AZ-801 is a retired Microsoft exam. Microsoft retired both AZ-800 and AZ-801 on September 30, 2026. An article that invites candidates to book AZ-801 today would therefore give the wrong advice. This page preserves the technical purpose of the historical exam—Windows Server security, high availability, disaster recovery, migration, and monitoring—while explaining the current route to Windows Server administrator certification. Candidates seeking an available exam should begin with Microsoft AZ-802: Administering Windows Server, not a retired AZ-801 booking guide.
The old Microsoft AZ-801 page remains useful when interpreting an earlier transcript, mapping legacy training, or reusing hands-on knowledge. It should not be confused with an active qualification. Microsoft’s current credential is Microsoft Certified: Windows Server Administrator Associate, associated with AZ-802. This is a qualification-route change, not a claim that prior exam passes or existing credentials have been converted automatically. Check your own transcript and the credential’s official eligibility information if you completed part of the old pathway.
What AZ-801 originally assessed—and why those skills still matter
AZ-801, Configuring Windows Server Hybrid Advanced Services, was the advanced companion to AZ-800 in the former Windows Server Hybrid Administrator Associate route. Its published areas included securing Windows Server in on-premises and hybrid settings, implementing high availability and disaster recovery, migrating workloads, and monitoring and troubleshooting. Those are enduring operational problems even though the exam code has retired. The useful content to preserve is the method for solving them, not unsupported promises about historic scores or question formats.
The critical distinction is between technical reuse and certification continuity. A failover-cluster troubleshooting lab remains valuable when preparing for a modern Windows Server administration role. A course specifically advertising a current AZ-801 registration, an old passing score as a current rule, or an unverified mapping of every AZ-801 question to AZ-802 does not. Microsoft maintains an official retired exams record that candidates should use rather than the availability status on an old third-party study page.
How the current AZ-802 route differs
AZ-802 combines a broad Windows Server administration remit into one current exam. Microsoft’s published skills cover Active Directory Domain Services, hybrid workloads, virtual machines, networking, storage and file services, securing Windows Server infrastructure, and monitoring and troubleshooting. In contrast, the previous pathway split core hybrid administration and advanced services across AZ-800 and AZ-801. A candidate who previously organized study into two separate sets of notes should now map topics to AZ-802’s current objective guide instead of assuming the old split still defines the exam.
For example, an AZ-801 study notebook may contain detailed disaster-recovery exercises but relatively little coverage of managing AD DS, Hyper-V, or core network services because those topics were taught for AZ-800. That notebook is not automatically a complete AZ-802 plan. Conversely, do not discard the disaster-recovery labs: they can still demonstrate administrative reasoning where current objectives and workplace needs overlap. The exam owner decides the new tested boundaries; your old lab archive is supporting material, not the authority.
Start with the full current AZ-802 study guide, then mark old topics directly covered, supporting background, or not named in current objectives. That last group need not be deleted from professional development, but it should not displace high-priority current objectives. Dates matter: this page describes the transition as of October 17, 2026, and future issuer changes can alter the recommended learning path.
Hybrid identity and security: understand the trust boundary
Windows Server remains a security-sensitive identity and workload platform in many mixed estates. Know where AD DS domain controllers live, which authentication and authorization components depend on them, how privileged access is assigned, and which on-premises services trust cloud identities or management systems. Entra ID and AD DS serve related but different identity functions; hybrid synchronization does not make their control planes interchangeable.
Start security troubleshooting by identifying the system and principal involved. An administrator who cannot manage a server could lack a role assignment, a network path, an applicable policy, or working authentication. Avoid granting broad local administrator privileges simply because a management operation fails. Check effective group membership, delegated rights, event logs, and the management endpoint first. Apply least privilege and dedicated administrative identities to reduce the impact of compromised credentials.
Protect remote administration separately from workload access. Windows Admin Center, PowerShell Remoting, Remote Desktop, and Azure Arc offer useful management options but require trusted access paths, endpoint restrictions, credential protection, and logging. The steps in Windows Admin Center setup are useful only after the organization has decided who may reach the management plane and how access is reviewed. “Reachable from anywhere” is not an acceptable default design target.
High availability is not disaster recovery
Failover clustering can keep eligible services available when a node fails, but it does not by itself protect against every storage, site, or security incident. Separate workload availability, data recoverability, and disaster scope. A cluster that survives a host failure may still fail when its shared storage or dependent identity service becomes unavailable. A replicated copy can be recent but not usable if corruption or ransomware has propagated with it.
Define recovery time objective (RTO), recovery point objective (RPO), and the failure scope before selecting mechanisms. A small service might accept several hours of restoration using tested backups. A critical database might need local high availability and independently protected recovery copies. Windows Server Failover Clustering, Storage Replica, backup products, and Azure recovery tooling solve different portions of this design; choose only features supported by the actual workload, topology, licensing, and operating-system version.
Test the complete restoration sequence rather than relying on green backup-job indicators. Restore identity or name-resolution prerequisites where required, recover storage and service data, validate permissions, and demonstrate application access from a real client. Write down the failure you tested and the recovery you observed. Without that evidence, even a highly redundant architecture may have an unknown business recovery capability.
Migrating Windows Server workloads without moving the failure
A hybrid migration should start with an inventory of services, dependencies, data flows, certificates, identities, scheduled tasks, file shares, and licensing requirements. “Move the server” is rarely the full task. A line-of-business application may depend on a service account, a legacy SMB path, a DNS alias, or a firewall rule not documented in its virtual-machine inventory. Discover and test those dependencies before committing to a cutover window.
For a file-service migration, measure source volume and change rate, verify NTFS and share permissions, plan namespace or DNS cutover, rehearse incremental synchronization, and define who can roll back. For workload migrations involving Azure, document the responsibility split for operating-system patching, identities, network connectivity, backups, and monitoring. Migration success is not just a booted virtual machine; it includes access, performance, recoverability, and the ability to operate the new environment.
Azure Arc can provide an inventory and management connection for eligible non-Azure servers, but it is not a substitute for networking, identity design, application compatibility testing, or backup validation. The correct architecture may leave a workload on-premises while extending governance and monitoring to it. A hybrid plan should be chosen from constraints and measurable outcomes rather than assuming that every service must be migrated immediately.
Monitor and troubleshoot from the affected service outward
Choose monitoring based on how the business experiences failure. Host CPU and disk metrics help, but they do not establish that a user can authenticate, reach the application, open a file share, or complete a transaction. Pair operating-system health with service-level checks, event correlation, capacity trends, and alerts for security and backup problems. When using Azure Monitor or another platform, verify data collection, permissions, alert rules, and retention rather than assuming connected servers automatically provide complete visibility.
For a hybrid incident, first identify what still works. If an administrator can reach a server by IP but a client cannot reach it by hostname, investigate DNS before rebuilding a cluster. If on-prem users authenticate but remote users fail, inspect the remote access, identity, name-resolution, and network paths as distinct failure domains. If only recently migrated services fail, compare endpoint settings, access policies, dependencies, and routing with the known pre-migration state.
PowerShell is valuable for obtaining reproducible evidence and checking configuration across multiple systems, but scripts should use narrowly scoped credentials and report changes explicitly. The Active Directory PowerShell commands can support identity-administration tasks; they are not a reason to bypass change control or audit requirements. The useful troubleshooting habit is collecting a before/after record so operators can distinguish correlation from a proven repair.
How to reuse AZ-801 material responsibly for AZ-802
If you already invested in AZ-801 training, sort the material by real capability rather than preserving its old chapter order. Keep current labs for hybrid security, cluster recovery, server migration, and operational telemetry where they apply. Update deprecated product naming and feature behavior; add missing AZ-802 topics such as modern AD DS administration, workload management, virtualization, networking, and file services. Then compare each topic to Microsoft’s active exam guide and remove unsupported claims about which questions will appear.
A strong lab sequence begins with a small Windows Server environment, configures an identity and network baseline, applies constrained management access, introduces a recoverable failure, and uses event logs and monitoring to find and prove the cause. A second sequence can exercise a file-share migration and rollback. The goal is not to create theatrical scenarios; it is to demonstrate exactly which dependency changed and what evidence shows that availability, security, or recoverability improved.
For readers choosing among Microsoft certifications, the action today is straightforward: review the current AZ-802 exam requirements and the current Windows Server Administrator Associate credential, then decide which of your older AZ-800/AZ-801 materials are still technically useful. Do not purchase an AZ-801 exam booking on the assumption that its historical title signals an active test. The exam changed; the underlying discipline of secure, observable, recoverable Windows Server operations still matters.