On 22/04/2014 08:56, Gregory Farnum wrote:
> On Monday, April 21, 2014, Loic Dachary <[email protected] 
> <mailto:[email protected]>> wrote:
> 
>     Hi,
> 
>     I would like to allow users to create,use and delete RBD volumes, up to X 
> GB, from a single pool. The user is a Debian GNU/Linux box using krbd. The 
> sysadmin of the box is not trusted to have unlimited access to the Ceph 
> cluster but (s)he is not malicious either. Permissions and quota are 
> safeguards to prevent mistake.
> 
>     While it seems possible to grant access to a single pool to a given cephx 
> client with
> 
>        ceph-authtool -n client.foo --cap osd 'allow rwx pool=customer-pool
> 
>     and the cap parser suggests even more flexibility
> 
>        https://github.com/ceph/ceph/blob/master/src/mon/MonCap.cc#L329
> 
>     the documentation states that it should not be done
> 
>        
> http://ceph.com/docs/master/rados/operations/auth-intro/#cephx-limitations
> 
>     Suggestions about how to approach this use case are most welcome :-)
> 
>     Cheers
>     --
>     Loïc Dachary, Artisan Logiciel Libre
> 
> 
> That looks fine to me. The documentation is just pointing out that cephx keys 
> are per-host, not per-user.
> -Greg 
> 

Cool :-)

Here is how it could go then

# use import because ceph auth get-or-create does not allos for --set-uid
* ceph-authtool /tmp/keyring --create-keyring --name client.me --gen-key 
--set-uid 123 --cap osd 'allow rwx pool=foobar'
* ceph auth import -i /tmp/keyring
# creating the pool via client.me will implicitly set auid 
https://github.com/ceph/ceph/blob/dumpling/src/mon/OSDMonitor.cc#L2551
* ceph --name client.me osd pool create foobar 12
# record the desired quota but it is not enforced
* ceph osd pool set-quota foobar max_bytes 1T

On a daily basis a cron job uses ceph df to act on the quotas.

-- 
Loïc Dachary, Artisan Logiciel Libre

Attachment: signature.asc
Description: OpenPGP digital signature

_______________________________________________
ceph-users mailing list
[email protected]
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com

Reply via email to