How to configure Windows Server Failover Clustering for shared storage

Create a new failover cluster to host your SQL database or file services with redundancy. These steps target Windows Server 2022 and Windows Server 2025 with Hyper-V or iSCSI storage. Follow the procedure to configure the cluster role and ensure the shared disk is accessible to all nodes.

Prerequisites

• Windows Server 2022 or 2025 installed on at least two physical nodes.
• Shared Storage (SAN, iSCSI, or SMB 3.0) connected to all nodes with identical LUNs.
• Static IP addresses configured on the network for the cluster network.
• Administrative privileges to join the cluster and create roles.
• Quorum witness (File Share or Cloud Witness) configured for odd-numbered node clusters.

Step 1: Open the Failover Clustering Manager

Launch the Failover Clustering Management console to begin the wizard. You must run this as an administrator. Open Server Manager and click the Tools menu, then select Failover Clustering.

Step 2: Run the New Cluster Wizard

Start the setup process by selecting the New Cluster option from the dashboard. Click New Cluster and then choose Create a new cluster. In the wizard, click Next until you reach the Cluster Name and Cluster Network page. Enter a unique name for the cluster, such as SQLCluster01, and specify the static IP address for the cluster network.

Step 3: Select the Cluster Nodes

Select the server names that will participate in the cluster. Ensure all nodes are listed in the Available pane before moving forward. Click the Select button to add the first node, then repeat for the second node. Verify that the list shows your server names, then click Next.

Step 4: Validate the Cluster Configuration

Run the cluster validation wizard to check for hardware or configuration issues. Click Validate Configuration and then select Validate a Configuration for a New Cluster. On the next screen, check the boxes for Cluster, Network, Storage, and Server. Click Next to start the deep scan.

Cluster Validation Wizard - Progress
Validating Cluster Configuration...
Validating Network Configuration...
Validating Storage Configuration...
Validation Complete. No errors found.

Step 5: Create the Cluster

Review the validation results and click Repair if any errors appear. If the scan passes, click Create Cluster to finalize the setup. The wizard will create the cluster object and register the nodes. Click Finish when the process completes. You will see a success message indicating the cluster is online.

Step 6: Add the Shared Storage to the Cluster

Open the cluster manager and expand the Cluster node in the left pane. Right-click Cluster and select Add Disk. Select the shared LUN from the list of available disks. Click Add to include the storage in the cluster resource pool. The disk will appear as a Cluster Shared Volume (CSV) or a standard cluster disk.

Step 7: Promote the Nodes

Convert the nodes from witness nodes to active cluster nodes. Right-click the Cluster name in the left pane and select Properties. Click the Quorum tab to ensure the witness is set correctly. Then, right-click each node name and select Promote to join the cluster fully. Wait for the status to change to Online.

Verify the installation

Open PowerShell as Administrator and run the following command to confirm the cluster status. The output should show the cluster name and the node states as Online.

Get-Cluster -Name SQLCluster01 | Select-Object Name, State
Get-ClusterNode -Cluster SQLCluster01 | Select-Object Name, State

You will see output similar to this:

Name                 State
----                 -----
SQLCluster01         Online
Node01               Online
Node02               Online

Troubleshooting

If the cluster fails to start, check the Event Viewer for Failover Clustering errors. Common issues include network timeouts or storage access failures. Use the Cluster Validation Wizard again to isolate specific hardware faults. Ensure the shared storage is not in a Disconnected state in the storage management console. If the quorum is lost, verify the witness path is accessible and the cluster network has a valid route. Restart the Cluster Service if the nodes do not respond to heartbeats.