AZ-802

Free AZ-802 practice questions

10 questions from our AZ-802 practice tests, each with the correct answer and a written explanation. Read them with the answers shown, or switch to quiz mode and answer first.

Back to the AZ-802 practice exams

Question 1 of 10, Deploy and manage AD DS

A trust exists between two forests. Security requires that users from the trusted forest can authenticate only to two specific file servers in your forest and nowhere else. What should you configure?

  1. AName suffix routing on the trust so that only the file server host names are routed to the trusted forest
  2. BSID filtering on the trust, then remove the trusted users from the Authenticated Users group in your forest
  3. CForest-wide authentication on the trust, then add the trusted users to a domain local group on each file server
  4. DSelective authentication on the trust, plus Allowed to Authenticate on the two computer objects Correct answer

Answer and explanation: Selective authentication blocks authentication from the trusted forest by default until the Allowed to Authenticate right is granted on specific computer objects. Forest-wide authentication lets the users authenticate to any computer, and group membership on the file servers does not restrict authentication elsewhere. SID filtering removes foreign SIDs from tokens, and name suffix routing controls which UPN suffixes are routed, so neither limits targets.

Question 2 of 10, Deploy and manage AD DS

A branch RODC must be installed by a local technician who is not a domain admin. Domain admins must create the computer account and specify the delegated installer in advance. What should you do?

  1. APre-stage the RODC account and name the technician as its delegated administrator Correct answer
  2. BGrant the technician Full Control on the Domain Controllers OU and allow them to run dcpromo with an answer file
  3. CAdd the technician to the Enterprise Read-only Domain Controllers group and let them run Install-ADDSDomainController
  4. DAdd the technician to the Allowed RODC Password Replication Group so they can complete promotion

Answer and explanation: A staged RODC installation lets a domain admin pre-create the RODC account and name a delegated administrator who can then attach the server during promotion without domain admin rights. The Enterprise Read-only Domain Controllers group is for RODC computer accounts, not installers. Full Control on the Domain Controllers OU is excessive and dcpromo answer files are obsolete, and the password replication group only controls which credentials are cached.

Question 3 of 10, Manage Windows Server instances and workloads in a hybrid environment

An administrator connects to Server1 with Enter-PSSession and then runs a command that reads a file share on Server2. The command fails with access denied, although the administrator has rights to the share. You need to fix this with the MOST secure approach. What should you do?

  1. AEnable CredSSP authentication on both servers and connect with -Authentication CredSSP
  2. BAdd the administrator account to the local Administrators group on Server2 and reconnect
  3. CConfigure resource-based Kerberos constrained delegation on Server2 that trusts Server1 Correct answer
  4. DConfigure unconstrained Kerberos delegation for the Server1 computer object in Active Directory Users and Computers

Answer and explanation: This is the PowerShell second-hop problem, and resource-based Kerberos constrained delegation (set with PrincipalsAllowedToDelegateToAccount on Server2) lets Server1 delegate only to Server2 without passing credentials. CredSSP works but caches reusable credentials on Server1, and unconstrained delegation lets Server1 impersonate users to any service, both less secure. Adding local admin rights does not solve the missing credential on the second hop.

Question 4 of 10, Manage virtual machines

A domain controller runs as a Hyper-V VM and its time drifts whenever the host's clock is changed. The domain hierarchy should be the only time source for the DC. What should you do?

  1. ADisable Heartbeat integration services on the VM so the host stops sending time updates
  2. BDisable the VMICTimeProvider in the guest so the DC follows the domain hierarchy Correct answer
  3. CEnable the Data Exchange integration service so that the DC publishes its time to the host
  4. DConfigure the host to sync time from the DC by running w32tm /config /syncfromflags:domhier

Answer and explanation: Domain controllers should not take routine time from the Hyper-V host, so the VMIC time provider inside the guest is disabled (or the time integration service is turned off), letting the DC follow the domain hierarchy. Heartbeat does not control time. Making the host sync from the DC can help but does not stop the guest receiving host time, and Data Exchange shares metadata rather than time.

Question 5 of 10, Implement and manage an on-premises and hybrid networking infrastructure

You need to make sure that DNS records in an AD DS-integrated zone replicate to every domain controller running DNS in the forest, including those in other domains. Which replication scope should you choose?

  1. ATo all DNS servers running on domain controllers in this forest Correct answer
  2. BTo all domain controllers in this domain for Windows 2000 compatibility
  3. CTo all DNS servers listed on the zone's Name Servers tab
  4. DTo all DNS servers running on domain controllers in this domain

Answer and explanation: The forest-wide scope stores the zone in the ForestDnsZones application partition, which replicates to every DC that runs DNS across all domains. The domain scope stops at the domain boundary, the Windows 2000 compatible scope stores the zone in the domain partition for the current domain only, and the Name Servers tab governs zone transfers rather than AD replication.

Question 6 of 10, Manage storage and file services

A two-node failover cluster uses local NVMe drives in each node. You need to create highly available storage that pools the local drives across both nodes, without a SAN. What should you implement?

  1. AStorage Spaces on each node with a parity virtual disk shared by iSCSI
  2. BA Scale-Out File Server that uses the local NVMe drives as cluster shared disks
  3. CStorage Spaces Direct with a two-way mirror volume and a cluster witness Correct answer
  4. DStorage Replica in synchronous mode between the two nodes' local volumes

Answer and explanation: Storage Spaces Direct pools local drives across cluster nodes into a software-defined storage pool, and a two-node deployment uses two-way mirroring with a witness for quorum. Standalone Storage Spaces is per server, and sharing it over iSCSI creates a single point of failure. Storage Replica replicates volumes but does not create a pooled cluster volume, and SOFS on its own cannot use local disks as shared storage without S2D.

Question 7 of 10, Manage storage and file services. Choose 2.

You have two file servers in different sites hosting the same data. Users must access the data by using one UNC path and be directed to the server in their own site, and changes must be kept in sync. Which two components should you configure? (Choose two.)

  1. AA DFS Replication group for the shared folders Correct answer
  2. BA DFS namespace with folder targets on both servers Correct answer
  3. CStorage Replica asynchronous replication between the volumes
  4. DWork Folders sync shares on both servers
  5. EA Scale-Out File Server role on a failover cluster

Answer and explanation: A DFS namespace gives one UNC path and uses site awareness to refer clients to the closest folder target, while DFS Replication keeps the folder contents in sync between servers. Storage Replica replicates a volume to a destination that is not accessible while it is a replica. A Scale-Out File Server requires a cluster with shared storage, and Work Folders syncs user data to devices rather than between servers.

Question 8 of 10, Secure Windows Server infrastructure

After enabling an inbound Windows Firewall rule, you must require that SMB traffic between two management servers is authenticated and encrypted by using computer certificates, while traffic to other hosts remains unchanged. What should you create?

  1. AA Windows Firewall profile setting that blocks unauthenticated inbound traffic
  2. BA Group Policy that enables SMB signing and SMB encryption on all servers
  3. CA connection security rule of the Server-to-server type with IPsec Correct answer
  4. DAn outbound firewall rule with the Allow the connection if it is secure action

Answer and explanation: A server-to-server connection security rule defines IPsec authentication (such as computer certificates) and encryption between specific endpoints. The allow-if-secure action only works when a connection security rule exists to establish IPsec. Enabling SMB signing and encryption everywhere affects all hosts and does not use certificate authentication, and a profile-wide block is far broader than required.

Question 9 of 10, Monitor and troubleshoot Windows Server environments

An Arc-enabled server shows Disconnected in the Azure portal although it is running. What should you run on the server to diagnose the problem?

  1. ATest-NetConnection 168.63.129.16 -Port 80
  2. BGet-AzConnectedMachine -Name Server01
  3. Cazcmagent check and azcmagent show Correct answer
  4. DRepair-AzVMGuestAgent -VMName Server01

Answer and explanation: The azcmagent check command tests connectivity to the required Azure endpoints and azcmagent show reports the agent status and last heartbeat. Get-AzConnectedMachine reads the Azure resource state but cannot diagnose local connectivity. 168.63.129.16 is the Azure host platform address that on-premises machines do not use, and there is no Repair-AzVMGuestAgent cmdlet for Arc machines.

Question 10 of 10, Monitor and troubleshoot Windows Server environments

A domain user was deleted by mistake an hour ago. The AD Recycle Bin was enabled last year. You need to restore the user with all its group memberships. What should you do?

  1. ARestore the system state of a domain controller from last night's backup and restart it
  2. BRun Get-ADObject with -IncludeDeletedObjects and pipe the object to Restore-ADObject Correct answer
  3. CRestart a domain controller in DSRM and perform an authoritative restore of the user
  4. DRecreate the user with the same name and add it back to the original groups

Answer and explanation: With the Recycle Bin enabled, deleted objects keep all link-valued attributes such as group memberships, and Restore-ADObject (or the Administrative Center) brings them back online. An authoritative restore works but is far more disruptive and unnecessary. Recreating the user produces a new SID, and a non-authoritative system state restore would be overwritten by replication.