Created attachment 1311635 [details] logs Version-Release number of selected component (if applicable): ceph version 12.1.2-1.el7cp How reproducible: always Steps to Reproduce: 1. check for resharded bucket using 'radosgw-admin reshard list' 2. Put/delete a new object in that bucket while it is in resharding state 3. object creation and deletion operations should be unsuccessful during resharding.
This is the correct behavior as this is online resharding. The blocking is done inside radosgw and is transparent to the User. We block the operation till the resharding is done and than execute it. You need to confirm that the resharded bucket index is correct, i.e it has all the operations that happened during resharding.