VM connected to NSX segment loses network connectivity after HCX RAV Migration in Single vCenter Environment

Products

• VMware NSX
• VMware HCX

Issue/Introduction

Virtual machines experience total network connectivity loss immediately following a successful HCX Replication Assisted vMotion (RAV) migration within a single vCenter environment.

• The migration completes successfully in HCX.
• The VM is powered on and present in the destination inventory.
• The VM is unable to pass any network traffic (heartbeat loss, application timeout).
• We observed this issue on virtual machines that were added to the Distributed Firewall using their VM names

Environment

VMware NSX 

VMware HCX

Cause

The root cause is a Distributed Firewall (DFW) policy mismatch triggered by a forced change in the VM's Instance UUID (vc.uuid).

In a single vCenter HCX RAV migration, vCenter detects a temporary UUID conflict between the active source VM and the registered shadow VM at the destination. To resolve this conflict, vCenter enforces uniqueness by assigning a new vc.uuid to the target VM.

Resolution

To restore connectivity, you must manually revert the VM's Instance UUID to match the original source identity.

1. Power off the affected virtual machine.
   
   
2. Locate the VM's folder on the datastore and download/edit the .vmx configuration file.
   
   
3. Modify the vc.uuid parameter to match the original source UUID.
   
   
4. Save the file and Power on the VM.
   
   
5. Verify that the VM now correctly populates within its intended NSX Security Group and that connectivity is restored.

For more details: Changing or keeping a UUID for a moved virtual machine

Additional Information

MAC address is not retained using the "Retain MAC" option after migrating a virtual machine using HCX within the same VC (Single vCenter Operation)

• For instructions on downloading Broadcom products, see Download Broadcom Products and Software.
• To contact support, see Contact Broadcom Support.