Failed to add a new disk for one vm in WSFC

Products

• VMware vSphere ESX 8.x

Issue/Introduction

When attempting to add a new virtual disk (VMDK) to a Virtual Machine that is part of a Windows Server Failover Cluster (WSFC), the operation fails with the following error: "Insufficient disk space on datastore".

Environment observations:

• The vSphere Datastore capacity is at 97% utilization.

• The VM uses Thick Provisioned disks (typically Eager Zeroed, required for WSFC shared storage).

• Inside the Guest OS (Windows), Disk #: is nearly full, while other drives have ample free space.

• Standard guest-level space reclamation (UNMAP) is not effective because the disks are thick-provisioned at the virtualization layer.

 

 

Environment

VMware vSphere ESX 8.x

Cause

The datastore is at 97%, it has reached a critical threshold where the filesystem cannot safely guarantee the block allocation for a new thick disk. Furthermore, since the existing disks are thick-provisioned, the space is physically locked on the datastore regardless of how much "empty" space Windows says it has.

Resolution

To resolve the immediate "Insufficient Space" error without adding more physical hardware to the datastore:

1. Identify disks within the Guest OS that have high free space (e.g., Disk D: or Disk E:).

2. Copy/Move non-critical data from the nearly full Disk #: to these underutilized disks.

3. Once the data is moved, delete the original files on Disk #:.