Hey Checkyourlogs Fans,

Upgrading a Hyper-V cluster running Storage Spaces Direct (S2D) from Windows Server 2019 to Windows Server 2022 or 2025 can be complex. We follow a best-practice approach that minimizes downtime and ensures a seamless upgrade. This process involves leveraging Veeam Backup & Replication (VBR) to replicate VMs to a temporary host, freeing up cluster nodes for a complete rebuild instead of performing an in-place upgrade.

A screenshot of a computer Description automatically generated

Why Avoid In-Place Upgrades?

While in-place upgrades are technically possible, they can lead to:

  1. ReFS Compatibility Issues: In-place upgrades may not update the ReFS version to the latest supported by the newer OS, potentially limiting performance and features.
  2. Lingering Legacy Configurations: Older settings and drivers may carry over, potentially causing instability.
  3. Missed Opportunities for Optimization: A fresh rebuild allows you to reconfigure and optimize your cluster for new workloads.

Veeam VBR: A Critical Tool for Seamless Migration

Using Veeam VBR, we replicate all VMs to an alternate host. This approach offers several advantages:

  • Minimal Downtime: Veeam VBR’s planned failover and fallback capabilities ensure the VMs remain operational on temporary hardware during the upgrade.
  • Annual DR Test: This process doubles as a live disaster recovery test, validating your failover procedures.
  • Streamlined Failback: Once the new cluster is ready, Veeam’s resyncing capabilities ensure that all VM metadata and replication jobs are seamlessly re-established.

Step-by-Step Process

1. Preparation

  • Verify that Veeam VBR is updated to the latest version.
  • Ensure the temporary host has sufficient resources to accommodate the replicated VMs.
  • Validate the backup health of all VMs before starting the replication process.

2. Replication to Temporary Host

  • Configure replication jobs in Veeam VBR:
    1. Select the source cluster as the replication source.
    2. Choose the temporary host as the replication target.
    3. Schedule replication jobs to run frequently to minimize data loss during failover.
  • Initiate planned failover to the temporary host.

3. Rebuild the Cluster

  • Wipe and reinitialize the existing cluster nodes.
  • Install Windows Server 2022/2025.
  • Configure Storage Spaces Direct and recreate the cluster.
  • Verify the cluster’s health and performance.
  1. Failback to the New Cluster

A screenshot of a computer Description automatically generated

  • Once the new cluster is ready, initiate planned failback in Veeam VBR:
    1. Resynchronize VMs from the temporary host to the new cluster.
    2. Verify VM integrity and performance after failback.
  • Update Veeam replication jobs to reflect the new cluster as the target.

5. Post-Migration Tasks

  • Update Hyper-V Integration Services on all VMs to match the new OS version.

A screenshot of a computer Description automatically generated

  • Test backups and replication jobs to ensure they are functioning as expected.
  • Validate the latest ReFS version is in use and optimized for performance.
  • Document the new cluster configuration and any changes.

Benefits of This Approach

  1. Minimized Downtime: The VMs remain operational on the temporary host during the cluster rebuild.
  2. Validated DR Plan: Live failover and failback testing ensure disaster recovery readiness.
  3. Optimal Configuration: A clean build allows the implementation of best practices and leveraging new OS features.
  4. Seamless Job Resync: Veeam automatically synchronizes metadata, ensuring smooth operations post-migration.

Conclusion

Upgrading a Hyper-V cluster with S2D is a critical task that requires careful planning and execution. By leveraging Veeam VBR’s replication capabilities, you can minimize downtime, ensure a robust disaster recovery process, and rebuild your cluster with the latest OS and features. This process ensures a smoother upgrade and sets the stage for improved performance and reliability.

Thanks,

Dave