This article describes a scenario where a "leftover" VMX process remains on an ESXi host after a virtual machine has been moved or failed over. This process is associated with a virtual machine that is no longer present in the host's inventory.
Symptoms
The operation is not allowed in the current state".This issue occurs when hostd is restarted during a storage outage, such as an All Paths Down (APD) scenario. If vSphere HA is enabled, it may unregister the VM to start it on a different host, leaving the original VMX process active on the source host.
To resume normal operations, the leftover VMX process must be terminated manually.
esxcli vm process list esxcli vm process kill -t force -w [WorldID]For further assistance with unresponsive virtual machines, see Powering off an unresponsive virtual machine on an ESXi host.