Storage pool won’t claim replacement SAS drive due to stale metadata

Question

Hello,

After replacing a failed SAS drive in one of our Storage Spaces Direct (S2D) cluster nodes, we expected the new disk to be automatically detected and added back into the storage pool. Instead, the replacement drive is showing Unhealthy with an Operational Status of Stale Metadata, and the storage pool refuses to claim it as an available physical disk.

We've already refreshed the storage subsystem, rescanned the disks, and verified that the hardware and RAID controller are functioning correctly. However, the stale metadata appears to be preventing the disk from being recognized as a clean candidate for the existing pool, leaving the repair process unable to continue.

Could someone please explain the recommended method to safely clear the stale Storage Spaces pool metadata directly from the physical disk so that it can be reclaimed by the S2D storage pool without risking the integrity of the remaining cluster disks ?

Answer 1

The disk shows StaleMetadata because old Storage Spaces pool metadata is still on it. S2D needs the drive to be clean, so the way to clear that metadata is to run Reset‑PhysicalDisk on the replacement disk after checking it isn’t in any pool. That clears the leftover metadata and the disk becomes usable for the S2D pool again.

Answer 2

Stale metadata means Storage Spaces found old metadata on the replacement disk. The documented recovery path is:

1. First try a repair on the affected virtual disks:

      Repair-VirtualDisk
   

   Stale metadata should be temporary, and if the disk returns to OK, no reset is needed.
2. If the disk does not transition back to OK, reset the physical disk to wipe the old pool metadata:

      Reset-PhysicalDisk
   

   This erases all data on that disk.
3. After the reset, run the repair again:

      Repair-VirtualDisk
   

4. To confirm whether the disk is now eligible to be pooled, check its pooling state and reason:

      Get-PhysicalDisk | Format-Table FriendlyName,MediaType,Size,CanPool,CannotPoolReason
   

   If CanPool is False, CannotPoolReason shows why the disk still is not eligible.

If the replacement disk is healthy but still is not automatically added to the pool, AutoPool may be disabled. In that case:

• Check AutoPool:

    Get-StorageSubsystem Cluster* | Get-StorageHealthSetting | select "System.Storage.PhysicalDisk.AutoPool.Enabled"
  

• If AutoPool is disabled, either manually add the disk:

    Get-PhysicalDisk -CanPool $true
    Get-StoragePool -IsPrimordial $False
    $disks = Get-PhysicalDisk -CanPool $true
    Add-PhysicalDisk -StoragePoolFriendlyName "FriendlyName_from_step2" -PhysicalDisks $disks
  

• Or enable AutoPool:

    Get-StorageSubsystem Cluster* | Set-StorageHealthSetting -Name "System.Storage.PhysicalDisk.AutoPool.Enabled" -Value True
  

Important: Reset-PhysicalDisk is destructive for the selected disk. Use it only on the replacement drive that contains no needed data.

---

• Troubleshoot Storage Spaces and Storage Spaces Direct health and operational states
• Troubleshoot Storage Spaces and Storage Spaces Direct health and operational states
• Replace failed drives on Azure Local