Virtual machines experience a complete loss of network connectivity immediately following a vMotion migration to a new ESXi host. Initial network symptoms may mimic known vSphere networking faults, but validation confirms that the vSphere vSwitch and port group configurations are healthy and operating as expected.
VMware vSphere
The issue is caused by a third-party physical switch configuration feature for ARP suppression, which drops the Gratuitous ARP (GARP) or Reverse ARP (RARP) packets generated by the ESXi host during a vMotion. This prevents the physical network from updating its MAC address table with the virtual machine's new port location.
Access the management interface or command line of the third-party physical switch connected to the destination ESXi host uplinks.
Review the active configuration on the switch ports connected to the ESXi host.
Disable the ARP suppression feature or explicitly configure the switch to permit Gratuitous ARP (GARP) and Reverse ARP (RARP) broadcasts originating from the ESXi host.
Apply and save the updated physical switch configuration.
Perform a test vMotion of a virtual machine to the destination ESXi host while running a continuous ping to verify that network connectivity is seamlessly maintained.
For ruling out native vSphere vSwitch configuration issues when diagnosing post-vMotion connectivity loss, refer to Broadcom KB 318873.