How to Update Windows Server Failover Cluster Nodes Safely

Applies to: Windows Server 2025 / 2022 / 2019 / 2016 / 2012 R2 / 2012 / 2008 R2 / 2008 / 2003

Updating a Windows Server Failover Cluster should be done one node at a time so clustered workloads remain available while each node is patched and restarted. The basic method is simple: drain roles from a node, pause it, install updates, restart it, validate it, and then return it to the cluster.

The original version of this article covered service packs on Windows Server 2003, 2008, and 2008 R2. Current Windows Server versions use the same rolling-maintenance idea, but modern Failover Clustering also provides Cluster-Aware Updating (CAU) to automate much of the process.

In This Article

• 1. Before You Start
• 2. Manual Rolling Update Process
• 3. Cluster-Aware Updating
• 4. Validate the Cluster After Updating
• 5. Important Notes
• 6. Key Takeaways

1. Before You Start

Before installing updates, confirm that the cluster is healthy and that workloads can move between nodes without errors.

• Confirm all expected cluster nodes are online.
• Check that clustered roles and resources are healthy.
• Verify storage and cluster networks before maintenance.
• Confirm that the remaining nodes have enough capacity to host the workloads from the node being patched.
• Use the same approved update source and update baseline across all cluster nodes.

If the cluster is already unhealthy, fix that condition before starting patching. Updating should not be used as a way to clear an existing cluster problem.

2. Manual Rolling Update Process

If you patch the cluster manually, work on only one node at a time.

1. In Failover Cluster Manager, select the first node.
2. Select Pause > Drain Roles. This moves clustered workloads away from the node before maintenance.
3. Confirm that the node no longer owns the workloads that must remain available.
4. Install the approved Windows updates, drivers, firmware, or other maintenance items required for that node.
5. Restart the node if required.
6. After the node comes back, verify that the Cluster Service is running and that the node has rejoined the cluster correctly.
7. Select Resume to return the node to normal cluster participation.
8. Validate the node before moving to the next cluster member.
9. Repeat the same process for each remaining node.

PowerShell provides the same drain and resume operations through Suspend-ClusterNode -Drain and Resume-ClusterNode. Microsoft also uses the pause-and-drain method in its current failover cluster maintenance and rolling-upgrade guidance.

3. Cluster-Aware Updating

Cluster-Aware Updating (CAU) automates the process of updating clustered servers while maintaining workload availability. During an updating run, CAU coordinates the update process across nodes instead of requiring an administrator to perform each drain, update, restart, and resume step manually.

Microsoft currently documents CAU for Windows Server 2016, 2019, 2022, and 2025. CAU can run in remote-updating mode or as a clustered role in self-updating mode.

CAU is usually the cleaner option for regular Windows patching when the cluster and update process meet its requirements. Manual rolling maintenance is still useful when the update requires vendor-specific steps, firmware work, application coordination, or other actions that CAU does not handle.

4. Validate the Cluster After Updating

Do not consider the maintenance complete only because every node rebooted successfully. Validate the cluster after the final node is updated.

• Confirm all nodes show the expected state.
• Confirm clustered roles and resources are online.
• Check that workloads can move between nodes.
• Review Failover Clustering events for new errors or warnings.
• Validate application-specific health where applicable.
• If workloads have preferred owners, move them back only after the cluster is stable.

5. Important Notes

A normal monthly patching cycle is different from a cluster operating system rolling upgrade. Moving from one Windows Server release to another has additional version-mixing, upgrade-order, and cluster functional-level requirements. Follow the dedicated Microsoft rolling-upgrade procedure for that scenario.

Also check the application or workload documentation before patching. SQL Server, Hyper-V, Exchange-related infrastructure, third-party clustered applications, and storage products can have their own maintenance requirements in addition to the Windows Failover Cluster steps.

6. Key Takeaways

• Patch one cluster node at a time.
• Drain clustered roles before taking a node into maintenance.
• Validate each node before continuing to the next one.
• Use Cluster-Aware Updating when it fits the workload and update process.
• Treat operating system version upgrades as a separate rolling-upgrade procedure.

References

• Microsoft Learn – Cluster-Aware Updating overview
• Microsoft Learn – Cluster-Aware Updating requirements and best practices
• Microsoft Learn – Cluster OS rolling upgrade