Description of problem: When the cinder backend is removed, it is not anymore possible to remove the cinder volumes. The volumes remain in deleting state forever. Even running openstack volume delete --force -- There are workarounds to fix this in the DB, but they come with the risk that they become outdated when OpenStack is changing (e.g. https://access.redhat.com/solutions/2578061) they might have undocumented side effects. Version-Release number of selected component (if applicable): How reproducible: Each time you remove the cinder backend (in this case ceph). Steps to Reproduce: 1. Add a cinder backend (e.g. ceph) 2. Stop the backend or removed it form network) 3. Try to delete any volume that was stored on the cinder backend Actual results: The volume remains in deleting state. Expected results: The volume to be deleted Additional info:
Bug 1638820 was opened for OSP13 Blueprint upstream: https://blueprints.launchpad.net/cinder/+spec/cinder-delete-from-db
*** Bug 1638820 has been marked as a duplicate of this bug. ***
Hello, Can we get a clear set of instructions for removing orphaned cinder volumes from an OSP 10 database? Cory Oldford
This feature was reviewed during PTG, the idea is to add an option in cinder unmanage to forcefully unmanage a volume which would nicely clean it from the environment. Adding PM ack for OSP 18.0