On Fri, Jan 22, 2016 at 9:19 AM, James Gallagher <james.np.gallagher@xxxxxxxxx> wrote: > Hi there, > > Got a quick question regarding the CephFileSystem. After implementing the > setup from the quick start guide and having a Admin-Node, Monitor/Metadata > Server, OSD0 and OSD1. If I were to implement the mount on the Admin-Node at > the location /mnt/mycephfs, if I were then to transfer files to > /mnt/mycephfs - will the files be then distributed amongst OSD0 and OSD1, > where if the replication level is set at 2, the files should then be on both > nodes? > > If not, what implementation should I be looking to do, in order for the > files to be distributed amongst these devices? > All file data in CephFS is chunked (into 4MB pieces, by default) and stored as standard RADOS objects. So assuming you run with the defaults, yes, all the file data will be on both OSDs if you're running with only two. -Greg > Thanks, > > J. Gallagher > > _______________________________________________ > ceph-users mailing list > ceph-users@xxxxxxxxxxxxxx > http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com > _______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
 |