rotating disk upgrade on Windows-Server 2022 Storage Spaces Direct 2-Node Cluster

Question

Hello,

I have a question regarding the replacement of drives with larger capacity for Storage Spaces Direct.

We have a 2-Node Hyper-V Failover Cluster that uses Storage Spaces Direct.
Each server has 1x 8 slot drive cage, with 6x 960GB SSDs inside.
We use nested two-way mirror for our volumes for resilience mode.

The pool capacity is runnung low and we need to expand the pool beyond what two extra 960GB SSDs could provide. An third node is no option.

The plan is to replace all 960GB with 3,84TB SSDs.
It is known that expanding a S2D pool with the 3,84TB SSDs leave unusable capacity.
So the idee is to fill the 8-slot cage with 2x 3,84TB SSDs, expand the pool, decomission 2x 960GB and repeat the cycle until all 960GB SSDs have been replaced by the larger SSDs.

The Question is will S2D be able to use the 3,84 SSDs with there full usable capacity after the swap from 960GB and the pool will now be overall expanded?
Did some already try this method?

Otherwise we have to rebuild the pool from scratch, and restore the VMs from our backups

Answer 1

Hello Willenbrecht, Jannik,

Thank you for your question. Your approach to gradually replacing 960GB SSDs with 3.84TB SSDs in a 2-node Storage Spaces Direct (S2D) cluster is supported, and many organizations have successfully used similar strategies. However, drive symmetry is a key factor in how S2D allocates usable capacity.

In a 2-node cluster, S2D expects identical drive configurations across nodes. When mixing drive sizes, some capacity may remain unused until symmetry is restored. This means that during the transition phase, the 3.84TB drives may not be fully utilized until all 960GB drives are replaced on both nodes.

To maximize usable capacity:

• Ensure each node has matching drive counts and sizes.
• Use Get-StoragePool and Get-PhysicalDisk to monitor capacity and allocation.
• Consider running Optimize-StoragePool after each replacement cycle.
• Avoid mixing cache and capacity tiers unless explicitly configured.

If this guidance helps, please click “Accept Answer” so others can benefit too.

Thank you so much!!

Best regards,

QQ.

Answer 2

When replacing drives in a Storage Spaces Direct (S2D) setup, especially when upgrading to larger capacity drives, it is important to follow the proper procedure to ensure that the new drives are utilized effectively. In your case, replacing the 960GB SSDs with 3.84TB SSDs can be done in the manner you described, and S2D should be able to utilize the full capacity of the new drives after they are added to the pool.

Here’s a suggested approach:

1. Install the new 3.84TB SSDs in the empty slots of the drive cage.
2. Expand the storage pool to include the new drives. S2D should recognize the new capacity and allow you to use it.
3. Decommission the 960GB SSDs one at a time, ensuring that data is migrated appropriately to the new drives before removing them from the pool.
4. Repeat the process until all 960GB SSDs have been replaced with the 3.84TB SSDs.

This method allows you to gradually upgrade your storage without needing to rebuild the entire pool from scratch. However, keep in mind that while S2D can use the full capacity of the new drives, the overall usable capacity may still be affected by the resiliency settings (nested two-way mirror in your case). Therefore, it is advisable to monitor the usable capacity after each upgrade to ensure it meets your needs.

If you encounter any issues or limitations during this process, consulting the relevant documentation or support resources may provide additional guidance specific to your setup.

---

References:

• Nested resiliency for Storage Spaces Direct
• Plan volumes on Azure Local and Windows Server clusters