The traffic on the cluster network is all communication between OSDs. No other ceph daemon other than OSDs use that network. All communication to the OSDs from any other daemon or service happens on the public network.
For a write on a replicated pool with size 3 the client writes to the primary osd across the public network and then the primary osd sends the other 2 copies across the cluster network to the secondary OSDs. So for writes the public network uses N bandwidth while the cluster use 2N bandwidth for the replica copies. Seeing as the write isn't acknowledged until all 3 copies are written it makes no sense to have a faster public network.
I would probably suggest just ditching the separate cluster network and only have the 1 public network for both types of traffic. If you want them to be isolated, use the faster network as the cluster network.
On Wed, May 9, 2018, 4:46 PM Gandalf Corvotempesta <gandalf.corvotempesta@xxxxxxxxx> wrote:
As subject, what would happen ?
_______________________________________________
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
