Hello, I run a ceph nautilus cluster where we use rbd for data storage and retrieval. THe rbd images metadata are in a replicated pool with 1+2 copies. The data are placed in a pool that uses erasure coding with a 4+2 profile. Now I am unsure about the exact meaing or min_size for both pools. The metadata pool has a min_size of 2, the erasure coded pool has a min_size of 5. The clusters redundancy is based on hosts. As far as I understand "min_size", this would mean that the cluster would be able to access data in the pools if one host would die. In case two hosts would be offline at the same time the metadata-pool would stop delivering data, the ec-Pool would basically be able to deliver data since min_size is irrelevant to ec-pools instead the pools K-value (==4) determines the minimal number of available OSDs to continue working. In the end I would no longer be able to read data in this situation because there is only one working OSD remaining for the replicated pool and min_size is 2. If this is correct, than does it make sense to increase the replication from 3 to 4 for the metadata pool so that I would be able to read and write data even if two hosts should go down at the same time? Is my understanding correct? Thank you very much for your help Rainer -- Rainer Krienke, Uni Koblenz, Rechenzentrum, A22, Universitaetsstrasse 1 56070 Koblenz, Tel: +49261287 1312 Fax +49261287 100 1312 Web: http://userpages.uni-koblenz.de/~krienke PGP: http://userpages.uni-koblenz.de/~krienke/mypgp.html _______________________________________________ ceph-users mailing list -- ceph-users@xxxxxxx To unsubscribe send an email to ceph-users-leave@xxxxxxx
 |