Revision history [back]

If you are using LVM with ISCSI connections its likely you are severing the connection to the target before it can gracefully detach. Basically detach the volume prior to deleting the instance, then the volume will be in an available state, remember you can only delete a volume which is in an available state.

If the volume is stuck in detaching or error i'd recommend trying 'cinder reset-state --state available $volid' or following the suggestions mentioned here https://ask.openstack.org/en/question/1378/cannot-remove-volume/