Description of problem:
Currently dynamic resharding can cause inconsistencies between replicated zones.
Each zone may decide to reshard the same bucket.
This will cause the bucket instance id to be different and may also can cause different shard number for the bucket between the zones.
We should automatically detect it is a multisite environment and disable resharding.
Version-Release number of selected component (if applicable):
How reproducible:
Steps to Reproduce:
1.
2.
3.
Actual results:
Expected results:
Additional info:
Hi,
In order to test this issue:
1. create a Ceph cluster with multisite configuration
2. make sure dynamic resharding is enabled
3. Create a bucket with number of object exceeding the limit for resharding (number of shard * max number of object of shards)
4. make sure the bucket is not resharded (number of shards is does not increase)
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, 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-2017:3387