On Mon, May 11, 2015 at 1:59 PM, Timofey Titovets <nefelim4ag@xxxxxxxxx> wrote:
> Year, np
>
> # begin crush map
> tunable choose_local_tries 0
> tunable choose_local_fallback_tries 0
> tunable choose_total_tries 50
> tunable chooseleaf_descend_once 1
> tunable chooseleaf_vary_r 1
> tunable straw_calc_version 1
> tunable allowed_bucket_algs 54
>
> # devices
> device 0 osd.0
> device 1 osd.1
> device 2 osd.2
> device 3 osd.3
> device 4 osd.4
> device 5 osd.5
> device 6 osd.6
> device 7 osd.7
> device 8 osd.8
> device 9 osd.9
> device 10 osd.10
> device 11 osd.11
>
> # types
> type 0 osd
> type 1 host
> type 2 chassis
> type 3 rack
> type 4 row
> type 5 pdu
> type 6 pod
> type 7 room
> type 8 datacenter
> type 9 region
> type 10 root
>
> # buckets
> host srv-lab-ceph-node-03 {
> id -6 # do not change unnecessarily
> # weight 2.720
> alg straw
> hash 0 # rjenkins1
> item osd.11 weight 0.910
> item osd.8 weight 0.450
> item osd.9 weight 0.450
> item osd.10 weight 0.910
> }
> host srv-lab-ceph-node-01 {
> id -4 # do not change unnecessarily
> # weight 1.940
> alg straw
> hash 0 # rjenkins1
> item osd.0 weight 0.450
> item osd.1 weight 0.450
> item osd.2 weight 0.130
> item osd.3 weight 0.910
> }
> host srv-lab-ceph-node-02 {
> id -5 # do not change unnecessarily
> # weight 2.720
> alg straw
> hash 0 # rjenkins1
> item osd.4 weight 0.450
> item osd.5 weight 0.450
> item osd.7 weight 0.910
> item osd.6 weight 0.910
> }
> rack rack3 {
> id -3 # do not change unnecessarily
> # weight 7.380
> alg straw
> hash 0 # rjenkins1
> item srv-lab-ceph-node-03 weight 2.720
> item srv-lab-ceph-node-01 weight 1.940
> item srv-lab-ceph-node-02 weight 2.720
> }
> room k402b {
> id -2 # do not change unnecessarily
> # weight 7.380
> alg straw
> hash 0 # rjenkins1
> item rack3 weight 7.380
> }
> root default {
> id -1 # do not change unnecessarily
> # weight 7.380
> alg straw
> hash 0 # rjenkins1
> item k402b weight 7.380
> }
>
> # rules
> rule replicated_ruleset {
> ruleset 0
> type replicated
> min_size 1
> max_size 10
> step take default
> step choose firstn 0 type room
> step choose firstn 0 type rack
> step choose firstn 0 type host
> step chooseleaf firstn 0 type osd
> step emit
> }
>
> # end crush map
>
> 2015-05-11 12:53 GMT+03:00 Ilya Dryomov <idryomov@xxxxxxxxx>:
>> On Mon, May 11, 2015 at 12:37 PM, Timofey Titovets <nefelim4ag@xxxxxxxxx> wrote:
>>> Sorry list, i just can't find how send bugs to tracking system -_-
>>>
>>> How reporduce:
>>> 1. Use crush rule
>>> rule replicated_ruleset {
>>> ruleset 0
>>> type replicated
>>> min_size 1
>>> max_size 10
>>> step take default
>>> step choose firstn 0 type room
>>> step choose firstn 0 type rack
>>> step choose firstn 0 type host
>>> step chooseleaf firstn 0 type osd
>>> step emit
>>> }
>>> 2. inject into cluster
>>> 3. do:
>>> rbd map <rbd name>
>>>
>>> crushtool passed test successful, but if you inject map into cluster
>>> (i use 0.94.1)
>>> All kernel rbd Clients who use rbd map, immadentely crush.
>>> i don't know which kernel not affected, but i catch it with 4.0.1
>>>
>>> With long stucktrace in kernel and message like "Hard lookup detected"
>>> Sorry but i realy have a problems with find logs =_=
>>
>> Can you paste your entire crushmap?
I think this is http://tracker.ceph.com/issues/9492. It's fixed in
kernel 4.1, but isn't marked for backporting. I'll see if it can be
backported to a few older kernels.
Thanks,
Ilya
--
To unsubscribe from this list: send the line "unsubscribe ceph-devel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
