Description of problem: Client completed FFU from 13 to 16.2.3 with ceph backend. Now tried to create a 5GB volume in one of the ceph pool and it fails. In cinder-scheduler logs we see: Insufficient free virtual space (-16916.4GB) to accommodate thin provisioned 5GB volume on host After reaching out to engineering it is thought it could be this that would need backported: https://bugs.launchpad.net/cinder/+bug/1960206 I asked the client for 'ceph df -f json' and 'ceph --version' to confirm if its the case. If not then we need to understand if the calculation changed somewhere else during the FFU. We have sosreport with logs in debug. I will put below in a private comment the data we found. Thank you. Version-Release number of selected component (if applicable): OSP16.2.3 How reproducible: 100% Steps to Reproduce: 1. FFU to 16.2.3 with ceph 2. 3. Actual results: Can't create a volume on that pool Expected results: Create a volume on this pool Additional info: sosreport with cinder in debug.
The fix is present in the candidate build, namely: python3-cinder-15.6.1-2.20230906144857.el8ost.noarch openstack-cinder-15.6.1-2.20230906144857.el8ost.noarch The calculation of the available space, as shown by the cinder-scheduler logs, considers the STORED field ("stored") and not the USED field ("bytes_used"), as expected.
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory (Red Hat OpenStack Platform 16.2.6 bug fix and enhancement advisory), and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHBA-2024:1519