On 01/29/2018 04:14 PM, Niklas wrote:
Thank you for the answer.
It seems like a good solution to use uniform in this case.
But adding this in the crush map result in failure to compile crush map.
datacenter stray {
alg uniform
hash 0
}
$: crushtool -c /tmp/crush.text -o /tmp/crush
in rule 'hdd' step take default has no class information
rule hdd {
id 2
type replicated
min_size 1
max_size 3
step take default class hdd
step chooseleaf firstn 0 type datacenter
step emit
}
Rule 'hdd' has nothing to do with datacenter 'stray'. So using uniform
anywhere in crush map disables/removes new class information?
Good question! I didn't try a combination of the two yet. It's possible
that it doesn't work indeed.
Wido
On 2018-01-29 15:42, Wido den Hollander wrote:
On 01/29/2018 03:38 PM, Niklas wrote:
When adding new OSDs to a host, the CRUSH weight for the datacenter
one level up is changed to reflect the change.
What configuration is used to stop ceph from automatic weight
management on the levels above the host?
In your case (datacenters) you might want to change the 'alg' to
uniform instead of straw(2).
This 'alg' doesn't support weights, but if all your DCs are always
equal you can use that.
Source:
http://docs.ceph.com/docs/master/rados/operations/crush-map-edits/#crush-map-bucket-hierarchy
Wido
_______________________________________________
ceph-users mailing list
ceph-users@xxxxxxxxxxxxxx
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
_______________________________________________
ceph-users mailing list
ceph-users@xxxxxxxxxxxxxx
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
_______________________________________________
ceph-users mailing list
ceph-users@xxxxxxxxxxxxxx
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
_______________________________________________
ceph-users mailing list
ceph-users@xxxxxxxxxxxxxx
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
