Hi Jiri,

I have checked this on a dist/dist-rep  volume and this is not a problem for me.

Requesting you to run " find . | xargs stat" on the mount point and once this 
is finished collect the quota information from the all servers. Also, please 
share your setup information.

Thanks,
Saurabh

Hi Jiri,

If a limit of 600GB is set on / then Quota translator will allow you to
store 600GB of data under / and any writes after that will not be allowed.
The same rule applies to replicate aswell, here it will consider only one of
the copies of the data for calculating the sizes.


Junaid


On Thu, Jun 16, 2011 at 3:04 AM, Jiri Lunacek <[email protected]>wrote:

> Hi all,
>
> I am having trouble understanding how quota works, or how to successfully
> set it up and use it. Some best practices list of quota feature would be
> great.
>
> My problem is this. We have a volume with replicate 2, 2 bricks, quota
> enabled, usage limit on / set to 600GB, gluster packages 3.2.1.
>
>  gluster volume quota virtual_782 list
> on storage server 1 shows:
>
>        path              limit_set          size
>
> ----------------------------------------------------------------------------------
> /                    644245094400         277136318464
>
> on storage server 2 shows:
>
>        path              limit_set          size
>
> ----------------------------------------------------------------------------------
> /                    644245094400         277160824832
>
> on machine, which has the volume mounted (which is also a node in the same
> cluster holding no bricks) shows:
>
>        path              limit_set          size
>
> ----------------------------------------------------------------------------------
> /                         600GB
>
>
> writes to the share fail with: -1 (Disk quota exceeded)
>
> I am confused what the quota really means. From our setup it seems that the
> quota translator sums used size from all servers and compares it to the set
> quota.
> I'd expect it (and it seems like that from the docs) that it should allow
> (with replica 2) 600GB on each server.
>
> If this is expected behavior I think this is a subject for documentation.
>
> TIA for any clear information in this matter.
> Jiri
> _______________________________________________
> Gluster-users mailing list
> [email protected]
> http://gluster.org/cgi-bin/mailman/listinfo/gluster-users
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: 
<http://gluster.org/pipermail/gluster-users/attachments/20110618/2492b32b/attachment.html>

------------------------------

_______________________________________________
Gluster-users mailing list
[email protected]
http://gluster.org/cgi-bin/mailman/listinfo/gluster-users

Reply via email to