Details
-
Bug
-
Resolution: Fixed
-
Critical
-
7.1.0, 7.1.1
-
Untriaged
-
1
-
Unknown
Description
What's the issue?
Backup resumption for AWS/Azure/GCP does not work as expected. When populating the staging the backup sub-command "forces" population; this results in the plan being overwritten with the one from GCP (which is out-of-date). Ultimately this means resume doesn't resume, it starts from scratch and re-does the backup.
What's the fix?
We should re-evaluate the decision to make the backup sub-command force population (e.g. override the rev-id logic). The rev-id logic will catch this, and result in skipping population, ending with a correct resumption.
Attachments
Issue Links
- relates to
-
MB-53764 [CBM] Investigate how to further improve staging directory population performance
- Resolved
For Gerrit Dashboard: MB-53671 | ||||||
---|---|---|---|---|---|---|
# | Subject | Branch | Project | Status | CR | V |
180183,2 | MB-53671 Remove the ability to force override staging population | neo | backup | Status: MERGED | +2 | +1 |
180379,1 | Merge branch 'neo' into master | master | backup | Status: MERGED | +2 | +1 |