On Tue, Oct 25, 2016 at 2:46 PM, J David <j.david.lists@xxxxxxxxx> wrote: > Are long-running RBD clients (like Qemu virtual machines) placed at > risk of instability or data corruption if they are not updated and > restarted before, during, or after such an upgrade? No, we try very hard to ensure forward and backwards compatibility. However, since firefly is EOL and our testing capacity is finite, I don't believe we perform any direct tests between firefly clients and jewel clusters. > If so, what are the potential consequences, and where in the process > should they be upgraded to avoid those consequences? In general, I would recommend upgrading the librbd clients after the cluster is fully upgraded. It really shouldn't matter unless you are attempting to use new CRUSH map / RBD features without the necessary backing support in the cluster. Assuming your VM environment is properly set up, you can use live migration to transparently upgrade the running librbd version within VMs. [1] http://docs.ceph.com/docs/master/releases/ -- Jason _______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
 |