Hi，Nigel I read some PPT document and find that when n (n<M) host get down, the PG become degraded, is it mean that client still could write to the degraded PG even if there not have enough host to hold all the K+M chunks? (like the replicated PG). And if n=M (M host get down), what is state the pg will be ?
-- Yueliang Sent with Airmail On June 24, 2015 at 15:10:46, Nigel Williams (nigel.d.willi...@gmail.com) wrote: On Wed, Jun 24, 2015 at 4:29 PM, Yueliang <yueliang9...@gmail.com> wrote: > When I use K+M hosts in the EC pool, if M hosts get down, still have K hosts > active, Can I continue write data to the pool ? If your CRUSH map specifies a failure-domain at the host level (so no two chunks share the same host) then you will be unable to write to the pool. If instead the failure-domain is OSD then with enough OSDs pool writes would still be accepted. > Since there only have K > hosts, not K+M hosts, When client write a data to EC pool , Primary OSD will > split the data to K data pieces，but how about the M coding pieces? is it > still be calculated and where it should be hold ? Same as above, with failure-domain = host then there would be nowhere to put the M coding pieces.
_______________________________________________ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
