Bug #39495
rgw: swift object expiry fails when a bucket reshards
% Done:
0%
Source:
Tags:
Backport:
nautilus, mimic, luminous
Regression:
No
Severity:
3 - minor
Reviewed:
Description
A bucket reshard causes swift object expiry to fail. While further swift client requests will still return a 404 as we check for obj expiry times in GET requests, the object is never deleted on disk, and the delete hint is already trimmed as we assume a non-findeable bucket means the bucket was deleted
$ swift upload foobar foo -H "X-Delete-After: 100" $ radosgw-admin bucket reshard foobar --num-shards 4 $ sleep 100 && radosgw-admin objects expire --debug-rgw=20 $ radosgw-admin object stat --bucket foobar --object foo # returns
Related issues
History
#1 Updated by Abhishek Lekshmanan almost 4 years ago
- Description updated (diff)
#2 Updated by Abhishek Lekshmanan almost 4 years ago
- Status changed from New to 17
- Pull request ID set to 27817
#3 Updated by Abhishek Lekshmanan almost 4 years ago
followup fixes added in https://github.com/ceph/ceph/pull/27870
#4 Updated by Casey Bodley over 3 years ago
- Status changed from 17 to Pending Backport
#5 Updated by Nathan Cutler over 3 years ago
- Copied to Backport #39740: nautilus: rgw: swift object expiry fails when a bucket reshards added
#6 Updated by Nathan Cutler over 3 years ago
- Copied to Backport #39741: mimic: rgw: swift object expiry fails when a bucket reshards added
#7 Updated by Nathan Cutler over 3 years ago
- Copied to Backport #39742: luminous: rgw: swift object expiry fails when a bucket reshards added
#8 Updated by Nathan Cutler over 3 years ago
- Pull request ID changed from 27817 to 27870
#9 Updated by Nathan Cutler about 2 years ago
- Status changed from Pending Backport to Resolved
While running with --resolve-parent, the script "backport-create-issue" noticed that all backports of this issue are in status "Resolved" or "Rejected".