So, I did find this:
http://community.gluster.org/q/quota-limit-not-respected/
in which the user answered their own question saying a remount on
clients was needed...which...IF that is the case..is a big deal
breaker for us. We've got a few thousand systems mounting this volume,
running long running HPC jobs, so remounting when we provision new
space on this isn't an option. Again, IF this is teh case, anyway to
get clients to pick up the new quotas without a unmount/remount? Some
sort of volfile refesh?
--
Matthew Nicholson
Research Computing Specialist
Harvard FAS Research Computing
matthew_nicholson at harvard.edu
On Mon, Mar 11, 2013 at 9:43 AM, Matthew Nicholson
<matthew_nicholson at harvard.edu> wrote:> So, running a dist-replica 5x2 volume, quotas are on:
>
> features.quota: on
>
> however, I've got a few areas where users have gone right over there
> quota like it wasn't there at all:
>
> root at sum1-gstore01 Lukin_Lab]# gluster volume quota gstore list
> path limit_set size
>
----------------------------------------------------------------------------------
> /Labs/Lab2 15TB 15.1TB
> /Labs/Lab1 20TB 31.1TB
>
> Any idea where to start with this? is there a re-calc that can be
> fired off or a crawl? I badly need to limit these directories.
>
>
> --
> Matthew Nicholson
> Research Computing Specialist
> Harvard FAS Research Computing
> matthew_nicholson at harvard.edu