Hello all, I have a naive question about the way and the maximum rebuild speed for erasure coding pool. I did some search, but could not find any formal and detailed information about this. For pool recovering, the way Ceph works(to my understanding) is: each active OSD scrubs the drive, and if it find any degraded PGs, it will then try to recover/rebuild it, and distribute the recovered PGs to the remaining OSDs? For erasure pool, suppose I have 10 nodes, each has 10 6TB drives, so in total 100 drives. I make a 4+2 erasure pool, failure domain is host/node. Then if one drive failed, (assume the 6TB is fully used), what the maximum speed the recovering process can have? Also suppose the cluster network is 10GbE, each disk has maximum 200MB/s sequential throughput. I suppose that each OSD(or master OSD?) will try to rebuild by reading in other related shards and rebuild the lost one, and then redistribute it to other OSD? Besides, when recovering, will mgr, mds, mon also work for it? Thanks! Best, Feng _______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
 |