MS-102 domain - 26% of the exam

Implement and Manage Identity and Access

Implement and Manage Identity and Access is 26% of the MS-102 exam. These are the objectives it covers, each with practice questions, with every answer explained.

The domain in numbers

52
Practice questions
3
Objectives
2
Easy
30
Medium
20
Hard

Where people go wrong

  • Giving Cloud Sync the traits of Connect Sync

    The wrong answers lend Microsoft Entra Cloud Sync features that belong to Microsoft Entra Connect Sync: a 30-minute scheduler, staging mode, device writeback and an external SQL Server. Cloud Sync runs on a cloud-managed scheduler that synchronises every two minutes, and its configuration is managed in the Microsoft Entra admin center. It gets high availability from additional active provisioning agents on different servers, which fail over automatically.

    See it in a sample question
  • Misreading how Conditional Access policies apply

    The wrong answers misjudge how Conditional Access policies apply. Options pick one policy by creation order or a numeric priority, have a report-only policy prompt or warn users, or let a Disabled per-user MFA state exempt a user. Conditional Access has no priority field: when one policy requires MFA and another requires a compliant device, the user must satisfy both. Report-only policies run Phase 1 and log the outcome, but Phase 2 enforcement does not affect the user. Policies are enforced after first-factor authentication, so per-user MFA state does not gate them.

    See it in a sample question
  • Configuring an authentication method in the wrong place

    The wrong answers move a method's settings to a control that does not own them: Conditional Access, the legacy MFA service settings or per-user MFA. In the Authentication methods policy, each method has an Enable and Target tab that can include a specific group, and each target can pick its own authentication mode, such as Any for a passwordless group and Push for All users. Microsoft Authenticator OTP has its own control, not the hardware or third-party OATH sections.

    See it in a sample question
  • Blurring pass-through authentication and hash sync

    The wrong answers merge pass-through authentication (PTA) with password hash synchronisation (PHS). One has PHS silently rescue a sign-in when every PTA agent is offline; others have PTA store hashes in the cloud, need a paid edition or purge existing hashes. PTA does not fail over to PHS automatically. It forwards validation to on-premises domain controllers, needs no paid Microsoft Entra ID edition and leaves synced hashes in place.

    See it in a sample question
  • Misplacing Password Protection components

    The wrong answers misplace Microsoft Entra Password Protection components and misstate what happens when no policy is available. Options install the DC Agent on the Microsoft Entra Connect server, have domain controllers contact the internet, require schema changes, or have the agent reject password changes or shut itself down. The DC Agent runs on every domain controller and the Proxy service on any domain-joined server, forwarding policy downloads so domain controllers never reach the internet. With no policy available, the agent accepts the password and logs an event.

    See it in a sample question

Objectives in this domain

Sample questions from this domain

Free sampleImplement and Manage Identity and Accessmedium

Your Cloud Sync deployment currently runs a single provisioning agent. The CIO wants seamless failover if the host server is rebooted for patching. Which configuration matches the documented Cloud Sync high-availability model?

  • APlace the existing single agent behind a hardware load balancer that distributes inbound SCIM traffic.
  • BInstall additional active provisioning agents on different servers so Cloud Sync fails over automatically. Correct
  • CPromote the existing agent host to a domain controller so failover can follow the AD FSMO topology.
  • DConfigure the active agent in staging mode and trigger Restart sync from the portal each patch window.
Cloud Sync HA is achieved by installing multiple active agents on separate servers. The Cloud Sync overview lists high availability through multi-agent support as a key capability: multiple active provisioning agents deployed across different servers provide automatic failover without configuration changes. Synchronisation continues seamlessly when individual agents become unavailable.

Why A is wrong: Cloud Sync agents maintain outbound persistent connections via Azure Service Bus; load balancers cannot mediate that pattern, and the documented HA model is multi-agent.

Why B is correct: Correct. The Cloud Sync overview lists high availability through multi-agent support as a key capability: multiple active provisioning agents deployed across different servers provide automatic failover without configuration changes.

Why C is wrong: Cloud Sync HA is independent of AD FSMO roles; making the agent host a domain controller is unnecessary and not the documented pattern.

Why D is wrong: Staging mode is a Connect Sync concept for disaster recovery, not Cloud Sync HA; manual Restart sync is operational triage, not HA.

Free sampleImplement and Manage Identity and Accessmedium

A policy is in report-only mode in the Conditional Access portal. Which behaviour matches the documented enforcement model for that policy?

  • ANeither phase runs and the policy is treated as if it were disabled.
  • BBoth phases run and the user sees prompts, but the access decision is recorded rather than enforced.
  • CThe policy is enforced but blocking outcomes are converted to a warning page in the browser.
  • DPhase 1 evaluation runs for the policy and logs results, but Phase 2 enforcement does not affect the user. Correct
Report-only policies execute Phase 1 (signals captured for impact analysis) but not Phase 2 (no actual block/grant action). The two-phase enforcement model says Phase 1 (collect session details) occurs for enabled policies and policies in report-only mode, while Phase 2 (enforcement) occurs only for enabled policies. That means a report-only policy logs what would have happened without changing the user's experience.

Why A is wrong: Report-only policies do run Phase 1; the difference from disabled is that they collect impact data for review.

Why B is wrong: Users are not prompted by report-only policies; only Phase 1 session details and the would-be outcome are recorded.

Why C is wrong: Report-only does not enforce or warn; it only logs the projected outcome for analysis.

Why D is correct: Correct. The two-phase enforcement model says Phase 1 (collect session details) occurs for enabled policies and policies in report-only mode, while Phase 2 (enforcement) occurs only for enabled policies.

Free sampleImplement and Manage Identity and Accesseasy

In the Microsoft Entra Authentication methods policy, an admin wants Microsoft Authenticator available only to a single department's security group while the rest of the tenant keeps the method disabled. Which target scope satisfies that requirement on the method's Enable and Target tab?

  • AEnable the method for All users and exclude every group except the department's group.
  • BLeave the method Off and configure a Conditional Access policy that targets only that group.
  • CEnable the method and target the department's security group as the included group. Correct
  • DEnable the method for All users and rely on per-user MFA settings to restrict access.
The Authentication methods policy supports per-group inclusion as a first-class targeting mode, which avoids the All-users-plus-excludes anti-pattern. Each method in the Authentication methods policy has an Enable and Target tab where the admin can include specific groups instead of All users. Targeting only the department's group enables Microsoft Authenticator for that group and leaves the method off for everyone else.

Why A is wrong: All users with broad excludes still flips the method on for every account not explicitly named, including new hires.

Why B is wrong: Conditional Access cannot enable an authentication method; the Authentication methods policy is the only place to turn it on.

Why C is correct: Correct. Each method in the Authentication methods policy has an Enable and Target tab where the admin can include specific groups instead of All users.

Why D is wrong: Per-user MFA service settings do not gate which users can use a method registered via the Authentication methods policy.

Free sampleImplement and Manage Identity and Accessmedium

Your tenant runs pass-through authentication with password hash synchronisation also enabled. If every Pass-through Authentication Agent goes offline at the same time, Microsoft Entra ID automatically authenticates managed-domain users against the stored hash so sign-ins continue. Is this statement correct?

  • AYes
  • BNo Correct
PTA has no automatic fallback to PHS - HA comes from multiple agents only. The PTA FAQ explicitly states pass-through authentication does NOT automatically fail over to password hash synchronisation. When all PTA agents are offline, managed-domain sign-ins fail even when password hashes are present in the cloud. High availability is achieved by running multiple PTA agents.

Why A is wrong: Enabling PHS alongside PTA only stages hashes for cases such as Identity Protection leaked-credentials detection or a planned manual switchover; it never silently rescues an inflight PTA sign-in.

Why B is correct: Correct. The PTA FAQ explicitly states pass-through authentication does NOT automatically fail over to password hash synchronisation.

Free sampleImplement and Manage Identity and Accessmedium

An organization is planning Microsoft Entra Password Protection for an on-premises Active Directory forest. Which architectural statement matches Microsoft's published design?

  • AThe DC Agent runs on the Microsoft Entra Connect server and pushes policy directly to domain controllers.
  • BDomain controllers must be granted outbound internet access so they can poll Microsoft Entra ID directly.
  • CThe DC Agent and Proxy service both require AD DS schema extensions to register their service connection points.
  • DThe DC Agent runs on every domain controller and the Proxy service runs on any domain-joined server in the forest. Correct
DC Agent on every DC plus Proxy service on any domain-joined server is the published two-tier architecture for on-premises Password Protection. The on-premises architecture documents two components: the Microsoft Entra Password Protection Proxy service running on any domain-joined machine in the forest, and the DC Agent service running on every domain controller. The Proxy forwards policy download requests; the DC Agent applies policy at password change time.

Why A is wrong: The DC Agent is installed on the domain controllers themselves, not on the Microsoft Entra Connect server.

Why B is wrong: A documented design principle is that domain controllers never have to communicate directly with the internet; the Proxy service forwards policy downloads on their behalf.

Why C is wrong: Design principles state no AD DS schema changes are required; the software reuses existing container and serviceConnectionPoint objects.

Why D is correct: Correct. The on-premises architecture documents two components: the Microsoft Entra Password Protection Proxy service running on any domain-joined machine in the forest, and the DC Agent service running on every domain controller.

Other domains in this exam

See also the MS-102 cert hub, the study guide, and the cheat sheet.

Examworthy is not affiliated with or endorsed by Microsoft. Original, blueprint-aligned practice material only.