Description of problem:
  I created a overcloud with 3 controllers and 11 computes.   1 controller nodes failed creation,  and the cluster is in create_failed  status.

   Since it got stuck in this process for over an hour, we decided to delete the stack and recreate the cluster.  However,   "heat stack-delete cluster-name" failed,  and the cluster won't go away.  

  Repeated the commands 10 times,  still no luck.    Eventually,  we rebooted the OSP director node,  still stuck in same state.  Then we set the overcloud nodes (as shown by nova list) to "active" status manually,  and sent the "heat stack-delete" command again.  This seem to trigger another round of deletion which cleared the stack.

Version-Release number of selected component (if applicable):  OSPD 9


How reproducible:

Not tried to reproduce

Steps to Reproduce:
1.
2.
3.

Actual results:


Expected results:


Additional info: