On Mon, Dec 2, 2013 at 9:43 AM, Gary Harris (gharris) <[email protected]>wrote:
> Hi, I have a question about CEP high availability when integrated with > OpenStack. > > > Assuming you have all openstack controller nodes in HA mode would your > actually > have an HA CEPH implementation as well meaning two Primary OSDs or both > pointing to the primary? > Ceph is clustered, replicated, distributed storage so the concepts of primary/secondary are bit different. Each osd will have its own local storage. Ceph will take care of replication between osds based on how you configure it. If an osd fails, clients will look for one of the replicas. Ceph will also take care of re-replicating the data around the cluster as needed. > > Or, do the client requests get forwarded automatically to the > secondary OSD should the primary fail? > (Excuse the simplifications):) > > So my assumption is that if the primary fails, the mon would detect and > a new primary OSD candidate would be presented to clients? > The mon will detect the error and update the crush map as needed so that clients can figure out where their data is. You can read about the crush map at http://ceph.com/docs/master/rados/operations/crush-map/. > > Thanks, > Gary > > > > _______________________________________________ > ceph-users mailing list > [email protected] > http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com > > -- Randall Smith Computing Services Adams State University http://www.adams.edu/ 719-587-7411
_______________________________________________ ceph-users mailing list [email protected] http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
