I found another problem relating to "swift_delay_auth_decision" If that is set to the default "no" and you configure cinder to do volume backups to swift and try to do a backup to a container you'll get the following error in the cinder_backup logs: 2022-07-03 19:31:59.315 18 ERROR cinder.backup.drivers.swift [-] Can not get Swift capabilities during backup drive r initialization.: swiftclient.exceptions.ClientException: Capabilities GET failed: http://10.10.20.62:8080/info 40 1 Unauthorized [first 60 chars of response] b'{"error": {"code": 401, "title": "Unauthorized", "message": ' (txn: txcb5b4cd60cdf4fa1bcb25-0062c234ff) 2022-07-03 19:31:59.315 18 ERROR cinder.backup.drivers.swift Traceback (most recent call last): 2022-07-03 19:31:59.315 18 ERROR cinder.backup.drivers.swift File "/var/lib/kolla/venv/lib/python3.8/site-package s/cinder/backup/drivers/swift.py", line 420, in check_for_setup_error 2022-07-03 19:31:59.315 18 ERROR cinder.backup.drivers.swift conn.get_capabilities() 2022-07-03 19:31:59.315 18 ERROR cinder.backup.drivers.swift File "/var/lib/kolla/venv/lib/python3.8/site-package s/swiftclient/client.py", line 2059, in get_capabilities 2022-07-03 19:31:59.315 18 ERROR cinder.backup.drivers.swift return get_capabilities((parsed, self.http_conn[1] )) 2022-07-03 19:31:59.315 18 ERROR cinder.backup.drivers.swift File "/var/lib/kolla/venv/lib/python3.8/site-package s/swiftclient/client.py", line 1682, in get_capabilities 2022-07-03 19:31:59.315 18 ERROR cinder.backup.drivers.swift raise ClientException.from_response( 2022-07-03 19:31:59.315 18 ERROR cinder.backup.drivers.swift swiftclient.exceptions.ClientException: Capabilities G ET failed: http://10.10.20.62:8080/info 401 Unauthorized [first 60 chars of response] b'{"error": {"code": 401, "t itle": "Unauthorized", "message": ' (txn: txcb5b4cd60cdf4fa1bcb25-0062c234ff) 2022-07-03 19:31:59.315 18 ERROR cinder.backup.drivers.swift 2022-07-03 19:31:59.317 18 ERROR oslo.service.loopingcall [-] Fixed interval looping call 'cinder.backup.manager.Ba ckupManager._setup_backup_driver' failed: swiftclient.exceptions.ClientException: Capabilities GET failed: http://1 0.10.20.62:8080/info 401 Unauthorized [first 60 chars of response] b'{"error": {"code": 401, "title": "Unauthorize d", "message": ' (txn: txcb5b4cd60cdf4fa1bcb25-0062c234ff) 2022-07-03 19:31:59.317 18 ERROR oslo.service.loopingcall Traceback (most recent call last): 2022-07-03 19:31:59.317 18 ERROR oslo.service.loopingcall File "/var/lib/kolla/venv/lib/python3.8/site-packages/o slo_service/loopingcall.py", line 150, in _run_loop 2022-07-03 19:31:59.317 18 ERROR oslo.service.loopingcall result = func(*self.args, **self.kw) 2022-07-03 19:31:59.317 18 ERROR oslo.service.loopingcall File "/var/lib/kolla/venv/lib/python3.8/site-packages/c inder/backup/manager.py", line 162, in _setup_backup_driver 2022-07-03 19:31:59.317 18 ERROR oslo.service.loopingcall backup_service.check_for_setup_error() 2022-07-03 19:31:59.317 18 ERROR oslo.service.loopingcall File "/var/lib/kolla/venv/lib/python3.8/site-packages/c inder/backup/drivers/swift.py", line 420, in check_for_setup_error 2022-07-03 19:31:59.317 18 ERROR oslo.service.loopingcall conn.get_capabilities() 2022-07-03 19:31:59.317 18 ERROR oslo.service.loopingcall File "/var/lib/kolla/venv/lib/python3.8/site-packages/s wiftclient/client.py", line 2059, in get_capabilities 2022-07-03 19:31:59.317 18 ERROR oslo.service.loopingcall return get_capabilities((parsed, self.http_conn[1])) 2022-07-03 19:31:59.317 18 ERROR oslo.service.loopingcall File "/var/lib/kolla/venv/lib/python3.8/site-packages/s wiftclient/client.py", line 1682, in get_capabilities 2022-07-03 19:31:59.317 18 ERROR oslo.service.loopingcall raise ClientException.from_response( 2022-07-03 19:31:59.317 18 ERROR oslo.service.loopingcall swiftclient.exceptions.ClientException: Capabilities GET failed: http://10.10.20.62:8080/info 401 Unauthorized [first 60 chars of response] b'{"error": {"code": 401, "titl e": "Unauthorized", "message": ' (txn: txcb5b4cd60cdf4fa1bcb25-0062c234ff) 2022-07-03 19:31:59.317 18 ERROR oslo.service.loopingcall 2022-07-03 19:31:59.340 18 INFO cinder.keymgr.migration [req-3d3dacb5-fdbb-428b-bc35-0cb9ab7183a4 - - - - -] Not mi grating encryption keys because the ConfKeyManager's fixed_key is not in use. 2022-07-03 19:32:09.384 18 ERROR cinder.service [-] Manager for service cinder-backup os4.cedar.mulhollon.com is re porting problems, not sending heartbeat. Service will appear "down". 2022-07-03 19:32:19.387 18 ERROR cinder.service [-] Manager for service cinder-backup os4.cedar.mulhollon.com is re porting problems, not sending heartbeat. Service will appear "down". 2022-07-03 19:32:29.388 18 ERROR cinder.service [-] Manager for service cinder-backup os4.cedar.mulhollon.com is re porting problems, not sending heartbeat. Service will appear "down". At that point you get an error message about cinder-backup docker having crashed every ten seconds until your hard drive fills up. Filling up the hard drive with repeated log messages is not good behavior either... Also note that cinder volume backups become un-managable as cinder_backup has crashed. So the backup can't be cancelled or deleted. Changing "swift_delay_auth_decision" to yes, then manually stopping and restarting the cinder_backup container makes it possible to cancel/delete the ongoing backup and stops logging the error message every ten seconds. Also NEW backup jobs will succeed and store the files into swift. I would strongly suggest that it is a bug for Kolla-Ansible to permit a user to configure "cinder_backup_driver: swift" at the same time as the default configuration of "swift_delay_auth_decision: no" because that'll crash the cinder_backup container as soon as a volume backup is attempted. At least as of Yoga. For various hardware compatibility reasons I cannot test on versions older than Yoga, sorry.