S2D tier optimization fails with 0x80070057 after node eviction

Question

Hello Microsoft Support team.

Running Optimize-Volume -TierOptimize on our S2D Cluster Shared Volume fails with 0x80070057 (Invalid Parameter).

Our investigation indicates the heat map used for NVMe/HDD tiering became corrupted after a node eviction.

What is the recommended procedure to clear the existing heat map data and rebuild it ?

Answer 1

Hello Fedda,

There is no supported Microsoft procedure to directly delete or reset ReFS tiering heat map metadata on an S2D Cluster Shared Volume. If Optimize-Volume -TierOptimize is failing with 0x80070057, treat it as a storage metadata issue and first validate the health of the Storage Spaces and virtual disk layers rather than attempting to clear heat map data manually.

Start by checking:

Get-StoragePool

Get-VirtualDisk

Get-PhysicalDisk

Repair-VirtualDisk -FriendlyName <VirtualDiskName>

Then run:

Optimize-StoragePool -FriendlyName <PoolName>

to rebalance storage after the node eviction.

If the virtual disk and pool report healthy but TierOptimize continues to fail, the recommended approach is to collect cluster logs and open a Microsoft support case. There is no public PowerShell cmdlet or supported registry setting to purge and rebuild ReFS heat map data independently. In severe metadata corruption cases, the only supported recovery path may be recreating the virtual disk/volume after data protection and restoration procedures are completed.