CEPH_FEATURE_OSD_CACHEPOOL always required when osd cache pool exist ?

CEPH_FEATURE_OSD_CACHEPOOL always required when osd cache pool exist ?

[Laurent Barbe]

Hi all,

About 0.75, when I created cache pool, kernel client like rbd or cephfs 
make error "feature set mismatch". (Even if the device is not on the 
cached pool)

$ rbd map rbd/myrbd    => OK

$ ceph osd pool create otherpool1 100 100
$ ceph osd pool create otherpool2 100 100
$ ceph osd tier add otherpool1 otherpool2
$ ceph osd tier cache-mode otherpool2 writeback

$ rbd map rbd/myrbd
kern log >> libceph: mon2 10.2.0.90:6789 feature set mismatch, my 
4a042a42 < server's 84a042a42, missing 800000000

Does the commit 
https://github.com/ceph/ceph/commit/431a13eb370310a383abceeb509fe204a519e087
fix this ?


Regards,
Laurent Barbe

Re: CEPH_FEATURE_OSD_CACHEPOOL always required when osd cache pool exist ?

[Ilya Dryomov]

On Tue, Jan 21, 2014 at 10:28 AM, Laurent Barbe <laurent@ksperis.com> wrote:
> Hi all,
>
> About 0.75, when I created cache pool, kernel client like rbd or cephfs make
> error "feature set mismatch". (Even if the device is not on the cached pool)
>
> $ rbd map rbd/myrbd    => OK
>
> $ ceph osd pool create otherpool1 100 100
> $ ceph osd pool create otherpool2 100 100
> $ ceph osd tier add otherpool1 otherpool2
> $ ceph osd tier cache-mode otherpool2 writeback
>
> $ rbd map rbd/myrbd
> kern log >> libceph: mon2 10.2.0.90:6789 feature set mismatch, my 4a042a42 <
> server's 84a042a42, missing 800000000
>
> Does the commit
> https://github.com/ceph/ceph/commit/431a13eb370310a383abceeb509fe204a519e087
> fix this ?

No, I don't think so.  I might be wrong, but I think the idea is to
require OSD_CACHEPOOL feature if any tiering relationships are present.
Work on the kernel client is on-going, tiering support should arrive in
3.14.

Thanks,

                Ilya

Re: CEPH_FEATURE_OSD_CACHEPOOL always required when osd cache pool exist ?

[Laurent Barbe]

Thank you for your reply Ilya.

However, this means that if we want to activate a pool with cache, it 
will be necessary to upgrade the kernel of all clients even if they do 
not use this feature. :(

Regards,

Laurent Barbe


Le 21/01/2014 10:46, Ilya Dryomov a écrit :
> On Tue, Jan 21, 2014 at 10:28 AM, Laurent Barbe <laurent@ksperis.com> wrote:
>> Hi all,
>>
>> About 0.75, when I created cache pool, kernel client like rbd or cephfs make
>> error "feature set mismatch". (Even if the device is not on the cached pool)
>>
>> $ rbd map rbd/myrbd    => OK
>>
>> $ ceph osd pool create otherpool1 100 100
>> $ ceph osd pool create otherpool2 100 100
>> $ ceph osd tier add otherpool1 otherpool2
>> $ ceph osd tier cache-mode otherpool2 writeback
>>
>> $ rbd map rbd/myrbd
>> kern log >> libceph: mon2 10.2.0.90:6789 feature set mismatch, my 4a042a42 <
>> server's 84a042a42, missing 800000000
>>
>> Does the commit
>> https://github.com/ceph/ceph/commit/431a13eb370310a383abceeb509fe204a519e087
>> fix this ?
>
> No, I don't think so.  I might be wrong, but I think the idea is to
> require OSD_CACHEPOOL feature if any tiering relationships are present.
> Work on the kernel client is on-going, tiering support should arrive in
> 3.14.
>
> Thanks,
>
>                  Ilya
>
--
To unsubscribe from this list: send the line "unsubscribe ceph-devel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html