Hi Ceph community, I’m trying to upload some file with 5GB size, through radosgw, I have 9 OSDs deployed on 3 machines, and my cluster is healthy. The problem is: the 5GB file is being uploaded to osd.0 and osd.1 ,which are near full, while the other OSDs have more space that can have this file .. why would ceph (may be radosgw) do this? Is there a misconfiguration to ceph or misunderstanding of how ceph works .. Another question: Are data are sharded by default on ceph, or we have to configure it? Or the problem is not related to sharding in the first place? I appreciate everybody’s help Thanks! Beanos _______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
 |