Repository navigation
CLDSRV-1007: enforce object lock in DeleteObjects regardless of bucket config - #6385
Conversation
Hello francoisferrand,My role is to assist you with the merge of this Available options
Available commands
Status report is not available. |
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files
... and 2 files with indirect coverage changes @@ Coverage Diff @@
## development/9.6 #6385 +/- ##
===================================================
- Coverage 90.86% 90.84% -0.02%
===================================================
Files 208 208
Lines 39275 39275
===================================================
- Hits 35686 35679 -7
- Misses 3589 3596 +7
Flags with carried forward coverage won't be shown. Click here to find out more. 🚀 New features to boost your workflow:
|
DeleteObjects skipped the per-version lock check when the bucket is not object-lock enabled, while DeleteObject enforces it from the object metadata alone. Lock state can exist on such buckets when metadata is written below the API (e.g. DR replication), so locked versions could be deleted through DeleteObjects only. Both paths now behave the same; it changes nothing for objects without lock state. Issue: CLDSRV-1007
aca73a1 to
5152cbf
Compare
Waiting for approvalThe following approvals are needed before I can proceed with the merge:
|
|
/approve |
|
I have successfully merged the changeset of this pull request
The following branches have NOT changed:
This pull request did not target the following hotfix branch(es) so they
Please check the status of the associated issue CLDSRV-1007. Goodbye francoisferrand. The following options are set: approve |
DeleteObjects let you delete a locked version (retention or legal hold) as long as the bucket itself wasn't object-lock enabled. DeleteObject has never worked that way: it enforces the lock based on the object alone, whatever the bucket config. This brings DeleteObjects in line.
You can't get into that state through the S3 API, but it happens when metadata is written behind it, typically by replication, where the DR bucket isn't lock-enabled but the replicated objects still need to stay protected.
Objects without a lock are unaffected.
Issue: CLDSRV-1007