Isn’t the right parameter “network=writeback” for network devices like RBD? Jan > On 08 Jun 2015, at 12:31, Andrey Korolyov <andrey@xxxxxxx> wrote: > > On Mon, Jun 8, 2015 at 1:24 PM, Arnaud Virlet <avirlet@xxxxxxxxxxxxxxx> wrote: >> Hi >> >> Actually we use libvirt VM with ceph rbd pool for storage. >> By default we want to have "disk cache=writeback" for all disks in libvirt. >> In /etc/ceph/ceph.conf, we have "rbd cache = true" and for each VMs XML we >> set "cache=writeback" for all disks in VMs configuration. >> >> We want to use one ocfs2 volume on our rbd pool. For this volume we want set >> cache=none. When we set cache=none in libvirt template for this hosts, it >> doesn't work. > > Can you please describe this more specifically? Does the libvirt > produce a launch string with cache=none or you are measuring the cache > (mis)presence in some other way? The rbd_cache setting and cache=xxx > for qemu should show a conjugate behavior. > >> The only way that's work is when we set "rbd cache = false" in >> /etc/ceph/ceph.conf. >> >> How can I set cache=none just for one volume specifically without modifing >> default settings in ceph.conf? >> >> >> Regards, >> >> Arnaud >> _______________________________________________ >> 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 _______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com