Hi,
My Ceph cluster is out-of-balance. The amount of PG's per OSD ranges
from about 50 up to 100 PG's per OSD. This is far from balanced.
Today, I have enabled the balancer module. But unfortunately, it doesn't
want to balance:
"Unable to find further optimization, or pool(s) pg_num is decreasing,
or distribution is already perfect"
The upmap_max_deviation setting was set to default (5). Changing this to
1, triggered the balancer module. It did 1 run, balanced a few PG's
(less than 5 PG's in total), then again it got stuck with the same
message:
"Unable to find further optimization, or pool(s) pg_num is decreasing,
or distribution is already perfect"
I have about 280 OSD's:
24 OSD's per node, 4 nodes per rack, 3 racks in total. Replica = 3. 1
replica per rack.
My disk sizes differs from 1.6T up to 2.4T.
Any advice on this?
Thanks!
Sinan
_______________________________________________
ceph-users mailing list -- ceph-users@xxxxxxx
To unsubscribe send an email to ceph-users-leave@xxxxxxx
