NIC NSG rule before the subnet NSG rule when predicting the outcome.

Candidates carry over a stateful-firewall or ACL mental model from other platforms and do not separate evaluation order from directionality.

What to do instead

Treat NSGs as stateful: a permitted inbound flow automatically allows the reply, so no mirror rule is needed. Evaluate inbound traffic as subnet NSG first, then NIC NSG, and outbound as NIC NSG first, then subnet NSG, with the first deny winning within each NSG.

Planning to query the Activity Log for events older than 90 days, or assuming enabling a diagnostic setting shortens or replaces the built-in retention.

The Activity Log is visible in the portal next to Log Analytics, so it reads like a log store with configurable long-term retention.

What to do instead

The Activity Log keeps 90 days natively and its retention is not user-configurable; to keep events longer, create a diagnostic setting that routes the Activity Log categories to a Log Analytics workspace (retention set on the workspace), a storage account for archive, or an event hub for streaming.

Writing a log-search alert rule when the scenario needs near-real-time notification on a platform metric, or creating an alert rule and forgetting that it does nothing until an action group is attached.

Both alert types live under Azure Monitor > Alerts, and the metric-versus-log distinction plus the alert rule/action group split is easy to miss under time pressure.

What to do instead

Choose metric alerts for platform metrics (evaluated as often as every minute, cheap, no query) and log search alerts for conditions that require a KQL query over a workspace (minimum evaluation frequency of 5 minutes). Always attach an action group, and remember action groups are reusable and can be shared across alert rules.

Exam requirements can change. Confirm current details with the exam provider.

Azure AZ-104 practice

200 questions. Each one lists the choices, the correct answer, and an explanation. With JavaScript, this list opens one question at a time. The checker at the top of this page saves progress in this browser when it is available.

  1. Question 1Manage Azure identities and governance

    Which Microsoft Entra built-in role can create and delete users and reset passwords for non-administrator users, while granting the least privilege needed for those tasks?

    1. Helpdesk Administrator
    2. Global Administrator
    3. User Administrator Correct answer.
    4. Privileged Role Administrator

    Explanation

    User Administrator can create and manage users and can reset passwords for non-administrator users, which matches the required tasks with the least privilege. Global Administrator is broader than needed, and Helpdesk Administrator cannot create users.

  2. Question 2Manage Azure identities and governance

    At which scopes can an Azure RBAC role assignment be created?

    1. Only the subscription and resource group levels
    2. Management group, subscription, resource group, or resource Correct answer.
    3. Only the individual resource and subscription levels
    4. Only the tenant root management group and its subscriptions

    Explanation

    Role assignments can be applied at a management group, a subscription, a resource group, or an individual resource, and lower scopes inherit assignments made higher in the hierarchy.

  3. Question 3Manage Azure identities and governance

    Which lock level prevents users from both modifying and deleting a resource?

    1. ReadOnly Correct answer.
    2. NoWrite
    3. CanNotDelete
    4. Delete

    Explanation

    A ReadOnly lock allows read operations only, so it blocks both updates and deletions. CanNotDelete blocks only deletion while still permitting modifications.

  4. Question 4Manage Azure identities and governance

    You apply tags to a resource group. What happens to the existing resources in that group?

    1. The tags are copied to resources only when the resource group has no locks
    2. Tags are not inherited by the resources in the group Correct answer.
    3. The tags are copied only to resources created after the tags were applied
    4. The tags are copied to each resource automatically

    Explanation

    Tags are not inherited from a resource group to the resources inside it. To propagate tags you must tag each resource, for example with an Azure Policy assignment that uses the modify effect and a remediation task.

  5. Question 5Manage Azure identities and governance

    You plan to delete a management group that contains three subscriptions. What must you do first?

    1. Move the subscriptions to another management group Correct answer.
    2. Assign the Owner role on the tenant root group
    3. Convert the subscriptions into resource groups
    4. Remove all resource locks from the subscriptions

    Explanation

    A management group can be deleted only when it contains no child management groups or subscriptions. Moving the subscriptions elsewhere empties the group so it can be removed.

  6. Question 6Manage Azure identities and governance

    Which group membership type lets you add members by using a rule that evaluates user attributes?

    1. Dynamic Device
    2. Microsoft 365
    3. Assigned
    4. Dynamic User Correct answer.

    Explanation

    A dynamic user group uses a membership rule that evaluates user attributes such as department or country, and membership updates automatically as attributes change. Assigned groups require manual membership management.

  7. Question 7Manage Azure identities and governance

    Which Azure construct serves as the billing and administrative boundary for resources?

    1. Subscription Correct answer.
    2. Resource group
    3. Directory
    4. Management group

    Explanation

    A subscription is the unit of billing and the primary administrative boundary for Azure resources. Resource groups organize resources inside a subscription, and management groups organize subscriptions.

  8. Question 8Manage Azure identities and governance

    A partner organization's employee needs to access resources by signing in with their own email address. Which user type should you create?

    1. Guest Correct answer.
    2. Member
    3. Service principal
    4. Managed identity

    Explanation

    Guest users are external identities invited into the directory through business-to-business collaboration and authenticate with their own credentials. Member accounts are for identities owned by your directory.

  9. Question 9Manage Azure identities and governance

    A user must be able to manage virtual machines in resource group RG1 and must not be able to grant access to anyone. You need to assign the least privileged role. What should you assign?

    1. Owner on RG1
    2. User Access Administrator on RG1
    3. Contributor on the subscription
    4. Virtual Machine Contributor on RG1 Correct answer.

    Explanation

    Virtual Machine Contributor grants management of virtual machines without the ability to manage access. Owner and User Access Administrator both include permission to create role assignments, which violates the requirement.

  10. Question 10Manage Azure identities and governance

    Which Azure Policy effect blocks the creation or update of a resource that violates the policy rule?

    1. Audit
    2. DeployIfNotExists
    3. Append
    4. Deny Correct answer.

    Explanation

    The Deny effect causes the resource request to fail when the resource does not meet the policy condition. Audit only records non-compliance, and Append and DeployIfNotExists alter or add to the request rather than rejecting it.

  11. Question 11Manage Azure identities and governance

    Which property of a custom Azure RBAC role definition lists the control-plane operations that the role permits?

    1. NotActions
    2. AssignableScopes
    3. Actions Correct answer.
    4. DataActions

    Explanation

    Actions lists the control-plane operations the role allows, while NotActions subtracts from that list. DataActions applies to data-plane operations, and AssignableScopes controls where the role can be assigned.

  12. Question 12Manage Azure identities and governance

    Which feature provides time-bound activation of privileged directory roles, optionally requiring approval and justification?

    1. Entitlement management with access packages
    2. Privileged Identity Management Correct answer.
    3. Access reviews for directory roles
    4. Conditional Access with authentication strength

    Explanation

    Privileged Identity Management makes users eligible for a role and requires them to activate it for a limited period, optionally with approval, justification, and multi-factor authentication. It reduces standing privileged access.

  13. Question 13Manage Azure identities and governance

    What is the main purpose of an access review in Microsoft Entra ID?

    1. To create accounts automatically for new employees
    2. To encrypt data stored in Azure Storage accounts
    3. To periodically confirm that users still need their access Correct answer.
    4. To require multi-factor authentication at every sign-in

    Explanation

    Access reviews let reviewers attest on a schedule that users or guests still require access, and the results can automatically remove access that is no longer needed. They are a governance control rather than an authentication or provisioning control.

  14. Question 14Manage Azure identities and governance

    You must be alerted when a subscription's monthly spend exceeds 80 percent of a planned amount. What should you create?

    1. A budget with an alert threshold at 80 percent Correct answer.
    2. A metric alert rule on the subscription
    3. A resource lock on the billing resource group
    4. A policy definition with the Audit effect

    Explanation

    Cost Management budgets track actual or forecast cost against a limit and raise alerts when configured thresholds are crossed. Azure Monitor metric alerts and Azure Policy do not evaluate cost data.

  15. Question 15Manage Azure identities and governance

    A move operation for a virtual machine fails because of a CanNotDelete lock. What should you do?

    1. Assign the Contributor role on the target resource group
    2. Delete the lock, move the resource, and then create the lock again Correct answer.
    3. Delete the virtual machine and redeploy it in the target group
    4. Convert the lock to a ReadOnly lock

    Explanation

    A lock on a resource or its dependencies blocks a move, so the lock must be removed before the move and can be recreated afterwards. Changing the lock type does not remove the restriction.

  16. Question 16Manage Azure identities and governance

    Users must be able to reset their on-premises Active Directory password from the Microsoft Entra self-service password reset portal. Which feature must you enable?

    1. Password writeback Correct answer.
    2. Group writeback
    3. Seamless single sign-on
    4. Password hash synchronization

    Explanation

    Password writeback sends password changes made in the cloud back to on-premises Active Directory, which is required for self-service reset of on-premises passwords. Password hash synchronization only flows hashes from on-premises to the cloud.

  17. Question 17Manage Azure identities and governance

    A user is assigned Reader at the subscription and Contributor on resource group RG1. RG2 is another resource group in the same subscription. What can the user do with the resources in RG2?

    1. Assign roles on the resources
    2. View the resources but not change them Correct answer.
    3. Nothing, because the conflicting assignments cancel each other
    4. Modify and delete the resources

    Explanation

    Role assignments are additive and inherited, so the user has Contributor in RG1 and Reader everywhere else in the subscription, including RG2. Reader permits viewing resources but not changing them.

  18. Question 18Manage Azure identities and governance

    You need to apply a set of 20 related policy definitions together and track their compliance as one assignment. What should you use?

    1. A policy initiative assigned to the scope Correct answer.
    2. A custom RBAC role containing 20 actions
    3. 20 individual resource locks
    4. A single policy definition with 20 parameters

    Explanation

    A policy initiative (policy set definition) groups multiple policy definitions so they can be assigned and reported as a single unit. Individual definitions would require separate assignments and separate compliance views.

  19. Question 19Manage Azure identities and governance

    A helpdesk team in the United Kingdom must be able to reset passwords only for UK users. What should you configure?

    1. An administrative unit containing the UK users Correct answer.
    2. A Conditional Access policy scoped to a group of UK users
    3. A dynamic device group filtered by operating system
    4. A custom RBAC role assigned at the subscription

    Explanation

    Administrative units restrict directory role permissions to a defined collection of users or groups, so a Helpdesk Administrator scoped to the unit can act only on its members. Conditional Access controls sign-in conditions rather than administrative scope.

  20. Question 20Manage Azure identities and governance

    You must require multi-factor authentication only for sign-ins that originate outside the corporate network. What should you configure?

    1. Security defaults in Microsoft Entra ID
    2. Self-service password reset with additional authentication methods
    3. A Conditional Access policy with a location condition Correct answer.
    4. Per-user multi-factor authentication

    Explanation

    A Conditional Access policy can combine a location condition with a grant control that requires multi-factor authentication, so only sign-ins from outside the trusted locations are challenged. Security defaults and per-user MFA apply to all sign-ins.

  21. Question 21Manage Azure identities and governance

    A deny assignment exists on a resource group because of a managed application. Which statement describes its effect?

    1. The deny assignment applies only to guest users
    2. Any subscription Owner can delete the deny assignment
    3. Users who hold the Owner role at the subscription bypass the deny assignment
    4. Users are blocked from the denied operations even when an RBAC role grants them Correct answer.

    Explanation

    Deny assignments are evaluated alongside role assignments and always win, so an operation is blocked even if an RBAC role allows it. They are created by services such as managed applications and are not editable by subscription Owners.

  22. Question 22Manage Azure identities and governance

    You create a custom role whose AssignableScopes contains only /subscriptions/Sub1. Where can this role be assigned?

    1. Only on Sub1 and the scopes beneath it Correct answer.
    2. On any subscription in the tenant
    3. Only on the tenant root management group
    4. On Sub1 and any management group above it

    Explanation

    AssignableScopes limits a custom role to the scopes listed, and those scopes include their children. A scope of Sub1 therefore permits assignment on Sub1, its resource groups, and its resources.

  23. Question 23Manage Azure identities and governance

    You need to allow a group to read blobs in a storage account only when each blob carries the tag Project=Contoso. What should you configure?

    1. A policy with the Append effect on the storage account
    2. A role assignment condition on a blob data role Correct answer.
    3. A ReadOnly resource lock on the storage account
    4. A deny assignment on the storage account

    Explanation

    Attribute-based access control conditions on a data-plane role assignment can require a blob index tag to match a value before access is granted. Resource locks and policy effects do not evaluate per-blob tags for authorization decisions.

  24. Question 24Manage Azure identities and governance

    You assign a policy with the modify effect to add a missing tag. After several hours, resources that existed before the assignment still lack the tag. What should you do?

    1. Create and run a remediation task for the policy assignment Correct answer.
    2. Change the policy effect to Audit
    3. Reassign the policy at a higher scope
    4. Restart the affected resources

    Explanation

    The modify effect applies automatically to new or updated resources, but existing resources must be corrected by a remediation task that uses the assignment's managed identity. Reassigning or restarting resources does not trigger the correction.

  25. Question 25Manage Azure identities and governance

    External partners must request access to a security group, have the request approved by their manager, and lose access automatically after 60 days. Which feature should you use?

    1. An access package in entitlement management Correct answer.
    2. A Privileged Identity Management role setting
    3. A dynamic group with a 60-day membership rule
    4. A Conditional Access policy with session controls

    Explanation

    Access packages in entitlement management provide request workflows with approval, defined assignment duration, and expiry, and they support external users through connected organizations. Dynamic groups cannot enforce approval or expiry workflows.

  26. Question 26Manage Azure identities and governance

    A user account was deleted 12 days ago. An administrator needs to restore the account. For how long after deletion does Microsoft Entra ID keep a deleted user account available for restoration?

    1. 90 days after deletion
    2. 7 days after deletion
    3. 30 days after deletion Correct answer.
    4. Deleted accounts can never be restored

    Explanation

    Deleted user objects are moved to a soft-deleted state and kept for 30 days, during which an administrator can restore the account with the same object ID and most of its properties. After the retention period expires, the object is permanently purged and cannot be recovered. Restores are performed from the Deleted users list in the portal.

  27. Question 27Manage Azure identities and governance

    What is the maximum number of tags that can be applied directly to a single Azure resource?

    1. 100 tags
    2. 15 tags
    3. 20 tags
    4. 50 tags Correct answer.

    Explanation

    Azure supports up to 50 tags per resource and per resource group. Attempting to apply more than that causes the deployment or update to fail. Tag names are limited to 512 characters and tag values to 256 characters.

  28. Question 28Manage Azure identities and governance

    How many levels of management groups can you create beneath the tenant root group?

    1. Ten levels
    2. Three levels
    3. Six levels Correct answer.
    4. Four levels

    Explanation

    A management group hierarchy supports up to six levels of management groups, not counting the root management group or the subscription level. Deeper nesting is rejected when the management group is created. Ten levels is a common distractor because it resembles other service quotas.

  29. Question 29Manage Azure identities and governance

    In an Azure Policy definition, which JSON element contains the conditions that are evaluated against a resource?

    1. then
    2. mode
    3. parameters
    4. if Correct answer.

    Explanation

    A policy definition is built from an if block that holds the condition being evaluated and a then block that holds the resulting effect. The mode element controls which resource types the policy evaluates, and parameters supply values at assignment time.

  30. Question 30Manage Azure identities and governance

    Licenses are assigned to a security group, but several group members never receive a license. What should you check first?

    1. That the group is a mail-enabled security group
    2. That the group has at least two owners assigned
    3. That each affected user has a usage location configured Correct answer.
    4. That each affected user holds the License Administrator role

    Explanation

    Group-based licensing requires every user to have a usage location, because license SKUs are restricted by country or region. Users without a usage location fail licensing and are reported as errors on the group. Membership type and group owners have no effect on license delivery.

  31. Question 31Manage Azure identities and governance

    A user is assigned the Global Administrator role in Microsoft Entra ID but reports that no resources are visible in the Azure subscription. What is the cause?

    1. The administrator must activate the role through PIM first
    2. Directory roles do not grant access to Azure resources by default Correct answer.
    3. The subscription belongs to a different Microsoft Entra tenant
    4. Azure role assignments can take up to 24 hours to propagate to the subscription

    Explanation

    Microsoft Entra directory roles and Azure RBAC are separate authorization systems. Global Administrator grants control over directory objects but no access to subscriptions or resource groups. The administrator must use the elevate access option in the tenant properties, which assigns User Access Administrator at the root scope.

  32. Question 32Manage Azure identities and governance

    You plan to move a virtual machine into a resource group in a different subscription. What is a requirement for the move?

    1. The target resource group must be in the same region as the source resource group
    2. The virtual machine must first be exported as an ARM template
    3. Both subscriptions must belong to the same Microsoft Entra tenant Correct answer.
    4. Both subscriptions must use the same billing account

    Explanation

    Moving resources between subscriptions requires both subscriptions to be associated with the same Microsoft Entra tenant, and the target resource provider must already be registered. The move is performed by the resource provider and typically preserves the resource ID of the resource group's new scope rather than redeploying from a template.

  33. Question 33Manage Azure identities and governance

    You need to identify resources that violate a naming convention without preventing their deployment. Which Azure Policy effect should you use?

    1. DeployIfNotExists
    2. Modify
    3. Deny
    4. Audit Correct answer.

    Explanation

    The Audit effect records a non-compliance entry for the resource without blocking or altering the request. Deny blocks the deployment, while Modify and DeployIfNotExists change or create resources during remediation. Audit is the standard choice for reporting-only governance.

  34. Question 34Manage Azure identities and governance

    A CanNotDelete lock is applied at the subscription scope. What is the effect on resources in that subscription?

    1. Only resources created after the lock is applied are affected
    2. Only resources in the subscription's first resource group are affected
    3. All resources in the subscription inherit the lock Correct answer.
    4. Only the subscription object itself is protected

    Explanation

    Management locks apply to the scope where they are created and are inherited by all child scopes, including resource groups and resources. A CanNotDelete lock at subscription scope therefore protects every resource in every resource group. Inherited locks cannot be removed at the child level; the lock must be removed at its own scope.

  35. Question 35Manage Azure identities and governance

    You have added a custom domain name to Microsoft Entra ID. Which DNS record is used to prove that you own the domain?

    1. A PTR record for the domain
    2. A CNAME record for the tenant name
    3. An SRV record named _sipfederationtls
    4. A TXT record containing the verification value Correct answer.

    Explanation

    Microsoft Entra ID verifies domain ownership by having you create a DNS record at the registrar that contains a unique verification value, typically a TXT record. Once the record is detected, the domain is marked as verified and can be used in user principal names.

  36. Question 36Manage Azure identities and governance

    A newly hired technician must register a FIDO2 security key before any password is issued to them. Which feature should you use?

    1. An access package
    2. A self-service password reset policy
    3. A Windows Hello for Business policy
    4. A Temporary Access Pass Correct answer.

    Explanation

    A Temporary Access Pass is a time-limited passcode that lets a user without a password or registered method sign in and register passwordless credentials. It can be issued as a one-time pass and limited to a specific lifetime. Self-service password reset and access packages do not address passwordless enrollment for a brand-new account.

  37. Question 37Manage Azure identities and governance

    You want to see what a new Conditional Access policy would block before you enforce it. Which policy state should you use?

    1. Report-only Correct answer.
    2. On
    3. Enabled for all users only
    4. Off

    Explanation

    Report-only mode evaluates the policy against sign-ins and records the results in the sign-in logs without enforcing the grant controls. This lets administrators measure impact before switching the policy to On. The Off state performs no evaluation at all.

  38. Question 38Manage Azure identities and governance

    A policy assignment that uses the DeployIfNotExists effect does not remediate non-compliant resources. What should you configure on the assignment?

    1. The Audit effect instead of DeployIfNotExists
    2. An exclusion for each non-compliant resource
    3. A CanNotDelete lock on the target resource group
    4. A managed identity with the roles required by the effect Correct answer.

    Explanation

    Remediation-capable effects such as DeployIfNotExists and Modify require the assignment to have a managed identity that holds the permissions needed to create or change resources. Without that identity, the assignment can evaluate compliance but cannot remediate. The identity is created on the assignment and granted a role at the assignment scope.

  39. Question 39Manage Azure identities and governance

    You must prevent guest users from enumerating the directory while still allowing them to sign in and use shared resources. Which setting should you change?

    1. Guest user access restrictions in external collaboration settings Correct answer.
    2. Conditional Access session controls
    3. The default user role permissions for member users
    4. The self-service sign-up setting for external users

    Explanation

    External collaboration settings include a guest user access restriction that controls whether guests can enumerate users, groups, and other directory objects. Setting it to the most restrictive option limits guests to their own profile object while leaving resource access intact. Conditional Access controls sign-in conditions, not directory read permissions.

  40. Question 40Manage Azure identities and governance

    An Azure Policy assignment is created at the scope of a management group. Which resources does the assignment evaluate?

    1. Only resources in the subscription where the assignment was created
    2. Only resources in the management group's parent subscription
    3. All resources in subscriptions under that management group, including nested management groups Correct answer.
    4. Only resources tagged with the management group name

    Explanation

    Policy assignments are inherited down the management group hierarchy, so an assignment at a management group applies to every child management group and every subscription within them. Individual child scopes can be excluded by using notScopes, but by default all resources beneath the scope are evaluated.

  41. Question 41Manage Azure identities and governance

    You delete a resource group that contains 12 virtual machines. What happens to the virtual machines?

    1. They are moved to the default resource group
    2. They are soft-deleted and can be restored for 30 days
    3. They are retained but become unmanaged
    4. They are deleted along with the resource group Correct answer.

    Explanation

    A resource group is a container, and deleting it deletes every resource it contains. The virtual machines, their disks, and network interfaces are removed with the group. Soft delete applies to some individual services and to directory objects, not to resource group deletion as a whole.

  42. Question 42Manage Azure identities and governance

    A resource group is tagged with CostCenter=1234, but a virtual machine created in that group has no tags. Why does the virtual machine lack the tag?

    1. Tag inheritance works only for resources in the same region
    2. The CostCenter tag name is reserved for billing and cannot be set
    3. Tags require the Owner role on each individual resource
    4. Tags are not inherited from a resource group to its resources Correct answer.

    Explanation

    Tags applied to a resource group or subscription are not automatically inherited by the resources inside them. To enforce inheritance you can assign a policy with the modify effect that appends or copies the parent tag onto child resources. Billing reports based on tags therefore depend on tags being present on the resources themselves.

  43. Question 43Manage Azure identities and governance

    An administrator must create 300 user accounts from a spreadsheet. What is the most efficient approach?

    1. Deploy an ARM template that defines each account
    2. Create a dynamic group that generates the accounts
    3. Synchronize the accounts with Microsoft Entra Connect
    4. Use bulk create users with the downloadable CSV template Correct answer.

    Explanation

    The bulk operations feature accepts a CSV file in the documented template format and creates many users in a single operation. It reports per-row results so errors can be corrected and retried. Microsoft Entra Connect is intended for ongoing synchronization from an on-premises directory, not a one-time import.

  44. Question 44Manage Azure identities and governance

    An employee will be away for six months. The account must not be able to sign in, but the mailbox and permissions must be preserved. What should you configure?

    1. Set the account's sign-in status to blocked Correct answer.
    2. Revoke the user's refresh tokens
    3. Delete the user account
    4. Remove all role assignments from the account

    Explanation

    Blocking sign-in prevents the account from authenticating while keeping the object, mailbox, group memberships, and permissions intact. Deleting the account would remove those relationships. Blocking is reversible by changing the sign-in status back to Allowed.

  45. Question 45Manage Azure identities and governance

    Self-service group management is enabled and a security group has designated owners. Who can approve requests to join the group?

    1. Only a User Administrator
    2. Only a Global Administrator
    3. Only the person who originally created the group
    4. The owners of the group Correct answer.

    Explanation

    When self-service group management is enabled, group owners can approve or reject join requests and manage membership without an administrator. Owner approval can also be delegated to members if the group is configured that way. Administrative directory roles are not required for routine membership changes.

  46. Question 46Manage Azure identities and governance

    Which sign-in method continues to authenticate users to cloud applications when the on-premises directory is unavailable?

    1. Federation with Active Directory Federation Services
    2. Certificate-based authentication
    3. Pass-through authentication
    4. Password hash synchronization Correct answer.

    Explanation

    Password hash synchronization stores a hash of the on-premises password hash in Microsoft Entra ID, so sign-in is validated entirely in the cloud. Pass-through authentication and federation both rely on on-premises components, so an outage in the datacenter interrupts sign-in.

  47. Question 47Manage Azure identities and governance

    You must block B2B collaboration with one specific partner tenant while continuing to allow all other external tenants. What should you configure?

    1. Guest user access restrictions
    2. Organization-specific cross-tenant access settings Correct answer.
    3. Conditional Access named locations
    4. An access review for guest users

    Explanation

    Cross-tenant access settings define default inbound and outbound collaboration rules and allow organization-specific overrides for individual tenants. Adding an organization-specific entry for the partner tenant lets you block B2B collaboration with that tenant only, leaving the defaults in place for everyone else.

  48. Question 48Manage Azure identities and governance

    Which statement about Azure resource groups is correct?

    1. Resource groups can be nested only inside a management group
    2. Resource groups cannot be nested inside one another Correct answer.
    3. Resource groups can be nested up to six levels deep
    4. Resource groups can be nested when they share the same region

    Explanation

    A resource group is a flat container and cannot contain another resource group. Organization above the resource level is provided by management groups and subscriptions, and within a resource group by tags. Six levels is the limit for management groups, not resource groups.

  49. Question 49Manage Azure identities and governance

    A company has no Microsoft Entra ID P1 licenses and must require multi-factor authentication for every user. What should the administrator enable?

    1. Self-service password reset
    2. A Conditional Access policy
    3. Security defaults Correct answer.
    4. Privileged Identity Management

    Explanation

    Security defaults are a preconfigured set of baseline protections available to every tenant at no extra cost, including requiring MFA registration and enforcement for all users. Conditional Access requires Microsoft Entra ID P1 or higher, and PIM requires P2. Security defaults are turned off automatically once Conditional Access policies are in use.

  50. Question 50Implement and manage storage

    Which redundancy option keeps three synchronous copies of a storage account's data within a single Azure region?

    1. Locally redundant storage (LRS) Correct answer.
    2. Read-access geo-redundant storage (RA-GRS)
    3. Geo-redundant storage (GRS)
    4. Zone-redundant storage (ZRS)

    Explanation

    LRS replicates each write three times within a single region, across separate fault and update domains. ZRS spreads copies across availability zones, and GRS/RA-GRS replicate to a second region.

  51. Question 51Implement and manage storage

    A workload continuously writes log records to a single blob that is later read sequentially from start to finish. Which blob type should be created?

    1. Page blob
    2. Append blob Correct answer.
    3. Block blob
    4. Azure managed disk

    Explanation

    Append blobs are optimized for append operations, where each new block is written to the end of the blob, which matches log-writing workloads. Block blobs are for discrete block uploads and page blobs are for random read/write patterns.

  52. Question 52Implement and manage storage

    Which redundancy option replicates a storage account's data across three availability zones within one region?

    1. Zone-redundant storage (ZRS) Correct answer.
    2. Geo-redundant storage (GRS)
    3. Locally redundant storage (LRS)
    4. Geo-zone-redundant storage (GZRS)

    Explanation

    ZRS keeps three copies of the data in separate availability zones of a single region, protecting against the loss of an entire zone. GRS and GZRS involve replication to a secondary region.

  53. Question 53Implement and manage storage

    Which protocol do Windows clients use by default to mount an Azure file share?

    1. FTP
    2. iSCSI
    3. SMB 3.0 Correct answer.
    4. NFS 3.0

    Explanation

    Azure Files exposes shares over SMB 3.0, which Windows clients can mount natively with a drive letter. NFS 3.0 is supported only for Linux clients on premium file share accounts.

  54. Question 54Implement and manage storage

    Which tool provides a command-line interface for copying data to and from Azure Blob storage and Azure Files?

    1. Azure Import/Export
    2. Robocopy
    3. AzCopy Correct answer.
    4. Azure Storage Explorer

    Explanation

    AzCopy is a command-line utility for copying data to and from Blob storage and Azure Files, and it supports synchronization and Microsoft Entra ID sign-in. Storage Explorer is a graphical tool.

  55. Question 55Implement and manage storage

    Which blob access tier offers the lowest storage cost while having the longest retrieval latency and an early deletion penalty?

    1. Cold
    2. Archive Correct answer.
    3. Hot
    4. Cool

    Explanation

    The Archive tier has the lowest per-gigabyte storage price, but blobs must be rehydrated before reading and deleting a blob before 180 days incurs an early deletion charge. Hot, Cool, and Cold tiers cost progressively more to store and less to access.

  56. Question 56Implement and manage storage

    Which managed disk type is designed to deliver the lowest latency for latency-sensitive workloads such as SAP HANA?

    1. Premium SSD
    2. Standard SSD
    3. Standard HDD
    4. Ultra disk Correct answer.

    Explanation

    Ultra disks provide the lowest latency and highest IOPS and throughput of the managed disk options, and they can be sized and adjusted independently. Standard and premium SSD disks have higher latency and less flexible performance settings.

  57. Question 57Implement and manage storage

    An administrator needs premium-tier Azure file shares. Which storage account kind must be created?

    1. StorageV2 (general-purpose v2)
    2. BlobStorage
    3. FileStorage Correct answer.
    4. BlockBlobStorage

    Explanation

    Premium file shares require the FileStorage account kind, which is dedicated to Azure Files and supports only premium performance. General-purpose v2 accounts support only standard file shares.

  58. Question 58Implement and manage storage

    An administrator wants blobs to move from the Hot tier to the Cool tier automatically 30 days after they were last modified. What should be configured?

    1. A lifecycle management rule Correct answer.
    2. A stored access policy
    3. Blob versioning
    4. An immutability policy

    Explanation

    Lifecycle management rules apply conditions such as days since last modification to transition blobs between Hot, Cool, Cold, and Archive tiers or to delete them. The other options govern retention or authorization rather than tier changes.

  59. Question 59Implement and manage storage

    You issue service SAS tokens to partners and must be able to revoke them before their expiry without regenerating the storage account key. What should you do?

    1. Associate each SAS with a stored access policy and delete the policy when revocation is needed. Correct answer.
    2. Enable blob soft delete on the container.
    3. Set each SAS start time to the moment it is issued.
    4. Restrict each SAS to HTTPS only.

    Explanation

    A stored access policy holds the start time, expiry, and permissions, and any SAS that references it inherits those constraints. Deleting, renaming, or shortening the policy invalidates linked SAS tokens without rotating the account key.

  60. Question 60Implement and manage storage

    Which SAS type is signed with a user delegation key obtained by using Microsoft Entra ID credentials instead of the storage account key?

    1. User delegation SAS Correct answer.
    2. SAS secured by a stored access policy
    3. Service SAS with an ad hoc policy
    4. Account SAS

    Explanation

    A user delegation SAS is secured with Microsoft Entra ID credentials and a user delegation key from the Blob service, so its lifetime is bounded by the caller's identity and permissions. Account and service SAS tokens are signed with an account key.

  61. Question 61Implement and manage storage

    An administrator initiates a customer-managed failover of a RA-GRS storage account. What is the outcome?

    1. The secondary region becomes the new primary, the account becomes locally redundant, and writes made after the last sync time may be lost. Correct answer.
    2. The account becomes read-only in both regions until the original primary region recovers.
    3. The account keeps its original redundancy, and the old primary becomes writable again once replication catches up.
    4. The secondary region becomes the new primary, and all data is guaranteed to be present with no loss.

    Explanation

    A failover promotes the secondary region and repoints the DNS names, and the account becomes locally redundant in the new primary region. Only data replicated before the reported Last Sync Time is guaranteed to exist in the new primary.

  62. Question 62Implement and manage storage

    Which Azure File Sync feature keeps frequently accessed files cached on the on-premises server while the rest of the namespace remains in Azure?

    1. Cloud tiering Correct answer.
    2. Volume Shadow Copy Service
    3. File share snapshots
    4. DFS Replication

    Explanation

    Cloud tiering leaves a reparse point on the server for tiered files, so the full namespace is visible while cold data occupies space only in the Azure file share. This reduces the local storage footprint without changing how users browse the share.

  63. Question 63Implement and manage storage

    Which statement correctly distinguishes a private endpoint from a service endpoint for Azure Storage?

    1. A private endpoint maps the storage account to a private IP address in a subnet of a virtual network. Correct answer.
    2. A private endpoint routes traffic over the Azure backbone while the storage account keeps its public endpoint.
    3. A service endpoint requires an on-premises VPN gateway, while a private endpoint does not.
    4. A service endpoint assigns a private IP address from the virtual network to the storage account.

    Explanation

    A private endpoint creates a network interface with a private IP address in your subnet, and clients reach the storage account through that address. A service endpoint extends the virtual network identity to the service over the Azure backbone while the storage account keeps its public IP address.

  64. Question 64Implement and manage storage

    What is required to encrypt a storage account at rest with a customer-managed key?

    1. Create a shared access signature scoped to the Key Vault and store it in the account settings.
    2. Register an application in Microsoft Entra ID and assign it the Storage Account Contributor role.
    3. Create a key in Azure Key Vault and grant a managed identity of the storage account access to that key. Correct answer.
    4. Store a storage account access key in Azure Key Vault and enable automatic key rotation.

    Explanation

    The storage account uses a system-assigned or user-assigned managed identity to call Key Vault for wrap and unwrap operations, so that identity needs key permissions on the vault. Storage account keys and app registrations play no part in this encryption flow.

  65. Question 65Implement and manage storage

    Which feature automatically retains the previous content of a blob when the blob is overwritten or deleted?

    1. Blob soft delete
    2. Container soft delete
    3. Blob versioning Correct answer.
    4. Scheduled blob snapshots

    Explanation

    Blob versioning writes a new version each time a blob is modified, so an earlier version can be restored after an accidental overwrite. Soft delete retains only the state of deleted blobs, not overwritten content.

  66. Question 66Implement and manage storage

    Which AzCopy command copies only new or changed files from a local directory to a blob container and also removes destination files that no longer exist in the source?

    1. azcopy copy with the --overwrite=false flag
    2. azcopy list with the --machine-readable flag
    3. azcopy copy with the --recursive flag
    4. azcopy sync with the --delete-destination flag Correct answer.

    Explanation

    azcopy sync compares source and destination and transfers only files that differ, and the --delete-destination option removes destination files absent from the source. The copy command moves files without reconciling deletions.

  67. Question 67Implement and manage storage

    An administrator changes an existing storage account from LRS to GRS. What is the result?

    1. Existing data is immediately readable from the secondary region endpoints once the change is saved.
    2. The account must be deleted and recreated because redundancy is fixed at creation.
    3. Only blobs written after the change are replicated to the secondary region.
    4. Existing data is replicated asynchronously to the secondary region while the account stays available. Correct answer.

    Explanation

    Redundancy can be changed on an existing account, and switching to GRS starts asynchronous replication of existing data to the paired region while the account remains online. Reads from the secondary region are not possible until replication completes, and only RA-GRS or RA-GZRS expose read endpoints.

  68. Question 68Implement and manage storage

    Which two features must be enabled on both the source and destination accounts before object replication of block blobs can be configured?

    1. Static website hosting and a custom domain
    2. Blob versioning and blob change feed Correct answer.
    3. Hierarchical namespace and NFS 3.0
    4. Blob soft delete and container soft delete

    Explanation

    Object replication relies on the change feed to detect new and modified blobs and on versioning to identify which version to copy. Both features must be enabled on the source and destination accounts.

  69. Question 69Implement and manage storage

    Which mechanism prevents two clients from writing to the same blob at the same time?

    1. A lease on the blob Correct answer.
    2. A shared access signature with a short expiry
    3. A snapshot of the blob taken before each write
    4. An immutability policy on the container

    Explanation

    Acquiring a lease gives the holder exclusive write access for the lease duration, and other clients receive an error until the lease is released, broken, or expired. SAS tokens authorize access but do not serialize writes.

  70. Question 70Implement and manage storage

    An application must read a blob that is currently stored in the Archive tier. What must happen before the data can be read?

    1. Copy the blob to a page blob in the same account.
    2. Create a snapshot of the blob, which restores read access.
    3. Read the blob with a ranged GET request, which is supported for archived blobs.
    4. Set the blob's access tier to Hot or Cool and wait for rehydration to complete. Correct answer.

    Explanation

    Archived blobs are offline, and setting the tier to Hot or Cool triggers rehydration, which can take several hours depending on the priority chosen. Only after rehydration completes can the blob be read.

  71. Question 71Implement and manage storage

    An administrator wants data encrypted twice at rest: once at the service level and again at the infrastructure level, using customer-managed keys. Which setting achieves the second layer?

    1. Enable infrastructure encryption when the storage account is created. Correct answer.
    2. Set the minimum TLS version to 1.2 on the storage account.
    3. Enable secure transfer required on the storage account.
    4. Store the customer-managed key in a managed HSM instead of a Key Vault.

    Explanation

    Infrastructure encryption adds a second layer of encryption at the physical storage layer on top of service-level encryption, and it can only be enabled when the account is created. TLS and secure transfer settings protect data in transit rather than adding a second at-rest layer.

  72. Question 72Implement and manage storage

    A regulation requires that blobs in a container cannot be modified or deleted for seven years, and the control must remain in force even if an account owner tries to remove it. What should be configured?

    1. A read-only stored access policy on the container
    2. Blob versioning combined with a lifecycle management rule
    3. Blob soft delete with a seven-year retention period
    4. A time-based retention policy on the container that has been locked Correct answer.

    Explanation

    A time-based retention policy defines how long blobs must remain unmodified, and a locked policy cannot be deleted or shortened, even by an owner with full control. Legal hold is the alternative when the retention period is not yet known.

  73. Question 73Implement and manage storage

    Which configuration is required to create an Azure file share that uses the NFS 3.0 protocol?

    1. A premium file share account with a private endpoint in a virtual network Correct answer.
    2. A general-purpose v2 account joined to a Microsoft Entra Domain Services domain
    3. A general-purpose v2 account with SMB security settings enabled
    4. A BlobStorage account with hierarchical namespace enabled

    Explanation

    NFS 3.0 file shares are supported only on premium file share accounts, and clients must reach the share over a virtual network through a private endpoint or service endpoint. NFS 3.0 shares do not use identity-based authentication.

  74. Question 74Implement and manage storage

    An application must locate blobs by custom attributes such as costCenter and project without enumerating every blob in the container. What should be used?

    1. Blob index tags queried with the Find Blobs by Tags operation Correct answer.
    2. A lifecycle management rule with a tag filter
    3. Blob metadata read while enumerating every blob in the container
    4. Snapshots of each blob retained for 30 days

    Explanation

    Blob index tags are key-value attributes stored with the blob and indexed by the service, so they can be queried with Find Blobs by Tags across a container or account. Metadata is not indexed and can be filtered only after listing blobs.

  75. Question 75Implement and manage storage

    Which of the following is a valid name for a new Azure storage account?

    1. mystorageaccount123 Correct answer.
    2. ab
    3. StorageAccount01
    4. stg-acct-01

    Explanation

    Storage account names must be between 3 and 24 characters long and may contain only lowercase letters and digits, and the name must be globally unique because it forms part of the service endpoint. Uppercase characters, hyphens, and names shorter than three characters are rejected at creation time.

  76. Question 76Implement and manage storage

    Which storage account setting causes requests made over unencrypted HTTP to be rejected?

    1. Hierarchical namespace
    2. Secure transfer required Correct answer.
    3. Allow Blob public access
    4. Minimum TLS version

    Explanation

    The Secure transfer required setting enforces HTTPS for all requests to the account, so plain HTTP calls fail. The minimum TLS version setting only governs which TLS protocol versions are accepted for connections that are already using HTTPS, and the other two settings control anonymous access and a namespace feature.

  77. Question 77Implement and manage storage

    A blob container is set to the Container level of anonymous access, and public access is permitted on the storage account. What does this grant to an unauthenticated client?

    1. Anonymous read of known blobs, without listing the container
    2. Anonymous read of blobs and enumeration of the container's contents Correct answer.
    3. No anonymous access, because every read must carry a SAS token
    4. Anonymous read, write, and delete access to every blob in the container

    Explanation

    The Container access level allows anonymous clients both to read blobs and to list the blobs in that container, whereas the Blob access level allows reads only when the blob name is already known. Anonymous access is never write-capable and only applies when the account-level public access setting is enabled.

  78. Question 78Implement and manage storage

    Which redundancy option keeps three synchronous copies of data in separate availability zones within a single region?

    1. Zone-redundant storage (ZRS) Correct answer.
    2. Read-access geo-redundant storage (RA-GRS)
    3. Locally redundant storage (LRS)
    4. Geo-redundant storage (GRS)

    Explanation

    Zone-redundant storage replicates data synchronously across three availability zones inside one region, so data survives the loss of a single zone. LRS keeps copies within one zone, while GRS and RA-GRS add an asynchronous copy in a distant secondary region.

  79. Question 79Implement and manage storage

    A lifecycle management rule is created with a blob prefix filter of logs/. Which blobs does the rule act on?

    1. Every block blob in a container named logs, regardless of blob name
    2. Only blobs that carry a blob index tag named logs
    3. Block blobs whose names begin with logs/ in any container Correct answer.
    4. All blobs in the account that were last modified more than 30 days ago

    Explanation

    Lifecycle rules are evaluated at the account level and the prefix filter is matched against the blob name, not the container name, so any block blob whose path begins with logs/ is included regardless of which container holds it. Tag-based filtering is a separate condition and is not implied by a name prefix.

  80. Question 80Implement and manage storage

    An administrator needs a scheduled report listing every blob in an account with its size, tier, and creation time for auditing. Which feature provides this?

    1. Storage Analytics logging
    2. Blob inventory Correct answer.
    3. Blob versioning
    4. Change feed

    Explanation

    Blob inventory produces scheduled manifest reports in CSV or Parquet format that list blobs, snapshots, versions, and container metadata, which is exactly suited to periodic auditing. The other features track individual changes, request diagnostics, or retain prior blob versions rather than generating a full inventory listing.

  81. Question 81Implement and manage storage

    An application must process an ordered, durable log of every create, modify, and delete event for block blobs in a storage account. Which feature should be used?

    1. Blob inventory reports
    2. Blob soft delete
    3. Change feed Correct answer.
    4. Storage metrics

    Explanation

    Change feed provides a durable, ordered, read-only log of changes to blobs together with a cursor that consumers can resume from, making it suitable for event-driven processing. Inventory reports are periodic snapshots of state, soft delete retains deleted data, and metrics aggregate performance counters.

  82. Question 82Implement and manage storage

    Which Azure Files capability provides a read-only, point-in-time view of an entire share that can be mounted or browsed to recover an earlier version of a single file?

    1. SMB multichannel
    2. File share snapshot Correct answer.
    3. File share soft delete
    4. Azure File Sync cloud tiering

    Explanation

    A file share snapshot is a read-only copy of a share at a moment in time; it can be mounted or accessed through the previous versions interface so individual files can be restored without affecting the live share. Soft delete only preserves deleted shares, cloud tiering manages caching, and SMB multichannel improves throughput.

  83. Question 83Implement and manage storage

    A data disk attached to a virtual machine hosts a database workload that performs mostly sequential writes. Which host caching option should be configured for that disk?

    1. None Correct answer.
    2. Read/write
    3. Write-through
    4. Read-only

    Explanation

    Host caching of None is recommended for write-heavy or sequential workloads because caching in front of the disk adds no benefit and can add latency for writes. Read-only caching suits read-heavy data, and read/write caching should be reserved for OS disks and avoided for database files.

  84. Question 84Implement and manage storage

    An administrator must transfer 80 TB of data to a storage account over a 10 Mbps internet connection. Which approach is most appropriate?

    1. Create an Azure Import/Export job and ship drives Correct answer.
    2. Map the container as a network drive and copy the files
    3. Run AzCopy with automatic resume over the existing link
    4. Enable Azure File Sync and let cloud tiering upload the data

    Explanation

    Transferring 80 TB over a 10 Mbps link would take months, so shipping encrypted drives to a Microsoft datacenter through an Azure Import/Export job is the practical option. AzCopy, network drive copies, and File Sync all depend on the same insufficient bandwidth.

  85. Question 85Implement and manage storage

    An administrator wants to use point-in-time restore for block blobs on a general-purpose v2 storage account. Which combination of features must be enabled on the account?

    1. Change feed and container soft delete
    2. Blob versioning, change feed, and blob soft delete Correct answer.
    3. Blob soft delete and object replication
    4. Blob versioning and a lifecycle management policy

    Explanation

    Point-in-time restore works by rewinding to a prior state using stored versions and a change log, so blob versioning, change feed, and blob soft delete must all be enabled on the account. Without versioning there is nothing to restore to, and without the change feed the service cannot determine which blobs to revert.

  86. Question 86Implement and manage storage

    Three virtual machines in a cluster must read and write the same managed disk concurrently. Which configuration is required?

    1. Enable Accelerated Networking on all three virtual machines
    2. Set maxShares to 3 on the disk and use a cluster-aware file system Correct answer.
    3. Create a disk access resource and attach the disk through a private endpoint
    4. Convert the disk to an unmanaged VHD stored in a page blob

    Explanation

    Azure shared managed disks support attaching one disk to multiple VMs when maxShares is set to a value greater than one and the disk type supports sharing; the guest operating system must use a cluster-aware file system to coordinate access. Accelerated networking and private endpoints affect connectivity, not disk sharing.

  87. Question 87Implement and manage storage

    A storage account is configured for geo-zone-redundant storage (GZRS). While the primary region is unavailable and no failover has been performed, what read access do clients have to the secondary region?

    1. Read access only for blobs stored in the Cool tier
    2. No read access; secondary reads require RA-GZRS Correct answer.
    3. Read access at any time, because GZRS is always readable
    4. Read access once the account failover completes

    Explanation

    GZRS combines zone redundancy in the primary region with asynchronous replication to one secondary region, but the data in the secondary region is not readable unless read-access geo-zone-redundant storage (RA-GZRS) is used. Failover promotes the secondary region rather than enabling ongoing secondary reads.

  88. Question 88Deploy and manage Azure compute resources

    You deploy an ARM template to a resource group by using Complete deployment mode. A storage account in the resource group is not defined in the template. What happens to the storage account?

    1. It is locked to prevent changes.
    2. It is deleted. Correct answer.
    3. It is moved to another resource group.
    4. It is retained unchanged.

    Explanation

    Complete mode treats the resource group as the declared target state, so resources that exist in the group but are absent from the template are deleted. Incremental mode leaves such resources untouched.

  89. Question 89Deploy and manage Azure compute resources

    What does the Azure CLI command az bicep build produce from a .bicep file?

    1. A compiled .NET assembly
    2. A parameter file with default values
    3. An ARM JSON template Correct answer.
    4. A Terraform configuration file

    Explanation

    Bicep is a declarative domain-specific language that transpiles to an ARM JSON template, which the Azure Resource Manager engine then deploys. The build step performs the transpilation.

  90. Question 90Deploy and manage Azure compute resources

    A virtual machine scale set has an autoscale rule that scales out when average CPU exceeds 70 percent for 10 minutes. What must you also configure to prevent the scale set from growing beyond 20 instances?

    1. A spot instance eviction policy
    2. An availability set for the scale set
    3. A maximum instance count of 20 Correct answer.
    4. A cooldown period of zero minutes

    Explanation

    Autoscale profiles define a minimum, maximum, and default instance count, and the maximum value caps how far any scale-out rule can grow the set. Without a maximum, a rule could keep adding instances while its metric condition holds.

  91. Question 91Deploy and manage Azure compute resources

    An App Service web app running on a Standard plan experiences sustained high CPU. You need to add more worker instances without changing the pricing tier. What should you do?

    1. Enable deployment slots
    2. Scale out to more instances Correct answer.
    3. Change to a higher pricing tier
    4. Create a second App Service plan

    Explanation

    Scaling out increases the number of instances in the existing App Service plan and stays within the current tier. Scaling up instead moves the plan to a larger SKU and may change cost and available features.

  92. Question 92Deploy and manage Azure compute resources

    You need an app setting to remain associated with the staging slot after a swap. What should you do?

    1. Add it to the production slot only
    2. Enable Always On for staging
    3. Store it in Azure Key Vault
    4. Mark it as a slot setting Correct answer.

    Explanation

    App settings and connection strings marked as deployment slot settings are sticky: they stay with the slot instead of traveling with the content during a swap. Unmarked settings move with the swapped content.

  93. Question 93Deploy and manage Azure compute resources

    You need to host App Service apps in an isolated, dedicated environment inside a virtual network. Which App Service feature should you use?

    1. An App Service Environment Correct answer.
    2. A deployment slot on a Standard plan
    3. Azure Container Instances
    4. An App Service plan in the Premium v3 tier

    Explanation

    An App Service Environment is a dedicated, isolated App Service deployment injected into a subnet of a virtual network, giving full network isolation. Multi-tenant plans, even Premium, run on shared App Service infrastructure with only outbound or inbound VNet integration options.

  94. Question 94Deploy and manage Azure compute resources

    You need to run a PowerShell script on a newly provisioned Windows virtual machine without signing in to the VM. Which mechanism should you use?

    1. A runbook in an Automation account
    2. A scheduled task created after sign-in
    3. The Custom Script Extension Correct answer.
    4. Azure Automation State Configuration

    Explanation

    The Custom Script Extension downloads and runs a script on the VM through the guest agent, so no interactive sign-in is required and the script runs as part of provisioning.

  95. Question 95Deploy and manage Azure compute resources

    A database VM requires the lowest possible disk latency and lets you set IOPS and throughput independently of disk size. Which managed disk type should you choose?

    1. Standard HDD
    2. Premium SSD
    3. Ultra Disk Correct answer.
    4. Standard SSD

    Explanation

    Ultra Disks are designed for the lowest latency and allow IOPS and throughput to be configured independently of capacity. Premium SSD offers high performance but with limits tied more closely to disk size.

  96. Question 96Deploy and manage Azure compute resources

    You plan to run a fault-tolerant batch workload on Spot VMs. You need the VM's managed OS disk to be kept when Azure evicts the VM. Which eviction policy should you configure?

    1. Deallocate Correct answer.
    2. Retain for a grace period
    3. Delete
    4. Hibernate

    Explanation

    With the Deallocate eviction policy, an evicted Spot VM is stopped and deallocated, so its managed disks are preserved and the VM can be started again later. The Delete policy removes the VM and its disks when eviction occurs.

  97. Question 97Deploy and manage Azure compute resources

    Administrators must connect to Azure VMs over RDP without public IP addresses on the VMs or open RDP ports. Which Azure service should you deploy?

    1. Azure Firewall
    2. Network Watcher
    3. Azure Bastion Correct answer.
    4. Azure VPN Gateway

    Explanation

    Azure Bastion brokers RDP and SSH sessions over TLS from the Azure portal through the Azure network, so the VMs need no public IP address and no inbound management ports.

  98. Question 98Deploy and manage Azure compute resources

    You need to minimize network latency between several VMs by keeping them on the same physical hardware cluster while still using availability sets. Which feature should you use?

    1. A NIC with accelerated networking
    2. An availability zone
    3. Virtual network peering
    4. A proximity placement group Correct answer.

    Explanation

    A proximity placement group is a placement constraint that keeps resources on the same hardware cluster or datacenter to reduce network latency between them. It is often combined with availability sets for resiliency within that cluster.

  99. Question 99Deploy and manage Azure compute resources

    A VM becomes unreachable after a host maintenance event. Restarting it does not help, and you must move it to a new Azure host without reconfiguring it. What should you do?

    1. Use the Redeploy operation on the VM Correct answer.
    2. Capture a VM image and redeploy it
    3. Resize the VM to a different size and back
    4. Delete and recreate the VM from the same disk

    Explanation

    Redeploy shuts down the VM, moves it to a new Azure host, and powers it back on while keeping the same disks, network interfaces, and configuration. It is the supported remedy for host-level connectivity problems.

  100. Question 100Deploy and manage Azure compute resources

    You run a short-lived batch job in Azure Container Instances and the container must not restart after its process exits. Which restart policy should you configure?

    1. UnlessStopped
    2. Always
    3. OnFailure
    4. Never Correct answer.

    Explanation

    The Never restart policy runs the container group once and does not restart containers when they exit, which suits one-shot jobs. Always and OnFailure both restart containers under their respective conditions.

  101. Question 101Deploy and manage Azure compute resources

    In an Azure Kubernetes Service cluster, which node pool hosts critical cluster pods such as CoreDNS?

    1. The spot node pool
    2. The virtual node pool
    3. The system node pool Correct answer.
    4. The user node pool

    Explanation

    AKS creates system node pools to run critical cluster components such as CoreDNS and metrics-server. User node pools are intended for application workloads and can be scaled or tainted independently.

  102. Question 102Deploy and manage Azure compute resources

    You need to protect virtual machines against the failure of an entire datacenter within a region. Which construct should you use?

    1. A proximity placement group
    2. A scale set with one instance
    3. Availability zones Correct answer.
    4. An availability set

    Explanation

    Availability zones are physically separate datacenters within a region, so distributing VMs across zones protects against the loss of a single datacenter. Availability sets only spread VMs across racks and update groups inside one datacenter.

  103. Question 103Deploy and manage Azure compute resources

    In an ARM template, which element ensures a virtual machine is created after its network interface?

    1. The dependsOn element Correct answer.
    2. The apiVersion property
    3. The condition element
    4. The copy element

    Explanation

    dependsOn declares an explicit ordering so that ARM creates the referenced resource first. ARM also infers some dependencies from functions such as reference, but dependsOn makes the ordering explicit.

  104. Question 104Deploy and manage Azure compute resources

    You need to distribute a customized VM image to multiple subscriptions and regions with versioning and automatic replication. Which resource should you create?

    1. A managed image in each subscription
    2. A snapshot of the OS disk
    3. An Azure Compute Gallery Correct answer.
    4. A VHD stored in a storage account

    Explanation

    Azure Compute Gallery stores image definitions and versions and can replicate them to multiple regions and share them across subscriptions or tenants. Managed images, VHDs, and snapshots are limited to a single region and subscription scope.

  105. Question 105Deploy and manage Azure compute resources

    A scale set with 50 instances runs a custom VM image. You need a new image version applied in batches with a configurable pause between batches. Which upgrade policy should you configure?

    1. Immutable
    2. Automatic
    3. Rolling Correct answer.
    4. Manual

    Explanation

    The Rolling upgrade policy applies updates in batches, with settings such as the batch size percentage and the pause time between batches, which limits downtime. Automatic applies the update without that batch control, and Manual requires you to update instances yourself.

  106. Question 106Deploy and manage Azure compute resources

    You need the OS disk of a stateless VM to use local storage so that it incurs no storage cost and is recreated when the VM is reimaged. Which configuration should you choose?

    1. A Premium SSD managed disk
    2. An OS disk stored in Azure Files
    3. An Ultra Disk
    4. An ephemeral OS disk Correct answer.

    Explanation

    An ephemeral OS disk uses the VM's local temporary storage, so there is no storage charge for the disk and it is recreated on reimage or redeploy. The trade-off is that the OS disk data is not durable.

  107. Question 107Deploy and manage Azure compute resources

    You need encryption performed inside the guest operating system, with keys protected in a Key Vault, for a VM's OS and data disks. Which feature should you use?

    1. Azure Disk Encryption Correct answer.
    2. Double encryption at rest
    3. Storage service encryption on the managed disks
    4. Customer-managed keys on the storage account

    Explanation

    Azure Disk Encryption uses BitLocker on Windows or dm-crypt on Linux inside the guest to encrypt OS and data volumes, with key material held in a Key Vault. Platform features such as storage service encryption protect data at the storage layer instead.

  108. Question 108Deploy and manage Azure compute resources

    You need to move an existing Azure VM to a different Azure region. Which approach is supported?

    1. Change the VM's Region property in the portal
    2. Move it with a resource group move operation
    3. Recreate it from a disk copied to the region Correct answer.
    4. Attach a new public IP in the target region

    Explanation

    A VM's region is fixed at creation, so moving regions means deploying the VM again in the target region from a copied disk, image, or template. Azure Resource Mover orchestrates this process.

  109. Question 109Deploy and manage Azure compute resources

    You need to back up Azure virtual machines by using Azure Backup. Which resource must you create first?

    1. A storage account with blob containers
    2. An Azure Automation account
    3. A Recovery Services vault Correct answer.
    4. A Log Analytics workspace

    Explanation

    Azure VM backup stores backup policies and recovery points in a Recovery Services vault, which must exist in the same region as the VMs before you can enable backup.

  110. Question 110Deploy and manage Azure compute resources

    The Azure Monitor Agent is delivered to an Azure virtual machine as what type of Azure resource?

    1. A network interface
    2. A runbook
    3. A VM extension Correct answer.
    4. A managed disk

    Explanation

    VM extensions are small agent packages published by Microsoft that ARM attaches to a virtual machine to add monitoring, configuration, or scripting capabilities, and the Azure Monitor Agent is installed this way.

  111. Question 111Deploy and manage Azure compute resources

    You need a serverless platform for a containerized microservice that scales to zero, supports revisions and traffic splitting, and requires no Kubernetes management. Which service should you choose?

    1. Azure App Service Web App for Containers
    2. Azure Container Apps Correct answer.
    3. Azure Kubernetes Service
    4. Azure Container Instances

    Explanation

    Azure Container Apps provides a managed, serverless container platform built on Kubernetes concepts with scale to zero, revisions, and traffic splitting, without exposing cluster management.

  112. Question 112Deploy and manage Azure compute resources

    You need to resize a running VM to a size that is not available on its current hardware cluster. What should you do first?

    1. Redeploy the VM
    2. Deallocate the VM Correct answer.
    3. Restart the VM
    4. Capture the VM image

    Explanation

    If the requested size is not supported on the cluster currently hosting the VM, the VM must be deallocated so Azure can release the current hardware and place the VM on a cluster that supports the new size.

  113. Question 113Deploy and manage Azure compute resources

    A company deploys four virtual machines into a single availability set. What does the availability set guarantee about where those VMs are placed?

    1. They are consolidated onto one rack to minimize network latency.
    2. They are automatically replicated to the paired Azure region.
    3. They are spread across fault domains and update domains. Correct answer.
    4. They are guaranteed to run on physical servers reserved for one tenant.

    Explanation

    An availability set distributes VMs across fault domains, which have separate power and network, and across update domains, which are rebooted in separate maintenance batches, protecting the workload from hardware failure and planned platform maintenance.

  114. Question 114Deploy and manage Azure compute resources

    Which resource defines the region, pricing tier, and compute capacity that an App Service app runs on?

    1. The deployment slot
    2. The App Service plan Correct answer.
    3. The App Service environment
    4. The host name binding

    Explanation

    An App Service plan defines the region, the pricing tier, and the number and size of the virtual machine instances, and every app assigned to the plan shares those compute resources.

  115. Question 115Deploy and manage Azure compute resources

    An App Service app runs in a production slot with a staging slot alongside it. What does a slot swap do?

    1. It routes a percentage of production traffic to the staging slot without changing the content.
    2. It deletes the staging slot and recreates it from the production app.
    3. It exchanges the content and settings of the two slots. Correct answer.
    4. It merges the files of both slots into the production slot and empties staging.

    Explanation

    A swap exchanges the content, configuration, and host names of the two slots, and App Service warms up the target slot first so requests do not hit a cold app. Slot settings marked as sticky stay with their slot instead of moving.

  116. Question 116Deploy and manage Azure compute resources

    Which action releases the compute resources of an Azure VM so that you stop being billed for its CPU and memory?

    1. Stopping the VM from inside the guest operating system.
    2. Deallocating the VM. Correct answer.
    3. Restarting the VM from the Azure portal.
    4. Deleting the VM's network interface.

    Explanation

    Only deallocation releases the compute allocation and returns the VM to a state where no CPU or memory charges apply. Disks and static public IP addresses continue to be billed separately.

  117. Question 117Deploy and manage Azure compute resources

    Which managed disk type can be attached only as a data disk and cannot be used as a VM OS disk?

    1. Premium SSD
    2. Standard HDD
    3. Ultra Disk Correct answer.
    4. Standard SSD

    Explanation

    Ultra Disks are supported only as data disks and are chosen when you need to set IOPS and throughput independently of disk size. They cannot host an operating system.

  118. Question 118Deploy and manage Azure compute resources

    Several containers are defined in one Azure Container Instances container group. What do those containers share?

    1. A storage account dedicated to all of their volumes.
    2. A separate restart policy for each container.
    3. A shared IP address and port namespace. Correct answer.
    4. A separate managed identity for each container.

    Explanation

    Containers in an ACI container group are scheduled together on the same host and share a DNS name label, an IP address, and a port namespace, so they can reach each other over localhost.

  119. Question 119Deploy and manage Azure compute resources

    Which Azure Functions hosting option bills per execution and scales to zero instances when there is no traffic?

    1. The Dedicated plan
    2. The Premium plan
    3. The Consumption plan Correct answer.
    4. An App Service Environment

    Explanation

    The Consumption plan allocates compute only while functions execute, scales automatically with load, and scales to zero when idle, with billing based on executions and execution time.

  120. Question 120Deploy and manage Azure compute resources

    You need a virtual machine scale set that can mix VM sizes, include existing VMs, and span availability zones. Which orchestration mode should you choose?

    1. Flexible orchestration mode. Correct answer.
    2. Automatic orchestration mode, which upgrades instances without an upgrade policy.
    3. Uniform orchestration mode, which uses a single VM model created from a template.
    4. Classic orchestration mode, which is intended for cloud services.

    Explanation

    Flexible orchestration mode lets a scale set contain VMs of different sizes and configurations, including existing VMs, and it can span availability zones. Uniform mode instead requires every instance to come from the same single VM model.

  121. Question 121Deploy and manage Azure compute resources

    A scale set must add instances when average CPU across the set exceeds 70 percent for 10 minutes and remove them when it falls below 30 percent. What should you configure?

    1. Instance protection on each running instance.
    2. The automatic OS image upgrade policy.
    3. A metric-based autoscale rule. Correct answer.
    4. A scheduled autoscale profile for business hours.

    Explanation

    Metric-based autoscale rules compare an aggregated metric such as CPU percentage against a threshold over a time window and then run a scale action. Separate scale-out and scale-in rules cover the two thresholds.

  122. Question 122Deploy and manage Azure compute resources

    A regulation requires that a workload run on physical servers that are not shared with any other Azure customer. Which option satisfies the requirement?

    1. Azure Dedicated Host Correct answer.
    2. An availability zone
    3. A proximity placement group
    4. An App Service Environment

    Explanation

    Azure Dedicated Host provisions physical servers that are dedicated to a single subscription, so no other customer's virtual machines are placed on that hardware.

  123. Question 123Deploy and manage Azure compute resources

    A VM that handles a very high packet rate shows high CPU usage caused by the networking stack. Which feature should you enable to offload that processing?

    1. Accelerated networking Correct answer.
    2. Just-in-time VM access
    3. Azure Bastion
    4. Azure Load Balancer

    Explanation

    Accelerated networking uses SR-IOV so the VM talks directly to the physical NIC, bypassing the host virtual switch and reducing CPU load and latency for high-throughput traffic.

  124. Question 124Deploy and manage Azure compute resources

    You plan to back up an App Service app, including its content and configuration, on a schedule. Which resource must exist in the same subscription first?

    1. A private endpoint on the App Service plan.
    2. A Recovery Services vault linked to the app.
    3. An Azure Backup policy applied to the App Service plan.
    4. A storage account in the same subscription. Correct answer.

    Explanation

    App Service backup writes app content and configuration to a storage account and container that you specify, and that storage account must be in the same subscription as the app.

  125. Question 125Deploy and manage Azure compute resources

    Pods in an AKS cluster stay in a Pending state because the node pool has no capacity. Which component adds nodes automatically in response?

    1. The cluster autoscaler Correct answer.
    2. The Horizontal Pod Autoscaler
    3. KEDA
    4. The Vertical Pod Autoscaler

    Explanation

    The cluster autoscaler watches for pods that cannot be scheduled and increases the node count of the node pool, then removes underused nodes later. Pod-level scalers change pods, not nodes.

  126. Question 126Deploy and manage Azure compute resources

    You must deploy an Azure Container Instances container group into an existing virtual network subnet so it can reach on-premises resources over a VPN. What must you configure on the subnet?

    1. Enable accelerated networking on the container group.
    2. Attach a network security group that allows all outbound traffic.
    3. Delegate the subnet to Azure Container Instances. Correct answer.
    4. Assign a public IP address to the container group.

    Explanation

    To place a container group in a virtual network, the subnet must be delegated to Microsoft.ContainerInstance/containerGroups, and the container group is then deployed with a reference to that subnet.

  127. Question 127Deploy and manage Azure compute resources

    You must enforce and automatically correct specific Windows features and registry settings on a group of VMs. Which VM extension should you use?

    1. The DSC extension. Correct answer.
    2. The Custom Script Extension.
    3. The VMAccess extension.
    4. The Azure Monitor Agent extension.

    Explanation

    The Desired State Configuration extension applies a configuration to the guest operating system and keeps the machine in that state, correcting drift on subsequent runs. A script extension runs once and does not enforce state.

  128. Question 128Deploy and manage Azure compute resources

    A VM stops responding during startup. You need to review the boot output to diagnose the failure. Which feature should you enable on the VM?

    1. Boot diagnostics Correct answer.
    2. Connection troubleshoot in Network Watcher
    3. Just-in-time VM access
    4. Azure Monitor VM insights

    Explanation

    Boot diagnostics captures serial console output and screenshots of the VM during startup, which you can review from the portal to see exactly where boot stopped.

  129. Question 129Deploy and manage Azure compute resources

    You must replicate Azure VMs to a secondary region and run a test failover without affecting the production VMs. Which service should you use?

    1. Geo-redundant storage replication
    2. Azure Resource Mover
    3. Azure Backup
    4. Azure Site Recovery Correct answer.

    Explanation

    Azure Site Recovery continuously replicates VM disks to a target region and can run a test failover into an isolated network, followed later by a planned failover and failback.

  130. Question 130Deploy and manage Azure compute resources

    A licensed database must keep high memory and disk throughput but use only 8 vCPUs to limit per-core licensing costs. Which VM size should you choose?

    1. A burstable B-series size.
    2. A constrained vCPU capable size. Correct answer.
    3. A memory-optimized M-series size that uses every vCPU.
    4. A GPU-optimized N-series size.

    Explanation

    Constrained vCPU capable sizes present the memory and storage throughput of a larger size while limiting the number of vCPUs visible to the guest, which reduces per-core licensing cost.

  131. Question 131Deploy and manage Azure compute resources

    You enable automatic instance repairs on a virtual machine scale set. What else must exist for the platform to detect unhealthy instances?

    1. The automatic OS image upgrade policy enabled.
    2. Instance protection applied to each instance.
    3. A health probe on the scale set. Correct answer.
    4. A Log Analytics workspace linked to the scale set.

    Explanation

    Automatic instance repairs depends on a health signal supplied either by a load balancer health probe or by the Application Health extension. Without that signal the platform cannot decide when an instance must be replaced.

  132. Question 132Deploy and manage Azure compute resources

    You need an Azure Resource Manager template to deploy three virtual machines that differ only in name. Which template construct should you use?

    1. The copy element on the resource. Correct answer.
    2. The dependsOn element with three values.
    3. A nested template with three separate resource declarations.
    4. A variable that concatenates the three names.

    Explanation

    The copy element combined with copyIndex creates a loop that instantiates the same resource definition multiple times, which is how a template declares a variable number of near-identical resources.

  133. Question 133Deploy and manage Azure compute resources

    Which option encrypts a VM's temporary disk and disk caches in addition to the OS and data disks, without installing an encryption extension in the guest?

    1. Encryption at host Correct answer.
    2. Confidential disk encryption with customer-managed keys
    3. Azure Disk Encryption
    4. Server-side encryption with platform-managed keys

    Explanation

    Encryption at host encrypts data at the host level, covering the OS disk, data disks, the temporary disk, and disk caches, and it does not use an in-guest extension such as Azure Disk Encryption.

  134. Question 134Deploy and manage Azure compute resources

    An autoscale profile with a scale-out rule of one instance rapidly adds many instances after a single short CPU spike. Which setting should you adjust to stop repeated actions?

    1. The metric time grain of the rule.
    2. The maximum instance count in the profile.
    3. The cool-down period of the autoscale rule. Correct answer.
    4. The default instance count of the scale set.

    Explanation

    The cool-down period is the wait time after a scale action before that rule can act again. Lengthening it prevents a burst of repeated scale-out actions from one spike and lets the metrics of new instances stabilize.

  135. Question 135Deploy and manage Azure compute resources

    An existing VM is a member of an availability set in a region that supports availability zones. The VM must become zone-redundant. What must you do?

    1. Recreate the VM from a snapshot with a zone specified. Correct answer.
    2. Move the VM to another availability set in the same region.
    3. Change the availability zone property of the existing VM in the portal.
    4. Add the VM to a proximity placement group in the target zone.

    Explanation

    A VM's availability zone and availability set assignment are fixed when the VM is created. To place the VM in a zone, delete it and recreate it in the required zone from a snapshot or image.

  136. Question 136Implement and manage virtual networking

    A subnet in an Azure virtual network uses the address range 10.0.1.0/24. How many IP addresses in this range can be assigned to virtual machines and other resources?

    1. 251 Correct answer.
    2. 254
    3. 256
    4. 250

    Explanation

    Azure reserves five addresses in every subnet: the first (network address), the second (default gateway), the third and fourth (Azure DNS mapping), and the last (broadcast address). A /24 contains 256 addresses, so 256 minus 5 leaves 251 assignable addresses.

  137. Question 137Implement and manage virtual networking

    A network security group has two inbound rules for TCP port 443: rule DenyHttps with priority 100 that denies traffic from any source, and rule AllowHttps with priority 200 that allows traffic from any source. An internet client tries to connect to port 443 on a VM in the associated subnet. What is the result?

    1. The connection is blocked only if the deny rule is also associated with the VM's network interface.
    2. The connection succeeds because allow rules always take precedence over deny rules.
    3. The connection is blocked because the rule with the lower priority number is processed first. Correct answer.
    4. The connection succeeds because the default rule AllowVnetInBound applies to the traffic.

    Explanation

    Network security groups process rules in priority order, and the lowest number is evaluated first. Once a rule matches the traffic, evaluation stops, so the priority 100 deny rule wins and the priority 200 allow rule is never reached.

  138. Question 138Implement and manage virtual networking

    Virtual network VNetA is peered with VNetB, and VNetB is peered with VNetC. No other connectivity, gateways, or route tables exist. Can a VM in VNetA reach a VM in VNetC?

    1. No, because a virtual network can be peered with only one other virtual network.
    2. Yes, because peering automatically propagates routes to all peered networks in the chain.
    3. No, because virtual network peering is not transitive. Correct answer.
    4. Yes, provided the three virtual networks use non-overlapping address spaces.

    Explanation

    Virtual network peering is a one-to-one relationship and is not transitive. Traffic from VNetA to VNetC requires a direct peering between those two networks or a routing device such as a firewall or gateway in VNetB with the appropriate peerings and forwarding settings.

  139. Question 139Implement and manage virtual networking

    You need traffic from a subnet to a specific destination prefix to be silently dropped before it leaves the virtual network. Which next hop type should the custom route use?

    1. Virtual network gateway
    2. None Correct answer.
    3. Virtual appliance
    4. VnetLocal

    Explanation

    A user-defined route with the next hop type None drops traffic that matches the address prefix. The Virtual network gateway next hop sends traffic to a VPN or ExpressRoute gateway, Virtual appliance sends it to a specified IP, and VnetLocal keeps it inside the virtual network.

  140. Question 140Implement and manage virtual networking

    You create a private DNS zone named contoso.internal, link it to a virtual network, and enable auto-registration on the link. What is the outcome?

    1. Public DNS records for the virtual machines are created in the contoso.internal zone.
    2. A conditional forwarder is created to the on-premises DNS servers.
    3. The virtual machines resolve names using only the Azure-provided DNS suffix.
    4. DNS records for the virtual machines' private IP addresses are registered in the zone automatically. Correct answer.

    Explanation

    When a virtual network link to a private DNS zone has auto-registration enabled, Azure creates and updates A records for the virtual machines in that virtual network using their private IP addresses. The records are visible only to resources that can resolve against the linked zone.

  141. Question 141Implement and manage virtual networking

    Which statement correctly distinguishes a service endpoint from a private endpoint?

    1. A service endpoint gives the PaaS resource an IP address taken from the subnet's range.
    2. Service endpoints can be used only with virtual machines, not with PaaS services.
    3. Both require a site-to-site VPN connection to function.
    4. A private endpoint gives the PaaS resource an IP address from the subnet, whereas a service endpoint does not. Correct answer.

    Explanation

    A private endpoint creates a network interface in your subnet with a private IP address that represents the PaaS resource. A service endpoint instead extends the virtual network identity to the service over the Azure backbone while the service keeps its public endpoint.

  142. Question 142Implement and manage virtual networking

    Which statement about Standard SKU public IP addresses is correct?

    1. They are provided at no charge when attached to a load balancer.
    2. They are always statically assigned and can be zone-redundant. Correct answer.
    3. They are dynamic by default and can be converted to static.
    4. They can be attached only to classic deployment model resources.

    Explanation

    Standard SKU public IP addresses are always static, are closed to inbound traffic unless an NSG allows it, and can be created as zone-redundant or zonal. The Basic SKU supported dynamic allocation and was not zone-aware.

  143. Question 143Implement and manage virtual networking

    A network security group named NSG-Subnet is associated with a subnet, and a different network security group named NSG-NIC is associated with a virtual machine's network interface in that subnet. How is inbound traffic to the VM evaluated?

    1. The rule with the numerically lowest priority across both groups determines the outcome.
    2. The subnet NSG is evaluated first, then the NIC NSG, and traffic must be allowed by both. Correct answer.
    3. The two network security groups are merged into a single combined rule set.
    4. Only the NSG associated with the network interface is evaluated.

    Explanation

    Inbound traffic is evaluated against the subnet-associated NSG first and then against the NIC-associated NSG, and both must permit the traffic. Outbound traffic is evaluated in the reverse order, starting with the NIC NSG.

  144. Question 144Implement and manage virtual networking

    Which Azure service can make routing decisions based on the URL path of an HTTP or HTTPS request?

    1. Azure Bastion
    2. Traffic Manager
    3. Azure Load Balancer
    4. Application Gateway Correct answer.

    Explanation

    Application Gateway operates at layer 7 and can route based on HTTP attributes such as URL path and host header. Azure Load Balancer works at layer 4, Traffic Manager resolves DNS, and Bastion provides remote session access.

  145. Question 145Implement and manage virtual networking

    You must connect to a virtual machine over RDP through the Azure portal without assigning a public IP address to the VM. Which service should you use?

    1. Azure Firewall
    2. Azure Bastion Correct answer.
    3. Network Watcher
    4. Azure VPN Gateway

    Explanation

    Azure Bastion provides browser-based RDP and SSH sessions to virtual machines through the Azure portal over TLS, so the VMs do not need public IP addresses or open inbound management ports.

  146. Question 146Implement and manage virtual networking

    Which subnet name must be used for the subnet that hosts a VPN gateway in a virtual network?

    1. AzureGateway
    2. GatewaySubnet Correct answer.
    3. Any name, as long as the subnet is at least /27
    4. GatewaySubnet only when ExpressRoute is used

    Explanation

    Azure identifies the gateway subnet by the exact name GatewaySubnet. The subnet should be at least /27 so that it has enough addresses for the gateway instances and future configuration changes.

  147. Question 147Implement and manage virtual networking

    Which statement about Azure ExpressRoute is correct?

    1. It can connect only virtual networks that are in a single Azure region.
    2. It provides a private connection through a connectivity provider that does not traverse the public internet. Correct answer.
    3. It is limited to a maximum circuit bandwidth of 1 Gbps.
    4. It encrypts traffic over the public internet using IPsec.

    Explanation

    ExpressRoute uses a dedicated private connection provisioned by a connectivity provider, so traffic does not go over the public internet. Circuits support a range of bandwidths and can be linked to virtual networks across regions through peering and gateways.

  148. Question 148Implement and manage virtual networking

    In a hub-and-spoke topology, you need all internet-bound traffic from the spoke subnets to be inspected by an Azure Firewall deployed in the hub. What should you configure?

    1. Virtual network peering with Allow forwarded traffic enabled and no other settings.
    2. Network security group rules on the spoke subnets that deny all outbound internet traffic.
    3. A route table associated with each spoke subnet that has 0.0.0.0/0 with next hop type Virtual appliance pointing to the firewall's private IP address. Correct answer.
    4. Azure Firewall application rules that reference FQDN tags scoped to the spoke virtual networks.

    Explanation

    Traffic follows the most specific matching route, and a user-defined default route overrides the system route to the internet. Pointing 0.0.0.0/0 at the firewall's private IP address forces spoke traffic through the appliance, where firewall policy can inspect it.

  149. Question 149Implement and manage virtual networking

    Which statement describes the function of an Azure NAT gateway?

    1. It is configured individually on each virtual machine network interface.
    2. It provides scalable outbound source network address translation for the subnets associated with it. Correct answer.
    3. It accepts inbound connections from the internet to virtual machines behind it.
    4. It requires a subnet named GatewaySubnet in the same virtual network.

    Explanation

    A NAT gateway provides outbound-only connectivity for the subnets it is associated with, translating private addresses to a static public IP or prefix. Because it does not accept unsolicited inbound connections, it is not a substitute for a public load balancer.

  150. Question 150Implement and manage virtual networking

    Azure DDoS Network Protection is enabled at which scope?

    1. Per resource group
    2. Per virtual network Correct answer.
    3. Per virtual machine
    4. Per load balancer

    Explanation

    DDoS Network Protection is applied to a virtual network and protects the public IP resources associated with it, such as those on load balancers and virtual machines. The free platform-level protection is always active but offers no tuning or attack analytics.

  151. Question 151Implement and manage virtual networking

    You need to know whether traffic for a specific five-tuple, including source IP, destination IP, and port, is allowed or denied by a network security group rule. Which Network Watcher tool should you use?

    1. Next hop
    2. Connection monitor
    3. Packet capture
    4. IP flow verify Correct answer.

    Explanation

    IP flow verify evaluates a specific five-tuple against the NSG rules applied to a virtual machine and reports whether the traffic is allowed or denied, along with the name of the rule responsible.

  152. Question 152Implement and manage virtual networking

    What do NSG flow logs record?

    1. An audit trail of changes made to network security group rules.
    2. The full packet contents captured from a virtual machine's network interface.
    3. The route changes that the platform applies to subnets over time.
    4. Allowed and denied flows through a network security group, including source, destination, and port, written to a storage account. Correct answer.

    Explanation

    Flow logs capture information about IP flows that traverse a network security group, including whether each flow was allowed or denied, and write the records to a storage account for analysis or ingestion into Traffic Analytics.

  153. Question 153Implement and manage virtual networking

    You enable the Microsoft.Storage service endpoint on subnet A but not on subnet B. Both subnets host virtual machines that access the same storage account. What is the effect?

    1. Storage traffic from both subnets automatically uses the Azure backbone.
    2. Only traffic from the subnet with the endpoint enabled can be permitted through the storage account's network rules. Correct answer.
    3. The storage account receives a private IP address from subnet A's address range.
    4. The endpoint blocks all storage traffic from subnet A until an NSG rule is added.

    Explanation

    A service endpoint extends the virtual network identity to the service, which lets the storage account firewall allow that specific subnet. Traffic from subnet B has no endpoint identity and is evaluated as public traffic.

  154. Question 154Implement and manage virtual networking

    You delegate a subnet to Microsoft.Web/serverFarms. What does the delegation accomplish?

    1. It reserves the subnet for that service and prevents other resource types from being deployed in it. Correct answer.
    2. It creates a private endpoint for the service inside the subnet.
    3. It assigns a public IP address to the delegated service.
    4. It grants the service permission to read the virtual network's NSG rules.

    Explanation

    Subnet delegation inserts a service association link so that the designated service can create its own resources in the subnet. Because the subnet is reserved for that service, you cannot deploy unrelated resources such as virtual machines into it.

  155. Question 155Implement and manage virtual networking

    You create a private endpoint for an Azure Storage blob account in a virtual network. On-premises clients must resolve the storage account name to the private IP address. What should you configure?

    1. An NSG rule that allows DNS traffic on port 53 from the on-premises network.
    2. A public DNS CNAME record that points the account name to the private IP address.
    3. A privatelink private DNS zone linked to the virtual network, with on-premises DNS forwarding to Azure DNS. Correct answer.
    4. A service endpoint on the on-premises network's address range.

    Explanation

    Private endpoints rely on the privatelink DNS zone, for example privatelink.blob.core.windows.net, to map the service name to the private IP address. On-premises resolvers must forward queries for that zone to Azure DNS or to a DNS forwarder in the virtual network.

  156. Question 156Implement and manage virtual networking

    Which requirement must be met before Accelerated Networking can be enabled on a virtual machine?

    1. The network interface must be associated with a Standard SKU public IP address.
    2. The subnet must have at least one service endpoint enabled.
    3. The VM size must be supported and the guest OS must include the required network drivers. Correct answer.
    4. The virtual network must contain a subnet named GatewaySubnet.

    Explanation

    Accelerated Networking uses SR-IOV, so it requires a supported VM size and a guest operating system with the appropriate drivers, such as the Mellanox driver on supported Linux images. It does not depend on gateways, public IPs, or service endpoints.

  157. Question 157Implement and manage virtual networking

    Two virtual networks in different Azure regions have non-overlapping address spaces and must communicate over the Microsoft backbone without a VPN gateway. What should you configure?

    1. A point-to-site VPN connection
    2. A site-to-site VPN connection
    3. An ExpressRoute circuit
    4. Global virtual network peering Correct answer.

    Explanation

    Global virtual network peering connects virtual networks across regions over the Microsoft backbone with low latency and no gateway requirement. Regional peering works the same way when both networks are in the same region.

  158. Question 158Implement and manage virtual networking

    You specify custom DNS server addresses in a virtual network's DNS settings. What happens for virtual machines in that virtual network?

    1. The custom servers receive queries only for names outside the virtual network.
    2. The Azure-provided DNS servers are queried first, then the custom servers.
    3. Queries from the virtual machines are sent to the custom DNS servers for name resolution. Correct answer.
    4. The custom servers must be reachable through a public IP address.

    Explanation

    When custom DNS servers are defined at the virtual network level, the virtual machines use those addresses for resolution instead of the Azure-provided resolver. The servers must be reachable from the virtual network, typically at a private IP address.

  159. Question 159Implement and manage virtual networking

    You need to review the routes that the Azure platform has applied to a virtual machine's network interface, including system routes and user-defined routes. Which Network Watcher capability should you use?

    1. Effective routes Correct answer.
    2. IP flow verify
    3. Connection monitor
    4. Packet capture

    Explanation

    Effective routes shows the combined set of system routes, user-defined routes, and learned routes that apply to a network interface or subnet. It is the standard way to confirm which route will be selected for a given destination prefix.

  160. Question 160Implement and manage virtual networking

    A spoke virtual network must use the VPN gateway located in the hub virtual network for on-premises connectivity. What must be configured on the peerings?

    1. Create a user-defined route in the spoke with next hop type Virtual network gateway.
    2. Enable Allow forwarded traffic on both peerings and leave the gateway settings unchanged.
    3. Deploy a GatewaySubnet in the spoke virtual network and enable Use remote gateways on both peerings.
    4. Enable Allow gateway transit on the hub-to-spoke peering and Use remote gateways on the spoke-to-hub peering. Correct answer.

    Explanation

    Gateway transit requires the peering from the hub to the spoke to allow gateway transit, and the peering from the spoke to the hub to use remote gateways. The spoke then reaches on-premises through the hub's VPN or ExpressRoute gateway without deploying its own.

  161. Question 161Implement and manage virtual networking

    A subnet is configured with the address range 10.0.1.0/24. How many IP addresses in this subnet can be assigned to network interfaces and other resources?

    1. 250
    2. 254
    3. 256
    4. 251 Correct answer.

    Explanation

    Azure reserves five IP addresses in every subnet: the first address (network), the next three addresses for Azure platform services, and the last address (broadcast). For a /24, 256 minus 5 leaves 251 usable addresses.

  162. Question 162Implement and manage virtual networking

    Which default inbound rule in a network security group permits traffic from resources in the same virtual network?

    1. AllowAzureLoadBalancerInBound
    2. AllowVnetOutBound
    3. AllowVnetInBound Correct answer.
    4. DenyAllInBound

    Explanation

    Every network security group contains default rules, including AllowVnetInBound with priority 65000, which permits traffic between resources in the same virtual network. These default rules cannot be deleted, but a custom rule with a lower priority number can override them.

  163. Question 163Implement and manage virtual networking

    You need traffic destined for 10.20.0.0/16 to be sent to a network virtual appliance that has the IP address 10.0.0.4. Which next hop type should the custom route use?

    1. VnetLocal
    2. Internet
    3. VirtualAppliance Correct answer.
    4. VirtualNetworkGateway

    Explanation

    The VirtualAppliance next hop type forwards matching traffic to a specified private IP address, typically a firewall or other network virtual appliance. The address of the appliance is entered in the next hop address field of the route.

  164. Question 164Implement and manage virtual networking

    You plan to create a virtual network peering between two virtual networks in the same region. Which requirement must be satisfied?

    1. Both virtual networks must use the same DNS servers.
    2. Their address spaces must not overlap. Correct answer.
    3. Both virtual networks must be in the same subscription.
    4. Both virtual networks must be in the same resource group.

    Explanation

    Peering requires that the address spaces of the two virtual networks do not overlap, because each virtual network must be able to route to the other unambiguously. Peering is supported across subscriptions, resource groups, and tenants.

  165. Question 165Implement and manage virtual networking

    You create a private DNS zone and want DNS records for virtual machines to be created and updated automatically when their IP addresses change. What should you configure?

    1. A virtual network link that has autoregistration enabled Correct answer.
    2. A static A record for each virtual machine
    3. A conditional forwarder on the virtual network's DNS servers
    4. An Azure DNS Private Resolver endpoint

    Explanation

    A virtual network link connects a private DNS zone to a virtual network, and enabling autoregistration on that link creates and maintains records for the VMs in the linked network. Autoregistration is a property of the link, not of the zone itself.

  166. Question 166Implement and manage virtual networking

    Dozens of web servers are spread across several subnets, and you must allow inbound TCP 443 to all of them without listing IP addresses in network security group rules. What should you use?

    1. A private endpoint for each web server
    2. A route table with a virtual appliance next hop
    3. An application security group as the rule destination Correct answer.
    4. A service tag for each subnet in the rule

    Explanation

    Application security groups group network interfaces by workload role, and an NSG rule can use the group as its source or destination. Adding a VM to the group automatically applies the rule to it, so no IP addresses need to be maintained.

  167. Question 167Implement and manage virtual networking

    You must distribute TCP 1433 traffic across three virtual machines, and the frontend must be reachable only from inside the virtual network. Which configuration should you use?

    1. A public load balancer with a Basic SKU public IP address
    2. An internal load balancer with a private frontend IP address Correct answer.
    3. An application gateway with a public frontend
    4. A public load balancer with a Standard SKU public IP address

    Explanation

    An internal load balancer uses a private IP address in a subnet as its frontend, so clients reach it only from within the virtual network or connected networks. This satisfies both the distribution requirement and the restriction to internal clients.

  168. Question 168Implement and manage virtual networking

    You are deploying a VPN gateway that must support point-to-site connections and coexist with an ExpressRoute circuit in the same virtual network. Which gateway type should you choose?

    1. RouteBased Correct answer.
    2. Basic
    3. PolicyBased with IKEv2 only
    4. PolicyBased

    Explanation

    Route-based VPN gateways support point-to-site connections, and they are the only type that can coexist with an ExpressRoute circuit in the same virtual network. Policy-based gateways support only site-to-site connections with static routing.

  169. Question 169Implement and manage virtual networking

    Internet users must reach an internal web server on TCP port 8080 by connecting to the Azure Firewall public IP address on port 80. Which kind of rule should you configure?

    1. An application rule in an application rule collection
    2. A service endpoint policy
    3. A network rule in a network rule collection
    4. A DNAT rule in a NAT rule collection Correct answer.

    Explanation

    Destination NAT rules translate the firewall's public IP address and port to a private IP address and port, which is exactly the port translation from 80 to 8080 described here. Network and application rules filter traffic but do not translate addresses or ports.

  170. Question 170Implement and manage virtual networking

    Virtual network encryption is enabled on VNetA, which is peered with VNetB. What else is required before traffic between the two networks is encrypted?

    1. Accelerated networking on every virtual machine
    2. An NSG rule that allows IPsec traffic on both networks
    3. A site-to-site VPN connection between the two networks
    4. Virtual network encryption enabled on VNetB Correct answer.

    Explanation

    Encryption is applied per virtual network, and for peered traffic to be encrypted, the setting must be enabled on each of the peered virtual networks. The virtual machines must also run on supported SKUs.

  171. Question 171Implement and manage virtual networking

    Network security group rules currently have no effect on traffic to a private endpoint in a subnet. Which subnet property must be changed to make the rules apply?

    1. Route table association
    2. PrivateEndpointNetworkPolicies Correct answer.
    3. Service endpoint policies
    4. PrivateLinkServiceNetworkPolicies

    Explanation

    The PrivateEndpointNetworkPolicies property on a subnet controls whether NSG rules and route tables are enforced for private endpoints in that subnet. Setting it to Enabled makes NSG rules apply, while the default of Disabled lets private endpoint traffic bypass them.

  172. Question 172Implement and manage virtual networking

    You must send all internet-bound traffic from Azure Firewall to an on-premises device for inspection. What should you configure on the firewall subnet?

    1. A DDoS protection plan on the firewall virtual network
    2. A NAT rule that redirects traffic to the on-premises network
    3. A service endpoint on the firewall subnet
    4. A 0.0.0.0/0 route with the VirtualNetworkGateway next hop Correct answer.

    Explanation

    Forced tunneling is implemented with a user-defined route for 0.0.0.0/0 that points to the virtual network gateway, causing internet-bound traffic to leave through the on-premises connection. The firewall's management traffic also requires handling, such as a management subnet.

  173. Question 173Implement and manage virtual networking

    VNetA is peered with VNetB, and VNetB is peered with VNetC. No other peering or gateway exists. Which statement about connectivity between VNetA and VNetC is correct?

    1. VNetA can reach VNetC only if all three networks are in the same region
    2. VNetA cannot reach VNetC unless routing is added Correct answer.
    3. VNetA can reach VNetC after the peerings are resynchronized
    4. VNetA can reach VNetC because peering is transitive

    Explanation

    Virtual network peering is not transitive, so a peering between A and B plus a peering between B and C does not connect A to C. Connectivity requires additional routing, such as a network virtual appliance in the hub or a gateway.

  174. Question 174Monitor and maintain Azure resources

    Several Azure Monitor alert rules must email the same on-call team, send an SMS, and call the same webhook. You want to define those notification endpoints once and reuse them. Which Azure resource should you create?

    1. Alert processing rule
    2. Action group Correct answer.
    3. Data collection rule
    4. Diagnostic setting

    Explanation

    An action group is a reusable collection of notification and action endpoints, such as email, SMS, voice, push, webhook, runbook, and Logic App, that alert rules reference. Defining one action group lets many alert rules share the same recipients and actions.

  175. Question 175Monitor and maintain Azure resources

    For how long does Azure Monitor retain platform metrics so that you can interactively explore them in Metrics Explorer?

    1. 7 days
    2. 400 days
    3. 30 days
    4. 93 days Correct answer.

    Explanation

    Platform metrics are retained for 93 days and can be charted interactively in Metrics Explorer during that period. To keep metric data longer, you must route it out, for example with diagnostic settings to a Log Analytics workspace.

  176. Question 176Monitor and maintain Azure resources

    What is the default retention period of the Azure Activity log for a subscription?

    1. 90 days Correct answer.
    2. 60 days
    3. 365 days
    4. 30 days

    Explanation

    The Activity log keeps subscription-level control-plane events for 90 days. For longer retention you route the events to a Log Analytics workspace, a storage account, or an event hub by using a diagnostic setting.

  177. Question 177Monitor and maintain Azure resources

    You want a consolidated list of recommendations for reducing cost, improving reliability, and tightening security across your subscriptions. Which Azure feature provides this?

    1. Azure Service Health
    2. Network Watcher
    3. Application Insights
    4. Azure Advisor Correct answer.

    Explanation

    Azure Advisor analyses configuration and usage telemetry and produces personalized recommendations in the categories of reliability, security, performance, cost, and operational excellence. It evaluates resources across the subscriptions you select.

  178. Question 178Monitor and maintain Azure resources

    A virtual machine is currently unreachable, and you need to determine whether an Azure platform event is responsible rather than a problem inside the guest operating system. Which feature should you check?

    1. Metrics Explorer
    2. Resource Health Correct answer.
    3. Service Health
    4. Azure Advisor

    Explanation

    Resource Health reports the health of an individual resource and distinguishes platform-initiated problems, such as an Azure hardware failure, from issues caused by your own configuration or the guest OS. Service Health, by contrast, reports service-level incidents and advisories for a subscription.

  179. Question 179Monitor and maintain Azure resources

    You need a central repository where log data from many Azure resources is stored and can be queried with KQL. What should you create?

    1. An Event Hubs namespace
    2. A general-purpose storage account
    3. A Recovery Services vault
    4. A Log Analytics workspace Correct answer.

    Explanation

    A Log Analytics workspace is the container in which log data is ingested and stored, and it is queried with Kusto Query Language. Diagnostic settings and monitoring agents send data into workspaces.

  180. Question 180Monitor and maintain Azure resources

    You must determine whether a specific inbound packet sent to a virtual machine is allowed or blocked by the network security group rules that apply to it. Which Network Watcher tool should you use?

    1. Next hop
    2. Packet capture
    3. IP flow verify Correct answer.
    4. Connection monitor

    Explanation

    IP flow verify evaluates a five-tuple of source, destination, port, protocol, and direction against the NSG rules applied to the VM's NIC and subnet, and it reports whether traffic is allowed or denied plus the name of the deciding rule.

  181. Question 181Monitor and maintain Azure resources

    You plan to protect Azure IaaS virtual machines by using Azure Backup. Where are the backup items, recovery points, and policies for those VMs stored?

    1. In a storage account in the same region as the VM
    2. In the VM's temporary disk
    3. In a Recovery Services vault Correct answer.
    4. In a Log Analytics workspace

    Explanation

    Azure Backup for IaaS virtual machines stores backup items, recovery points, and backup policies in a Recovery Services vault. Backup vaults are a different vault type used for workloads such as blobs and disks.

  182. Question 182Monitor and maintain Azure resources

    You must stream resource logs to a third-party SIEM in near real time. Which diagnostic settings destination should you configure?

    1. An event hub namespace Correct answer.
    2. A Log Analytics workspace
    3. An Azure Monitor workbook
    4. A storage account

    Explanation

    Diagnostic settings can send platform logs and metrics to a Log Analytics workspace, an Azure Storage account, or Event Hubs. Event Hubs provides the streaming pipeline that a SIEM or other near real-time consumer can subscribe to.

  183. Question 183Monitor and maintain Azure resources

    You need an alert when the number of failed sign-in events recorded in a Log Analytics workspace exceeds 50 within a 15-minute window. Which type of alert rule should you create?

    1. A metric alert rule
    2. An activity log alert rule
    3. A smart detection alert
    4. A log search alert rule Correct answer.

    Explanation

    Log search alert rules, also called scheduled query rules, run a KQL query against a workspace on a schedule and fire based on the number of results or an aggregated value. They are the correct choice whenever the condition is expressed over log records rather than a platform metric.

  184. Question 184Monitor and maintain Azure resources

    When an alert fires, you need Azure to run a PowerShell script that restarts a service on the affected machine. What should you add to the action group?

    1. An email, SMS, and voice action
    2. An ITSM connector
    3. A webhook
    4. An Automation runbook action Correct answer.

    Explanation

    Action groups support an Automation runbook action that starts a runbook in an Azure Automation account when the alert fires, which is the supported way to perform automated remediation. The runbook can contain the PowerShell logic that restarts the service.

  185. Question 185Monitor and maintain Azure resources

    You have a planned maintenance window. Alerts should still be generated and recorded, but no notifications should be sent while the window is active. What should you configure?

    1. Disable and then re-enable each alert rule
    2. An alert processing rule with suppression Correct answer.
    3. Set each alert rule's severity to Sev4
    4. Create an action group that has no actions

    Explanation

    Alert processing rules act on alert instances after they fire and can apply a suppression action to a chosen scope for a defined time window. During suppression the alerts still fire and are recorded, but their action groups are not invoked.

  186. Question 186Monitor and maintain Azure resources

    You need a KQL query that returns the number of heartbeat records for each computer. Which query clause produces that result when appended to the Heartbeat table?

    1. extend Number = count() and group by
    2. where Computer contains count()
    3. project Computer, count() by Computer
    4. summarize count() by Computer Correct answer.

    Explanation

    The summarize operator groups rows and applies aggregate functions to each group. Combining count() with by Computer produces one row per computer containing the number of records for that computer.

  187. Question 187Monitor and maintain Azure resources

    You need to visualize the call flow, dependencies, and failure rates between the components of a distributed application monitored by Application Insights. Which feature should you use?

    1. Smart detection
    2. Application Map Correct answer.
    3. Availability tests
    4. Live Metrics

    Explanation

    Application Map uses dependency telemetry to draw the components of a distributed application together with the calls between them, their latencies, and their failure rates. It is the standard view for understanding cross-component behaviour.

  188. Question 188Monitor and maintain Azure resources

    You are protecting a virtual machine with Azure Backup and need to change how often backups run and how long recovery points are kept. Where are those settings defined?

    1. In the vault's replication settings
    2. In the backup policy linked to the vault Correct answer.
    3. In the VM extension settings
    4. In the storage account lifecycle management rules

    Explanation

    A backup policy in the vault defines the schedule, the backup frequency, and the retention ranges for instant restore, daily, weekly, monthly, and yearly recovery points. Protected items inherit these settings from the policy they are associated with.

  189. Question 189Monitor and maintain Azure resources

    After you stop backup and delete a backup item from a Recovery Services vault, soft delete keeps the backup data recoverable for how long by default?

    1. 24 hours
    2. 7 days
    3. 14 days Correct answer.
    4. 30 days

    Explanation

    Soft delete is enabled by default on Recovery Services vaults and retains deleted backup data for 14 days. During that period the data can be undeleted; after it expires the data is permanently removed.

  190. Question 190Monitor and maintain Azure resources

    Before performing a real failover with Azure Site Recovery, how should you verify that the replicated virtual machines start correctly without affecting the production environment?

    1. Run a test failover to an isolated network Correct answer.
    2. Disable the Recovery Services vault
    3. Delete the replication and start the VMs manually
    4. Run an unplanned failover to production

    Explanation

    A test failover creates the replica VMs in a network you specify, typically an isolated one, so you can validate boot and connectivity without impacting production. Replication continues unchanged during and after the test failover.

  191. Question 191Monitor and maintain Azure resources

    You need continuous monitoring of latency and reachability between a virtual machine and an on-premises endpoint over time, with the ability to alert on failures. Which Network Watcher capability should you use?

    1. Next hop
    2. Packet capture
    3. Topology
    4. Connection monitor Correct answer.

    Explanation

    Connection Monitor runs continuous reachability and latency tests between endpoints such as Azure VMs, on-premises hosts, and URLs, stores the results over time, and integrates with Azure Monitor alerting. It is the tool designed for ongoing connectivity monitoring rather than one-off diagnostics.

  192. Question 192Monitor and maintain Azure resources

    You are deploying a new solution that must collect Windows event logs from a set of virtual machines into a Log Analytics workspace. Which approach should you use?

    1. Azure Monitor Agent with a data collection rule Correct answer.
    2. The legacy Log Analytics agent with a workspace key
    3. VM insights with the Dependency agent
    4. Diagnostic settings on each virtual machine

    Explanation

    Azure Monitor Agent is the current agent for guest operating system telemetry, and a data collection rule defines which data it collects and which workspace it is sent to. The legacy Log Analytics agent is retired, so new deployments should not rely on it.

  193. Question 193Monitor and maintain Azure resources

    You need to schedule monthly operating system patching for a group of Azure virtual machines by using Azure Update Manager. What defines the schedule and the machines it applies to?

    1. An alert rule
    2. A maintenance configuration Correct answer.
    3. A data collection rule with a transform
    4. A backup policy

    Explanation

    In Azure Update Manager, a maintenance configuration holds the patching schedule and is associated with the machines that should follow it. Update assessments and deployments then run against that schedule.

  194. Question 194Monitor and maintain Azure resources

    You have two Log Analytics workspaces, WS1 and WS2. You need to run one query from WS1 that also returns rows from a table in WS2. What should you use?

    1. A union of the two workspace tables by using join
    2. A cross-workspace query that uses workspace() Correct answer.
    3. A resource-centric query that uses union
    4. A saved search that reads from both workspaces

    Explanation

    The workspace() function references a table in another workspace, for example workspace('WS2').Heartbeat, and the query can be executed from any workspace the user can access. Access to both workspaces is required by the identity running the query.

  195. Question 195Monitor and maintain Azure resources

    You must ensure that telemetry sent to Azure Monitor from your virtual machines, and queries issued from your workstation, never traverse the public internet. Which Azure Monitor feature should you configure?

    1. Diagnostic settings with a storage account
    2. A service endpoint on the workspace subnet
    3. A private endpoint on each monitored VM
    4. An Azure Monitor Private Link Scope Correct answer.

    Explanation

    An Azure Monitor Private Link Scope ties workspaces and Application Insights components to a private endpoint in your virtual network so ingestion and query traffic stay on the Microsoft backbone. Clients then resolve the Azure Monitor endpoints to private IP addresses within the network.

  196. Question 196Monitor and maintain Azure resources

    A metric follows a repeating daily pattern with large peaks that are normal for the workload, and static thresholds produce constant false alarms. Which metric alert configuration best reduces false positives while still catching abnormal behaviour?

    1. A static threshold set to the daily peak
    2. A log search alert with a 1-minute frequency
    3. Dynamic thresholds Correct answer.
    4. An activity log alert on service health

    Explanation

    Dynamic thresholds apply machine learning to the metric's historical behaviour and compute adaptive boundaries, so seasonal peaks are treated as normal and alerts fire only when values deviate from the learned pattern. This avoids the manual tuning that static thresholds require for cyclical metrics.

  197. Question 197Monitor and maintain Azure resources

    You need Azure VM backups to run as often as every four hours. Which backup policy type supports that frequency?

    1. An operational backup policy
    2. A standard backup policy
    3. A snapshot-only policy
    4. An enhanced backup policy Correct answer.

    Explanation

    An enhanced policy for Azure VM backup supports multiple backups per day, as frequently as every four hours, along with longer retention options and support for Trusted Launch VMs. A standard policy allows only one backup per day.

  198. Question 198Monitor and maintain Azure resources

    You need to remove sensitive columns and drop unneeded rows from telemetry before it is stored in the Log Analytics workspace, while continuing to use the Azure Monitor Agent. What should you configure?

    1. A metric alert with dimensions
    2. An alert processing rule
    3. A data collection rule transformation Correct answer.
    4. A workspace table retention and archive setting

    Explanation

    A data collection rule can include a transformKql statement that filters, projects, or aggregates incoming records before ingestion. Because the transformation runs before the data is stored, removed columns never reach the workspace.

  199. Question 199Monitor and maintain Azure resources

    Your team must be notified whenever Microsoft reports an Azure service outage, health advisory, or planned maintenance event that could affect resources in your subscription. Which Azure capability should you configure?

    1. A metric alert rule that monitors the CPU percentage of each production virtual machine
    2. An activity log alert rule that fires on the Restart Virtual Machine administrative operation
    3. An Azure Service Health alert rule that sends notifications through an action group Correct answer.
    4. A weekly review of Azure Advisor recommendations for the subscription

    Explanation

    Service Health communicates Microsoft-initiated events such as service outages, planned maintenance, health advisories, and security advisories for the subscriptions and regions you select. A Service Health alert uses an action group to notify recipients when one of those events is published. The other features observe your own resources or your own control-plane operations rather than Microsoft-reported platform events.

  200. Question 200Monitor and maintain Azure resources

    You have a resource group named RG1 that currently contains 10 virtual machines, and more virtual machines will be added to RG1 over time. You need a single alert rule that raises an alert whenever any virtual machine in RG1 sustains average CPU usage above 90 percent, without editing the rule when new virtual machines are created. What should you create?

    1. An autoscale setting scoped to RG1
    2. A metric alert rule on each virtual machine, plus a new rule added each time a virtual machine is created
    3. A metric alert rule scoped to RG1 that uses the Percentage CPU metric for the virtual machine resource type Correct answer.
    4. An alert processing rule scoped to RG1

    Explanation

    A metric alert rule can be scoped to a subscription or resource group and filtered by resource type and location, so all resources matching that scope are evaluated, including ones created later. Scoping the Percentage CPU signal at the resource group level therefore covers future virtual machines with no rule edits. Rules scoped to individual resources or rules that do not generate alerts cannot meet the requirement.

Azure AZ-104 flashcards

32 cards. Read the term and the definition here. The deck at the top of this page is optional.

Azure RBAC scope levels
Permissions assigned at management group, subscription, resource group, or resource scope are inherited downward, and the union of applicable assignments determines effective access.
Azure Policy versus RBAC
Azure Policy evaluates and enforces resource properties and configurations, while RBAC controls who can perform actions; Policy does not grant permissions and RBAC does not enforce resource shape.
Management group purpose
A container above subscriptions used to organize subscriptions and apply governance, RBAC, and Policy at scale; assignments inherit to all child subscriptions and resource groups.
Resource lock types
CanNotDelete prevents deletion while allowing read and modify; ReadOnly prevents deletion and modification. Locks apply regardless of RBAC role, but authorized users can remove them.
PIM eligible activation
Privileged Identity Management provides just-in-time eligible role activation, often requiring approval, MFA, justification, and a time-bound duration instead of permanent active assignment.
Conditional Access evaluation
Signals such as user, group, location, device, application, and risk are evaluated to grant, block, or require controls such as MFA or a compliant device.
SSPR hybrid requirement
Self-service password reset for hybrid users requires password writeback to on-premises Active Directory and registered authentication methods; administrators usually need two methods.
LRS versus ZRS
LRS replicates three copies within one datacenter in a single region; ZRS replicates across three availability zones in the primary region, protecting against zone failure.
RA-GRS read access
Read-access geo-redundant storage replicates to a secondary region and provides read-only access to the secondary endpoint at any time without initiating a failover.
Blob access tiers
Hot is for frequent access, Cool for infrequent 30-day data, Cold for rare 90-day data, and Archive for offline data requiring rehydration; tiers affect storage and access costs.
Lifecycle management rule
A policy on a storage account that transitions blobs between tiers or deletes them based on conditions such as last-modified age, creation age, or prefix filters.
Stored access policy
A policy defined on a container or share that supplies start time, expiry, and permissions for a service SAS; changing or revoking it invalidates linked SAS tokens.
Azure Files identity auth
SMB access can use on-premises AD DS, Entra Domain Services, or Entra Kerberos for identity-based authentication; shared key and SAS are also possible but less granular.
Blob soft delete
Retains deleted blobs or snapshots for a configured retention period, allowing recovery; it is separate from blob versioning and from container soft delete.
Storage firewall default
When the firewall is enabled, requests are denied unless the client IP, virtual network service endpoint, or private endpoint is allowed; trusted Microsoft services can be exempted.
Availability set fault domains
VMs in an availability set are distributed across fault domains, which share power and network, and update domains, which share maintenance reboots within one datacenter.
Availability zones versus sets
Zones separate VMs across physically distinct datacenters within a region, protecting against datacenter failure; availability sets protect only against hardware and maintenance within one datacenter.
VMSS autoscale rule
A rule adjusts instance count based on a metric, schedule, or predictive model, with minimum and maximum bounds; scale-in and scale-out rules can differ.
Managed disk caching
None, ReadOnly, and ReadWrite apply to OS and data disks; ReadOnly is typical for read-heavy data disks, while ReadWrite is common for OS disks but can affect write durability.
Custom Script Extension
Downloads and runs a script on a VM after provisioning using a storage URI or inline command; it is useful for post-deployment configuration and runs once unless updated.
Proximity placement group
Constrains VMs or scale sets to physically close hardware to reduce network latency; it can limit available capacity and should be used only when latency matters.
App Service plan
Defines the region, operating system, pricing tier, and compute capacity for hosted apps; scaling the plan changes resources for all apps assigned to that plan.
NSG default rules
Built-in rules allow traffic within the virtual network and from Azure Load Balancer, deny other inbound internet traffic, and allow outbound internet; custom rules use priority.
VNet peering transitivity
Peering is nontransitive by default: spoke-to-spoke traffic does not flow through a hub unless a gateway, firewall, or user-defined route is configured to forward it.
Service versus private endpoint
A service endpoint extends the virtual network identity to a PaaS service over the Azure backbone; a private endpoint assigns a private IP address in the subnet to the service.
UDR next hop types
User-defined routes can specify virtual network gateway, virtual network, internet, virtual appliance, or none; the route with the longest prefix match takes precedence.
Azure Bastion purpose
Provides browser-based RDP and SSH to VMs through the Azure portal without public IP addresses or opening inbound management ports to the internet.
Load Balancer versus App Gateway
Azure Load Balancer operates at layer 4 for TCP and UDP; Application Gateway operates at layer 7 with HTTP routing, TLS termination, and optional web application firewall.
Private DNS zone link
A private DNS zone must be linked to each virtual network that needs to resolve its records; auto-registration can create VM records in linked virtual networks.
Metrics versus logs
Metrics are numeric time-series stored for near-real-time alerting and charts; logs are structured records queried with KQL in a Log Analytics workspace.
Action group
A reusable set of notification and action receivers, such as email, SMS, webhook, Logic App, Function, or Automation runbook, invoked by Azure Monitor alerts.
Diagnostic setting
Routes platform logs and metrics from a resource to a Log Analytics workspace, storage account, event hub, or partner solution; it is configured per resource.

Azure AZ-104 study guide

Manage Azure identities and governance

Scope and inheritance are the backbone

Every Azure RBAC assignment is a triple: a security principal (user, group, service principal, or managed identity), a role definition, and a scope. Scopes nest as management group > subscription > resource group > resource, and permissions flow downward. Assignments are additive, so a Reader at the management group plus Contributor at one resource group yields Contributor there. Owner manages everything including access; Contributor manages everything except access; User Access Administrator manages only access. Deny assignments, produced by managed applications and blueprints, override any allow.

Policy, RBAC, and locks answer different questions

RBAC answers who can act. Azure Policy answers what a resource may look like; it evaluates on create and update, and existing resources change only through a remediation task. Effects worth memorizing: Deny, Audit, Append, Modify, DeployIfNotExists, AuditIfNotExists, Disabled. An initiative is a group of policy definitions. Resource locks answer whether an action can complete: CanNotDelete and ReadOnly. Locks ignore RBAC, so even a subscription Owner must remove the lock before deleting the resource.

Identity plumbing

Entra ID holds users (member or guest), groups (assigned, dynamic user, dynamic device; dynamic membership needs P1 licensing), service principals, and managed identities. A system-assigned identity shares the resource lifetime; a user-assigned identity stands alone and can be attached to several resources. Treat SSPR, MFA, Conditional Access, and PIM as separate controls.

Worked rule of thumb

When a task is blocked, ask three questions in order: is there a lock, does Policy deny or modify, and does the role allow it? That order usually matches the answer choices. Tags do not inherit from a resource group automatically, so enforce them with a Modify policy.

Implement and manage storage

Choose the account type before you create it

Most storage questions are decided at creation time. Standard general-purpose v2 covers blobs, files, queues, and tables. Premium accounts are specialized for block blobs, file shares, or page blobs, each with its own redundancy options. Redundancy: LRS keeps three copies in one datacenter; ZRS spreads three copies across availability zones in one region; GRS copies LRS data asynchronously to a paired region; RA-GRS adds read access to that secondary; GZRS and RA-GZRS combine zones with region pairing. After a failover, the account becomes locally redundant in the former secondary region.

Blobs, tiers, and lifecycle

Blob access tiers are hot, cool, cold, and archive, set as an account default or per blob. Archive is offline: reads fail until rehydration completes, which can take hours (high priority is faster and costs more). Lifecycle management rules automate tiering and deletion using filters on base blobs, snapshots, and versions. Container access levels are private, blob, and container; many subscriptions now disable anonymous access entirely.

Securing and sharing data

Every account has shared key access; prefer Entra ID authorization with a user delegation SAS where possible. Service and account SAS tokens can be revoked only when tied to a stored access policy. Soft delete, versioning, and immutable (WORM) policies protect against deletion and tampering. Azure Files offers SMB and NFS shares with share snapshots.

Worked rule of thumb

Map requirements to redundancy first: read access during an outage means RA-GRS or RA-GZRS; zone-failure tolerance means ZRS or GZRS. Then choose tiers, because later replication conversion is restricted. Traps: assuming archive blobs are readable, assuming a SAS can be revoked without a stored access policy, and assuming the secondary endpoint is current.

Deploy and manage Azure compute resources

Pick the compute service, then the size

VMs, VM scale sets, App Service, Container Instances, AKS, and Functions each map to a scenario. For VMs, know the size families (B burstable, D general purpose, E memory optimized, F compute optimized, N GPU) and the disk types: Standard HDD, Standard SSD, Premium SSD, Premium SSD v2, and Ultra Disk. The temporary disk is not persisted, and only the deallocated state stops compute billing.

Availability is a design decision

An availability set uses fault domains and update domains inside one datacenter and targets 99.95%. Availability zones spread instances across datacenters in a region and target 99.99%. Neither can be changed after a VM is created, so the choice is permanent. Scale sets add autoscale rules and, when created with zones, zone balancing.

Deployment and configuration

ARM templates and Bicep deploy declaratively. Incremental mode adds and updates; complete mode deletes resources in the resource group that are absent from the template, a classic exam trap. Run what-if before deploying. VM extensions such as Custom Script and Desired State Configuration handle post-provisioning configuration.

App Service specifics

The App Service plan is the billing and scale unit, and several apps can share it. Scale up changes the plan tier for more CPU and memory; scale out adds instances and requires a tier that supports it. Deployment slots let you stage a version and swap it into production.

Worked rule of thumb

To resize a VM, check whether the target size exists on the current hardware cluster. If not, deallocate the VM, resize, then start it. Expect questions where a resize fails only because the VM is still running.

Implement and manage virtual networking

Address space and subnetting

Plan VNet address space so it never overlaps another VNet you might peer with or any on-premises range. Azure reserves five IP addresses in every subnet (network address, gateway, two for DNS, and broadcast), so a /24 yields 251 usable addresses. Some subnets require exact names: GatewaySubnet for VPN and ExpressRoute gateways, AzureFirewallSubnet for Azure Firewall, and AzureBastionSubnet for Bastion.

Filtering traffic

Network security groups contain inbound and outbound rules with priorities from 100 to 4096; the lower number wins. Default rules exist in every NSG (AllowVnetInBound 65000, AllowAzureLoadBalancerInBound 65001, DenyAllInBound 65500). NSGs attach to subnets and NICs, and both are evaluated. Application security groups group NICs so rules can reference a name instead of IP addresses.

Routing and connectivity

System routes handle VNet-local and internet traffic. User-defined routes override them, with next hop types such as virtual appliance, virtual network gateway, virtual network, or internet. Precedence is UDR, then BGP, then system routes. VNet peering is non-transitive: A peered to B and B peered to C does not connect A to C unless you add a route and a gateway, or use a hub firewall. Peering also requires non-overlapping address spaces.

Private access to PaaS

Service endpoints extend VNet identity to a PaaS service over the Azure backbone, and the service firewall allows that subnet. Private endpoints instead give the service a private IP in your subnet through Private Link, which needs a matching private DNS zone to resolve.

Worked rule of thumb

Troubleshoot connectivity in this order: DNS resolution, NSG rules, UDR next hop, then peering or gateway.

Monitor and maintain Azure resources

Metrics, logs, and the activity log differ

Platform metrics are numeric, near real time, and retained for 93 days; you can alert on them but cannot query them with KQL. Resource logs flow to a Log Analytics workspace and are queried with KQL; interactive retention defaults to 30 days and can extend to 730, with long-term archival to a storage account. The activity log records control-plane events such as who created, changed, or deleted a resource, and keeps 90 days on its own; route it with a diagnostic setting to keep it longer or alert on it.

Diagnostic settings and agents

A diagnostic setting sends a resource's logs and metrics to one or more destinations: Log Analytics workspace, storage account, Event Hubs, or a partner solution. The Azure Monitor Agent, configured through data collection rules, replaces the legacy Log Analytics agent. Application Insights adds request, dependency, and exception telemetry for applications.

Alerting

An alert rule combines a signal (metric, log query, or activity log), a condition, a severity from 0 to 4, and an action group that delivers email, SMS, voice, push, webhook, Logic App, runbook, or ITSM actions. Alert processing rules suppress notifications during maintenance windows. Service Health alerts cover platform incidents, Resource Health covers a specific resource, and Advisor offers recommendations.

Backup and network diagnostics

Azure Backup uses a Recovery Services vault, backup policies, and soft delete. Network Watcher provides IP flow verify, next hop, connection troubleshoot, and packet capture.

Worked rule of thumb

Use metric alerts for simple numeric thresholds evaluated quickly and cheaply, and log alerts for multi-condition or stateful checks. If a question asks how to prove a VM cannot reach a database endpoint, the answer is usually Network Watcher IP flow verify or connection troubleshoot.

Azure AZ-104 cheat sheet

Quick reference

Manage Azure identities and governance

  • Azure RBAC assignments at management group, subscription, resource group, or resource scope are inherited by child scopes; effective access is the union of applicable role assignments.

  • Azure Policy evaluates resource properties and can deny, audit, append, or deploy remediation; it does not grant permissions, while RBAC controls who can perform actions.

  • Management groups sit above subscriptions and can organize many subscriptions so that RBAC and Policy assignments inherit to all child subscriptions and resource groups.

  • Resource locks are CanNotDelete and ReadOnly; they apply to all users regardless of RBAC role, and only users with sufficient permissions at the lock scope can remove them.

  • Microsoft Entra Privileged Identity Management provides eligible, time-bound role activation with optional approval, MFA, and justification instead of permanent active assignments.

  • Conditional Access evaluates signals such as user, group, location, device, application, and sign-in risk to grant, block, or require controls such as MFA or a compliant device.

  • Self-service password reset for hybrid users requires password writeback to on-premises Active Directory; administrators typically need two authentication methods registered.

  • Azure Policy can enforce required tags on new or updated resources, but existing untagged resources require a remediation task to be brought into compliance.

Quick reference

Implement and manage storage

  • LRS keeps three copies in one datacenter in a single region; ZRS spreads three copies across availability zones; GRS stores six copies across two regions; RA-GRS adds read access to the secondary region.

  • Blob access tiers are Hot, Cool, Cold, and Archive; Cool has a 30-day minimum, Cold 90 days, and Archive 180 days, with early deletion charges if data leaves sooner.

  • Lifecycle management rules can transition blobs between tiers or delete current blobs, previous versions, and snapshots based on age, prefix, or index tags.

  • A stored access policy on a container or file share defines start time, expiry, and permissions for a service SAS; changing or deleting the policy revokes SAS tokens that reference it.

  • User delegation SAS is secured by Microsoft Entra credentials and is supported for Blob Storage; account and service SAS are secured by the storage account key unless a stored access policy is used.

  • When a storage account firewall is enabled, requests are denied by default unless the client IP address, virtual network service endpoint, or private endpoint is allowed; trusted Microsoft services can be exempted.

  • Azure Files supports SMB and NFS shares; identity-based SMB access can use on-premises AD DS, Entra Domain Services, or Entra Kerberos, while shared key and SAS authentication are also available.

  • Blob soft delete, blob versioning, and immutable WORM retention are separate features; soft delete retains deleted data for a configured period, and versioning keeps prior blob states.

Quick reference

Deploy and manage Azure compute resources

  • An availability set distributes VMs across fault domains and update domains within one datacenter; availability zones distribute VMs across physically separate datacenters in a region.

  • A Virtual Machine Scale Set autoscale profile defines minimum, maximum, and default instance counts plus scale-out and scale-in rules, which can include a cool-down period.

  • Managed disk types include Standard HDD, Standard SSD, Premium SSD, Premium SSD v2, and Ultra Disk; disk caching options are None, ReadOnly, and ReadWrite.

  • The Custom Script Extension downloads and runs a script after VM provisioning; it is useful for post-deployment configuration and does not run again unless the extension is updated.

  • A proximity placement group constrains VMs or scale sets to physically close hardware to reduce latency, but it can reduce available capacity and should be used only when latency requires it.

  • An App Service plan defines region, operating system, pricing tier, and compute capacity; scaling the plan affects every app assigned to that plan, while scaling out adds instances.

  • ARM and Bicep deployments are idempotent and support incremental or complete modes; complete mode can delete resources in the target resource group that are not defined in the template.

  • Azure Backup uses Recovery Services vaults for IaaS VM backups and Backup vaults for newer workloads; soft delete can retain deleted backup data for a configured period.

Quick reference

Implement and manage virtual networking

  • Virtual network address spaces must not overlap when peered, and peering is nontransitive by default; spoke-to-spoke traffic needs a gateway, firewall, or user-defined route to forward it.

  • Azure reserves five IP addresses in each subnet: the network address, three Azure-reserved addresses, and the broadcast address, so usable addresses are total minus five.

  • NSG rules use priority numbers from 100 to 4096, with lower numbers evaluated first; default rules include AllowVNetInBound, AllowAzureLoadBalancerInBound, DenyAllInBound, and outbound equivalents.

  • Application security groups group NICs so NSG rules can reference an application tier as source or destination instead of listing individual IP addresses.

  • User-defined routes support next-hop types Virtual network gateway, Virtual network, Internet, Virtual appliance, and None; the longest prefix match determines which route is used.

  • Azure Bastion provides browser-based RDP and SSH through the portal without public IPs on VMs and requires a dedicated subnet named AzureBastionSubnet of at least /26.

  • Azure Load Balancer operates at layer 4 for TCP and UDP; Application Gateway operates at layer 7 with HTTP routing, TLS termination, and optional Web Application Firewall.

  • A service endpoint extends the virtual network identity to a PaaS service over the Azure backbone, while a private endpoint assigns a private IP address in the subnet to the service.

Quick reference

Monitor and maintain Azure resources

  • Azure Monitor metrics are numeric time-series suited to near-real-time charts and alerts; logs are structured records queried with KQL in a Log Analytics workspace.

  • An alert rule combines a scope, condition, action group, severity, and evaluation frequency; the action group defines notifications and automated actions such as webhooks or runbooks.

  • A diagnostic setting routes platform logs and metrics from a resource to a Log Analytics workspace, storage account, event hub, or partner solution and is configured per resource.

  • Log Analytics workspaces have configurable retention for interactive queries and can archive data for longer periods; data collection rules define which data is collected and where it is sent.

  • Application Insights provides request, dependency, exception, and availability telemetry for applications, including smart detection and availability tests.

  • Azure Service Health reports Azure incidents, planned maintenance, and health advisories affecting your subscriptions; Resource Health reports the health of individual resources.

  • Azure Advisor provides recommendations across cost, security, reliability, operational excellence, and performance, but it does not automatically remediate every recommendation.

  • Network Watcher tools include IP flow verify, connection troubleshoot, packet capture, and NSG diagnostics for validating connectivity and security rule behavior.

Azure AZ-104 key terms

Microsoft Entra ID

Microsoft's cloud directory and identity service. It stores users, groups, and app registrations, and issues the tokens used to authenticate and authorize access to Azure, Microsoft 365, and custom applications.

Related: Tenant, Conditional Access, Managed identity
Tenant

A dedicated instance of Microsoft Entra ID created for an organization when it signs up for a Microsoft cloud service. It holds that organization's directory objects and is identified by a unique default domain name.

Related: Microsoft Entra ID
Resource group

A logical container holding related Azure resources for one solution. Resources are deployed, managed, and deleted together as a unit, and each resource belongs to exactly one resource group.

Related: Management group, Resource lock, Role-based access control (RBAC)
Management group

A container that organizes subscriptions into a hierarchy so that governance assignments, such as Azure Policy and role assignments, inherit down to every subscription beneath it.

Related: Resource group, Azure Policy
Role-based access control (RBAC)

Azure's authorization system. A role definition lists allowed actions, and assigning that role to a user, group, service principal, or managed identity at a scope grants those permissions at and below that scope.

Related: Managed identity, Resource lock
Azure Policy

A governance service that compares resources with rules called policy definitions. It can block noncompliant resource creation, audit existing resources, or deploy remediation, and definitions can be grouped into initiatives.

Related: Management group, Resource lock
Resource lock

A CanNotDelete or ReadOnly setting applied to a subscription, resource group, or resource. It blocks deletion or changes even for users who hold owner-level RBAC permissions.

Related: Resource group, Role-based access control (RBAC)
Managed identity

A service principal created and rotated automatically by Azure in Microsoft Entra ID. Code on a VM, app, or other resource uses it to request tokens for other services without storing secrets.

Related: Microsoft Entra ID, Role-based access control (RBAC)
Conditional Access

A Microsoft Entra ID policy engine that evaluates signals such as user, group, location, device state, and sign-in risk, then allows, blocks, or challenges access, for example by requiring multifactor authentication.

Related: Microsoft Entra ID
Storage account

A top-level Azure namespace that exposes blob, file, queue, and table endpoints under a unique name. It sets the redundancy, performance tier, and default access tier used by the data it holds.

Related: Storage redundancy, Access tier, Azure Files
Access tier

A per-blob or per-account setting (hot, cool, cold, or archive) that balances storage cost against access cost and latency. Archived blobs must be rehydrated before they can be read.

Related: Storage account, Lifecycle management rule
Azure Files

Fully managed SMB and NFS file shares hosted in a storage account. Shares can be mounted from Windows, Linux, and macOS, and on-premises servers can sync with them using Azure File Sync.

Related: Storage account, Shared access signature (SAS)
Shared access signature (SAS)

A signed token appended to a storage URL that grants specific permissions for a limited time. Account, service, and user delegation variants let clients reach data without the account key.

Related: Stored access policy, Storage account
Stored access policy

A named policy defined on a blob container, file share, queue, or table that stores start time, expiry, and permissions. SAS tokens tied to it can be revoked or adjusted centrally.

Related: Shared access signature (SAS)
Storage redundancy

The replication option chosen for a storage account: locally redundant within one region, zone redundant across availability zones, or geo-redundant to a paired region, with read-access variants of the geo options.

Related: Storage account
Lifecycle management rule

A rule on a storage account that automatically transitions blobs between access tiers or deletes them when conditions such as days since last modification or last access are met.

Related: Access tier, Storage account
AzCopy

A command-line tool for fast, resumable copies of files and blobs to, from, and between storage accounts, commonly used for migrations and bulk data movement.

Related: Storage account, Azure Files
Virtual machine

An IaaS compute resource with a chosen size, operating system image, disks, and network interface. You manage the guest operating system, patching, and scaling yourself.

Related: Availability set, Availability zone, VM extension
Availability set

A grouping of VMs that Azure spreads across fault domains and update domains inside a single datacenter, protecting workloads from hardware failure and planned maintenance at the same time.

Related: Availability zone, Virtual machine
Availability zone

A physically separate datacenter group within an Azure region with independent power and cooling. Deploying VMs across zones gives higher availability than an availability set.

Related: Availability set, Virtual machine
Virtual Machine Scale Sets

A compute resource that creates and manages a set of identical VMs from one configuration. It can autoscale on metrics or a schedule and integrates with load balancers.

Related: Virtual machine, Azure Load Balancer
ARM template

A JSON file, or the Bicep language that compiles to JSON, that declaratively describes the resources and configuration for a deployment so Azure Resource Manager can create or update them repeatably.

Related: Resource group
VM extension

A small package installed on a virtual machine after deployment that adds management capability, such as running a script, installing an agent, or joining a domain.

Related: Virtual machine
App Service plan

The compute backing for Azure App Service apps. It defines the region, operating system, pricing tier, and the number and size of instances that the hosted apps share.

Azure Kubernetes Service

A managed Kubernetes service in which Azure runs the control plane while you deploy node pools built from VMs or scale sets to host containerized workloads.

Related: Virtual Machine Scale Sets
Virtual network

A private network boundary in Azure. It defines an address space, holds subnets, and provides isolation and routing for the resources connected to it.

Related: Subnet, VNet peering
Subnet

A range of IP addresses within a virtual network's address space where resources such as VMs and private endpoints are placed. Network security groups and route tables attach at this level.

Related: Virtual network, Network security group
Network security group

A set of allow and deny rules that filter inbound and outbound traffic by source, destination, port, and protocol. It can be attached to a subnet, a network interface, or both.

Related: Subnet
VNet peering

A connection that links two virtual networks so resources communicate using private IP addresses over Microsoft's backbone. Peering can be regional or global and is not transitive by default.

Related: Virtual network
User-defined route

A route table entry that overrides Azure's default system routes to send subnet traffic to a chosen next hop, such as a firewall, network virtual appliance, or VPN gateway.

Related: Subnet, Virtual network
Azure Load Balancer

A layer-4 service that distributes TCP and UDP traffic across backend VMs or scale set instances using a public or internal frontend, health probes, and load-balancing rules.

Related: Virtual Machine Scale Sets, Application Gateway
Application Gateway

A layer-7 web traffic manager that routes HTTP and HTTPS requests based on URL paths and host headers, and can provide TLS termination, cookie affinity, and a web application firewall.

Related: Azure Load Balancer
Private endpoint

A network interface in a subnet that connects privately to a service such as storage or Azure SQL through Azure Private Link, giving that service a private IP inside your virtual network.

Related: Virtual network, Subnet
Azure Bastion

A managed jump-host service deployed into a dedicated subnet that gives browser-based RDP and SSH access to VMs over TLS without exposing public IP addresses on those VMs.

Related: Virtual network, Subnet
Azure Monitor

The platform that collects metrics, activity logs, and resource logs from Azure resources and lets you query, visualize, and create alerts on that telemetry.

Related: Log Analytics workspace, Action group, Diagnostic settings
Log Analytics workspace

A data store in Azure Monitor where log data from multiple resources is collected and queried with Kusto Query Language. Diagnostic settings and monitoring agents send data into it.

Related: Azure Monitor, KQL
Action group

A reusable list of notification and automation targets, such as email, SMS, webhook, Azure Function, or Logic App, that an alert rule invokes when it fires.

Related: Azure Monitor
Diagnostic settings

A per-resource configuration that sends platform metrics and resource logs to a Log Analytics workspace, a storage account, an event hub, or a partner solution.

Related: Azure Monitor, Log Analytics workspace
Recovery Services vault

An Azure Backup container that stores backup data, backup policies, and recovery points for virtual machines, file shares, and other workloads, and supports restore operations.

Related: Virtual machine
KQL

Kusto Query Language, the read-only query language used in Log Analytics and Azure Monitor to filter, summarize, joins, and chart log tables.

Related: Log Analytics workspace

Azure AZ-104 study plan

8 weeks · About 6 hours per week

Week 1

Manage Azure identities and governance

  • Create Entra ID users, security groups, and at least one dynamic membership group in a sandbox tenant

  • Assign built-in Azure RBAC roles at subscription, resource group, and individual-resource scope, then verify effective permissions with the access check tools

  • Configure self-service password reset and multifactor authentication for a pilot group of lab users

  • Map three common job functions to least-privilege role assignments and justify each choice in writing

  • Differentiate Entra ID directory roles from Azure RBAC roles using two concrete lab scenarios

Your checkpoint

A documented identity lab: 10 users, 3 groups, 4 least-privilege role assignments, and screenshots proving effective access for each assignment.

Week 2

Manage Azure identities and governance

  • Design and implement a management group hierarchy and place resources under it

  • Author a custom Azure Policy with a tagging requirement and an allowed-locations rule, then remediate noncompliant resources

  • Apply resource locks and consistent tag schema to a resource group and test the effect of each lock type

  • Configure a Cost Management budget with an alert threshold and tags-based cost filtering

  • Export a policy compliance report and interpret which resources remain noncompliant

Your checkpoint

A governance pack containing the management group diagram, exported policy compliance report at 90% or better after remediation, and a budget alert screenshot.

Week 3

Deploy and manage Azure compute resources

  • Author a reusable Bicep or ARM template that deploys a virtual network, network interface, and Linux VM from parameters

  • Deploy the same workload into an availability zone and into an availability set, and state the failure-domain difference

  • Attach and resize managed data disks, then change the VM size and confirm the expected downtime behavior

  • Deploy a virtual machine scale set with autoscale rules based on CPU and validate scale-out under synthetic load

  • Select an appropriate VM size and disk type for three given cost and performance scenarios

Your checkpoint

A template repository with three parameterized deployments plus captured evidence of a scale set scaling from two to four instances and back.

Week 4

Deploy and manage Azure compute resources

  • Create an App Service plan and web app, then configure a deployment slot and perform a slot swap

  • Bind a custom domain and TLS certificate to an App Service app and enforce HTTPS only

  • Deploy a containerized workload to Azure Container Instances and compare it against App Service and AKS for two scenarios

  • Configure App Service application settings and connection strings, and explain which ones are slot-sticky

  • Complete a timed 50-question practice set covering domains 1 and 3 and log every missed objective

Your checkpoint

A hosted App Service app reachable over HTTPS with a completed slot swap, plus a scored practice test above 75% and a written remediation list.

Week 5

Implement and manage storage

  • Create storage accounts with LRS, ZRS, and GRS replication and explain the durability and failover differences

  • Configure blob access tiers, a lifecycle management rule, soft delete, and versioning, then verify a blob moves to cool storage

  • Generate account SAS, service SAS, and stored access policies, then test each permission boundary

  • Create an Azure Files share, take a snapshot, and mount the share from a lab VM

  • Move data with AzCopy and compare it with Storage Explorer and the Azure CLI for the same task

Your checkpoint

A storage lab report showing before-and-after blob inventory after a lifecycle rule ran, plus a permissions matrix proving each SAS token works only as intended.

Week 6

Implement and manage virtual networking

  • Build a hub-and-spoke topology with virtual network peering and demonstrate which traffic does and does not flow by default

  • Author network security groups with prioritized rules and validate them with IP flow verify and effective security rules

  • Configure an Azure DNS private zone and confirm name resolution from a VM in a peered network

  • Deploy a load balancer with a health probe and backend pool, then remove one backend and observe probe failure

  • Configure a VPN gateway connection for a site-to-site design and document the required address space and gateway subnet

Your checkpoint

A connectivity test matrix recording the result of at least eight source-to-destination tests with the explaining NSG or peering rule for each.

Week 7

Monitor and maintain Azure resources

  • Create a Log Analytics workspace and route VM diagnostic data and activity logs into it

  • Write and save five KQL queries covering failed sign-ins, VM performance counters, and resource changes

  • Create metric alerts and log search alerts with an action group and trigger at least one on purpose

  • Configure Azure Backup for a VM, run a backup, and execute a file-level restore to a new location

  • Enable Network Watcher connection monitor or NSG flow logs and interpret the output

Your checkpoint

A monitoring pack with five saved KQL queries, five working alert rules, and a restore-test record showing recovery time and recovered file paths.

Week 8

Manage Azure identities and governance & Implement and manage storage & Deploy and manage Azure compute resources & Implement and manage virtual networking & Monitor and maintain Azure resources

  • Complete two full-length timed practice exams under exam conditions and review every incorrect answer

  • Rebuild the two weakest labs from memory without referencing notes to confirm retention

  • Re-read only the objectives flagged as missed across weeks 1 through 7 and close each gap with a hands-on task

  • Rehearse the exam interface features and time allocation for case-study style questions

Your checkpoint

Two full-length mock exams scored at 80% or higher plus a closed remediation checklist with zero open items.

Milestones

Week 2: Assign a least-privilege RBAC role and author a tagging policy from a blank template in under 20 minutes each, without consulting notes.

Week 4: Score 75% or higher on a timed 50-question set covering domains 1 and 3, with no objective missed twice.

Week 6: Score 75% or higher on a timed set covering domains 2 and 4, and correctly explain the outcome of all eight connectivity matrix tests.

Week 8: Score 80% or higher on two full-length timed mock exams and demonstrate every flagged objective hands-on at least once.

Azure AZ-104 exam day

Your checklist

Government-issued photo ID

Use the same name as your exam registration. For online proctoring it is typically one government-issued photo ID with signature; some test centres request a second form. Exact ID requirements vary by location - confirm with the provider before exam day.

Exam appointment confirmation and time zone

Note your scheduled start time, the delivery method (test centre vs. online proctored), and the time zone shown on your confirmation. Convert it to your local time to avoid a one-hour-early or one-hour-late arrival.

Quiet, private, well-lit room (online proctored only)

Proctors generally require a closed door, no other people, a clear desk, and a power outlet within reach. Details on permitted items (water in a clear container, tissues, etc.) vary - confirm with the provider.

Working webcam, microphone, and speakers

Built-in laptop hardware is usually fine. Wired or reliable headphones may not be permitted; check what the provider allows rather than assuming.

Stable internet connection plus a backup option

Have a phone with a hotspot or the provider's support number ready so a connection drop doesn't cost you the appointment.

Completed system/equipment test

Run the proctoring vendor's compatibility check on the exact machine you will test on, ideally 24-72 hours ahead so you have time to fix issues.

Cleared workspace and closed background applications

Before check-in, close email, chat, sync clients, and any app that might pop a notification. Notifications on screen can be flagged as a policy violation.

Confirmed the current exam format and section count

Microsoft role-based exams can include multiple-choice, case studies, drag-and-drop, and - on some exams at some times - interactive lab-style sections. The mix changes; do not build your timing plan on a rumour. needs_review - confirm the current format for your exam with the provider.

Approved testing accommodations (if you need them)

Accommodations generally must be requested and approved before the appointment, not on exam day. needs_review - confirm the request window with the provider.

Free Microsoft Learn practice assessment already taken

Do this a day or more before so results can guide your last review, not on the morning of the exam.

Login credentials for the scheduling vendor account

Keep the username and password accessible; a forgotten password five minutes before check-in is a common avoidable failure.

Rest, meal, and restroom handled before check-in

Treat the window as closed once you begin; plan breaks assuming you will not want to leave mid-exam.

Plan the day

72 hours before

Stop adding new topics. Re-read your notes only on the two or three objectives you score lowest on - typical weak spots for this exam are virtual networking (peering, routing, NSGs), storage redundancy and access tiers, and backup/recovery. Take the free practice assessment from Microsoft Learn once, review every wrong answer, and write down the reason it was wrong.

72 hours before

Reconcile your appointment: date, start time, time zone, delivery mode, and the address or check-in link. Reschedule now if anything conflicts - last-minute changes usually have cut-off rules, so confirm the provider's policy rather than guessing.

72 hours before

Run the equipment and internet check on the machine you will actually use. If testing online, test from the room you will sit in, not the kitchen table.

24 hours before

Do one light pass only: ARM resource hierarchy and RBAC scope inheritance, moving resources between subscriptions/resource groups, virtual network and subnet sizing, NSG vs. Azure Firewall use cases, load balancer vs. Application Gateway vs. Front Door, storage account kind/replication/access tier choices, and what Azure Backup vs. Azure Site Recovery each cover. Skim, don't memorise new facts.

24 hours before

Prepare logistics: ID out of the drawer, room tidied, desk cleared, charger packed or plugged in, a snack and water ready. Tell anyone sharing your space that you need quiet during the appointment window.

24 hours before

Sleep a normal night. Cramming until 2 a.m. reliably costs more marks than an extra hour of revision gains.

Morning of

Eat, hydrate, and re-read only your own one-page summary of mistakes from the practice test. No new material.

Morning of

Log in or arrive early. Check-in often opens roughly 30 minutes before the start time and includes ID verification and a room scan, so build in buffer rather than cutting it fine.

During the test

Work in passes. First pass: answer everything you know cold. Flag anything needing calculation or close reading - often subnet maths, cost/throughput comparisons, or 'which is cheapest / least effort' phrasing. Second pass: return to flagged items.

During the test

Watch the clock against the number of remaining questions and leave a few minutes at the end. Never leave a question blank - there is no penalty for a reasoned guess.

During the test

If a case study or lab-style section appears, read the final requirement first and only then the exhibits. Note that once you leave a case study set you may not be able to return to it - confirm the navigation rules shown on screen at the time.

After the test

Results are usually displayed on screen at the end and appear in your certification profile later. Save or print the score report; if you did not pass, it shows your performance by objective domain, which is the best study plan you can get.

Rules to check

  • Bring and present a valid government-issued photo ID whose name matches your registration. Name mismatches are a common reason candidates are turned away, so fix the registration before exam day. Specific accepted document types vary - confirm with the provider.

  • No notes, books, second screens, phones, smart watches, or unauthorised software are permitted during the exam. For online proctoring, keep the phone out of reach unless the proctor instructs you to use it for check-in.

  • You generally cannot leave during the exam for a break unless the exam explicitly offers one, and if a break is offered the clock may or may not stop. Check the on-screen instructions and confirm the provider's policy rather than assuming.

  • The exam is scored on a scaled scale, and the passing score is 700 out of 1000. That is not '70% correct' - some questions may be weighted differently and some may be unscored pretest items. Answer every question; do not try to game the unscored ones.

  • Attempt limits and waiting periods after a failed attempt are set by the provider and can change. needs_review - confirm the current retake policy, including any waiting period and whether a free retake is included with your purchase, with the provider.

  • Do not share, record, photograph, or discuss specific exam content afterwards. Doing so breaches the non-disclosure agreement you accept at the start of the exam and can invalidate your result and certification.

  • Certification is not permanent at every level: this certification typically requires periodic free renewal through an online assessment on Microsoft Learn. Confirm the renewal window for your certification in your learn profile rather than relying on memory.

  • If something goes wrong mid-exam - disconnection, proctor issue, power loss - stop and contact the provider's support immediately and get a case reference. Do not simply restart; the record of the incident is what supports a retake or extension.

Azure AZ-104 FAQ

How many Azure AZ-104 practice questions are available?

The practice set has 200 Azure AZ-104 questions. Each one includes the choices, the correct answer, and an explanation, and the set is free to read with no signup.

How many Azure AZ-104 flashcards are available?

The flashcard set has 32 Azure AZ-104 cards. Each card shows the term and its definition.

Which topics does the Azure AZ-104 study material cover?

The Azure AZ-104 study material has 5 sections, including Manage Azure identities and governance, Implement and manage storage, and Deploy and manage Azure compute resources.

What format facts does the catalog list for the Azure AZ-104 exam?

The catalog lists a duration of 100 minutes and these item types: multiple choice and interactive components for the Azure AZ-104 exam.