Sorry I was not precise enough. I will create 7 racks, the command I wrote was just there to show which one I would use. I will distribute 8 hosts into each rack. All pools are currently set to use the same default_replicated rule with 3 replicas. I have no EC Pool, RGW pool, but I have CephFS pools (with 3 replicas on both data and metadata) The OSD media is HHD all across, with separate 25G interface for public and cluster network. Client workload is 24/7/365 Ceph release is 16.2.15 (deployed/managed with ceph-ansible) Mon DBs are placed on NVME First time hearing about upmap-remapped.py, thanks- I will read up on it. _______________________________________________ ceph-users mailing list -- ceph-users@xxxxxxx To unsubscribe send an email to ceph-users-leave@xxxxxxx
 |