VCF Automation: VMDK Disk File Path Naming Behavior

Products

• VCF Automation
• VCF Operations/Automation (formerly VMware Aria Suite)

Issue/Introduction

Users may observe inconsistent VMDK file naming conventions depending on whether the deployment targets a Storage DRS (SDRS) cluster or a standalone datastore.

In VCF Automation, when we use an SDRS cluster for creating a deployment with a disk attached to the VM, the VMDK file name follows the convention <vm_name>_{index}.vmdk. This is observed to be different from deployments that use a single datastore (Non-SDRS), where the attached disk uses <disk_name>.vmdk. 

In general, the VMDK path when the disk is attached to the VM and present in the VM folder is: [datastore_name] vm_name/vmdk_name.vmdk. The vmdk_name generated by the following logic.

Type of Storage	vm-name	vmdk_name	Observation
SDRS	vmName-mcm-randomString	vmName-mcm-randomString_1.vmdk	Disk VMDK gets the VM name with an index
Non-SDRS	vmName-mcm-randomString	diskName-mcm-<randomString>.vmdk	Disk VMDK name is the name specified in the Cloud template (or custom name if any)

Environment

VCF Automation 9.x
Aria Automation 8.x

Cause

For deployments on standard datastores, the VMDK filename preserves the specific disk name generated by VCF Automation (whether defined in the Cloud Template or modified by Custom Naming policies). This results in the format vm_name/disk_name.vmdk. However, in SDRS-backed deployments, vCenter controls the placement and uses its native naming convention. Consequently, vCenter defaults the file name to match the VM name with a numeric suffix (e.g., vm_name/vm_name_{index}.vmdk).

Resolution

vCenter doesn't offer any API to rename the disk VMDK along with the backing file name once the disk is provisioned.

This is an expected behavior in Aria Automation 8.x and VCF Automation 9.x