Share via

Hyper-V - Creating Snapshots/Checkpoints on Clustered VMs

Guilherme Couto 0 Reputation points
2025-07-15T04:07:30.0233333+00:00

Hi all,

I’m running a Windows Server 2022 Datacenter environment with a Hyper-V failover cluster. All my VMs are highly available clustered roles on shared storage (CSV volumes). These VMs are production-critical workloads, including SQL Server databases.

We previously used Nutanix, where snapshot scheduling was a built-in feature, and we used it heavily for quick rollback scenarios. We’ve recently migrated this environment to Hyper-V and would like to continue leveraging frequent snapshots for some VMs.

Here’s what I’m planning:

  • I will use PowerShell to automate snapshot creation
  • Some VMs will have snapshots taken every 15 minutes (keeping only 1 copy), hourly (keep 6 copies only), and daily (1 copy daily)

I am fully aware of the storage implications

I will always ensure Production Checkpoints are used (not Standard Checkpoints)

We do have a proper backup solution in place that performs regular backups of all VMs, so this isn’t intended to replace backups. The reason for this approach is that for some client VMs we often need to perform quick restores, and restoring from backup takes significantly longer than simply reverting to a snapshot/checkpoint.

However, since migrating to Hyper-V, I’ve been reading that frequent checkpoints on clustered production VMs could create problems, and I want to be very cautious not to impact cluster health, stability, or performance.

My questions are:

  1. Is this frequent snapshot approach safe and supported in a clustered Hyper-V 2022 production environment, including for SQL workloads?
  2. Have others used this kind of snapshot scheduling successfully? Are there real-world examples where this has worked or caused issues?
  3. What risks (aside from storage consumption) should I be aware of, particularly around failover behavior or cluster integrity?

Any advice or real-life feedback would be appreciated. I want to make sure I’m not introducing potential problems by following this strategy.

Thanks in advance!

Windows for business | Windows Server | Storage high availability | Clustering and high availability
0 comments

1 answer

Sort by: Most helpful
Most helpful Newest Oldest
  1. Henry Mai 2,375 Reputation points Independent Advisor
    2025-08-04T01:39:14.4666667+00:00

    Hello, I am Henry and I want to share my insight about your concern.

    Based on the scenario you've described, here is my analysis and recommendation. I see that the primary risk is not storage space, but severe storage I/O performance degradation. When a checkpoint is deleted, the merge process is extremely I/O-intensive and can dramatically slow down all other VMs sharing that same storage (CSV).

    1. Is a frequent snapshot strategy safe and supported?
    • Not recommended or supported for critical production applications, especially SQL Server.
    • Reason: Hyper-V checkpoints (relying on .avhdx files) create layers of virtual disks. This leads to a significant degradation in I/O performance, particularly as the snapshot chain grows.
    • The merge operation, which occurs when a checkpoint is deleted, is also resource-intensive and impacts VM performance.
    1. Have others used this successfully?
    • Yes, but typically in development, testing, or non-critical workloads where the need for a quick rollback outweighs performance concerns.
    1. What risks should I be aware of?
    • Failover Behavior: A complex snapshot chain can complicate failover. A corrupted checkpoint file could lead to a failed failover and potential downtime.
    • Cluster Integrity: The constant creation and deletion of checkpoints place a high strain on the I/O of the Cluster Shared Volume (CSV) and cluster management services, which could lead to instability.
    • Data Loss/Corruption: Rolling back to a snapshot means losing all data changes since that point, which can cause data integrity issues for transactional applications like SQL Server.

    Final Advice:

    • Use Checkpoints sparingly: Reserve checkpoints for specific, short-term, controlled scenarios, such as before applying a major patch. Always delete the checkpoint as soon as the change is validated.
    • Thoroughly test: If you decide to proceed with this strategy, you must test it rigorously in a non-production environment with an identical workload to understand its real-world impact before deploying it in production.

    I hope this information is helpful.

    0 comments

Your answer

Answers can be marked as Accepted Answers by the question author, which helps users to know the answer solved the author's problem.