George,
Check the instructions here which should allow you to test your crush rules
without applying them to your cluster.
http://dachary.org/?p=3189
also, fwiw, we are not using an 'emit' after each choose (note these rules
are not implementing what you're trying to)-
# rules
rule replicated_ruleset {
ruleset 0
type replicated
min_size 1
max_size 10
step take default
step choose firstn 2 type room
step chooseleaf firstn 2 type host
step emit
}
rule ssd {
ruleset 1
type replicated
min_size 1
max_size 4
step take ssd
step choose firstn 2 type room
step chooseleaf firstn 2 type host
step emit
}
Bob
On Mon, Jul 11, 2016 at 9:19 AM, Gregory Farnum <[email protected]> wrote:
> I'm not looking at the docs, but I think you need an "emit" statement
> after every choose.
> -Greg
>
>
> On Monday, July 11, 2016, George Shuklin <[email protected]> wrote:
>
>> Hello.
>>
>> I want to try CRUSH rule with following idea:
>> take one OSD from root with SSD drives (and use it as primary).
>> take two OSD from root with HDD drives.
>>
>> I've created this rule:
>>
>> rule rule_mix {
>> ruleset 2
>> type replicated
>> min_size 2
>> max_size 10
>> step take ssd
>> step chooseleaf firstn 1 type osd
>> step take hdd
>> step chooseleaf firstn -1 type osd
>> step emit
>> }
>>
>> But I think I done something wrong - all PG are undersized+degraded (I
>> use 'size 3', have 2 SSD OSD and 5 HDD OSD).
>>
>> My noobie questions:
>>
>> 1) Can I use multiple 'take' steps in the single rule?
>> 2) How many emit I should/may use per rule?
>> 3) Is this a proper way to describe such logic? Or it should be done
>> differently? (How?)
>>
>> Thanks.
>> _______________________________________________
>> ceph-users mailing list
>> [email protected]
>> http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
>>
>
> _______________________________________________
> ceph-users mailing list
> [email protected]
> http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
>
>
_______________________________________________
ceph-users mailing list
[email protected]
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
