Bug 1802263
| Summary: | [RHOS 16.1] cinder backup-import is removing entries from DB when the volume backup exists. | |||
|---|---|---|---|---|
| Product: | Red Hat OpenStack | Reporter: | Darin Sorrentino <dsorrent> | |
| Component: | openstack-cinder | Assignee: | Brian Rosmaita <brian.rosmaita> | |
| Status: | CLOSED ERRATA | QA Contact: | Evelina Shames <eshames> | |
| Severity: | medium | Docs Contact: | Andy Stillman <astillma> | |
| Priority: | medium | |||
| Version: | 16.0 (Train) | CC: | brian.rosmaita, jelynch, ltoscano, senrique | |
| Target Milestone: | z9 | Keywords: | Triaged | |
| Target Release: | 16.1 (Train on RHEL 8.2) | |||
| Hardware: | Unspecified | |||
| OS: | Unspecified | |||
| Whiteboard: | ||||
| Fixed In Version: | openstack-cinder-15.4.0-1.20221003203219.58f0e73.el8ost | Doc Type: | Bug Fix | |
| Doc Text: |
Before this update, if you imported a backup record for a backup ID that currently existed, the import operation would correctly fail, but the existing backup record would incorrectly be deleted. With this update, the existing backup record is not deleted under this scenario.
|
Story Points: | --- | |
| Clone Of: | ||||
| : | 2131386 (view as bug list) | Environment: | ||
| Last Closed: | 2022-12-07 20:24:45 UTC | Type: | Bug | |
| Regression: | --- | Mount Type: | --- | |
| Documentation: | --- | CRM: | ||
| Verified Versions: | Category: | --- | ||
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | ||
| Cloudforms Team: | --- | Target Upstream Version: | ||
| Embargoed: | ||||
| Bug Depends On: | 2131386 | |||
| Bug Blocks: | 1802255 | |||
|
Description
Darin Sorrentino
2020-02-12 18:16:02 UTC
(In reply to Darin Sorrentino from comment #0) > > To create the scenario in order to test the backup export/import: Isn't the scenario already described by the steps reported above, with all the openstack commands? If it is correct, what is the purpose of the following steps? > > Create backup of a volume, note backup ID > Export the DB data using cinder backup-export > SSH to overcloud controller > Use podman to enter the galera container > Use mysql to access the cinder database > Delete from cinder.backup where id='<backup ID>' > Exit back out to Director > Confirm the backup is gone by listing them and perform the import as > detailed above. Also, does it work if you use the "cinder" instead of openstack volume? Luigi,
No, it's not. As far as I read, the purpose of the cinder backup is to restore the DB entries in the event of a catastrophic loss of data in the DB. In order to "imitate a catastrophic loss of data", one must actually go into the DB and delete the record, which is the steps I provided:
> Create backup of a volume, note backup ID
> Export the DB data using cinder backup-export
> SSH to overcloud controller
> Use podman to enter the galera container
> Use mysql to access the cinder database
> Delete from cinder.backup where id='<backup ID>'
> Exit back out to Director
> Confirm the backup is gone by listing them and perform the import as
> detailed above.
Once you do that, THEN you can use the cinder backup restore procedure to restore the volume to the DB as detailed in the BZ description.
The import/export commands are "cinder" commands and not "openstack" commands.
Are you asking if "cinder list" has a different output from "openstack volume list"? If that is what you're asking, no, it does not. Both show the same information.
Thanks. Reproduced. Fix is present in openstack-cinder-15.4.0-1.20221003203219.58f0e73.el8ost, which has the 'rhos-16.1-rhel-8-trunk-candidate' tag: https://brewweb.engineering.redhat.com/brew/buildinfo?buildID=2192209 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 16.1.9 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-2022:8795 |