Virtual Machines with Physical RDM Disks Failing to Migrate or Power On After Upgrading to ESXi 7.0 Update 3, Resulting in Error:"File system specific implementation of Ioctl[file] failed The source detected that the destination failed to resume"

Products

• VMware vSphere ESXi

Issue/Introduction

• After the recent upgrade to ESXi 7.0.3 Build: 18644231, virtual machines with Physical RDM disks can not migrate or be powered on.
• The following error message is displayed:
  
  File system specific implementation of Ioctl[file] failed
The source detected that the destination failed to resume.

The VM failed to resume on the destination during early power on.
Module Disk power on failed.
Cannot open the disk '/vmfs/volumes/<datastore uuid>/<vm folder>/<specific vmdk pointer to rdm>' or one of the snapshot disks it depends on.
Failed to lock the file File system specific implementation of OpenFile[file] failed
VM name : <vm name>
Datastore UUID: <datastore uuid>

Environment

VMware vSphere 7.0.x

Cause

When services like SNMP and PTP are running, these services may open RDM devices. A condition may occur where these services fail to close their open device handles properly due to a change in the closure mechanism introduced in ESXi 7.0 U3, resulting in device handles incorrectly being left open resulting in "failed to lock the file" conflicts when another host attempting to power on Virtual Machines with Physical RDM LUNS.

Resolution

This issue is resolved in ESXi 7.0 Update 3c.

Workaround

In some instances, rebuilding the RDM mapping works around the issue.