I've seen issues where a VM gets stuck in some state permanently. After that, no other operations succeed. As I also mentioned, you can reset-state as an admin user, then delete but this deadlock situation shouldn't occur in the first place. Or at least, a subsequent delete operation should go through. Otherwise its a pain in the ass to clean up. Basically, it is all to easy to get VMs stuck in a some permanent state
We have some scripts which build the VM, we observe that a VM can stay in “build” state forever and can’t be deleted without admin operation and pf9-ostackhost service restart on the hypervisor.