vSAN disk claiming failure due to existing partition metadata

Products

• VMware vSAN

Issue/Introduction

Host nodes are unable to claim SSDs when building or expanding a vSAN cluster. 

• Attempts to erase disk partitions via vSphere Client fail.
• CLI commands via partedUtil return: Error: Read-only file system during write.
• Disk status remains "Ineligible" or "Foreign".

 

Environment

• VMware vSAN 8.x
• VMware vSAN 7.x
• VMware vSAN 9.0
• VMware vSAN 9.1
• ESXi 7.0/8.0/9.0/9.1

Cause

The ESXi VMkernel flags foreign partition tables as active system disks to protect perceived OS data. This creates a software lock that prevents native ESXi management tools from overwriting the GUID Partition Table (GPT).

Resolution

Bypass the ESXi software layer to clear legacy metadata using one of the following methods:

Method 1: Out-of-Band Management (Recommended)

1. Log in to the hardware management interface (e.g., Dell iDRAC, HPE iLO).
2. Navigate to Storage > Physical Disks.
3. Select the affected disk and perform a Format, Initialize, or Sanitize operation.
4. Reboot the host and perform a storage rescan in ESXi: esxcfg-rescan -A.

Method 2: Live Partition Utility

1. Boot the affected host from a Live Linux USB (e.g., GParted Live).
2. Delete all existing partitions on the affected drive.
3. Initialize a new, blank GPT partition table. Leave space as unallocated.
4. Reboot into ESXi and claim the disk.

If the issue persists, see Contact Support.