On 2014/07/22, 8:05 AM, "Kumar, Amit" <ahkumar(a)mail.smu.edu> wrote:
Makes sense!
I will take this as a discussion item and see what we finalize on this.
Thank youall for your inputs.
It might not be too hard to add "lfs quota -p" to print only quota usage
for a user on that specific pool. There is already "lfs df -p" to print
space free/used for a pool (see how lfs_df->mntdf->llapi_search_ost(...,
poolname, ...) is used to filter out the OSTs to print), and something
similar could be done in print_obd_quota().
That said, I'm not a Lustre quota expert so there may be some pitfalls to
this approach that I'm missing, and this isn't something that we would be
working on soon, but if someone on the list wants to give this a try it
would be a nice little project.
Cheers, Andreas
-----Original Message-----
From: Mohr Jr, Richard Frank (Rick Mohr) [mailto:rmohr@utk.edu]
Sent: Monday, July 21, 2014 4:15 PM
To: Kumar, Amit
Cc: Dilger, Andreas; Robin Humble; hpdd-discuss(a)lists.01.org
Subject: Re: [HPDD-discuss] Lustre Pools vs Multiple Lustre File System
recommendation
On Jul 21, 2014, at 4:04 PM, "Kumar, Amit" <ahkumar(a)mail.smu.edu> wrote:
> A follow up question based on your input:
> I was hoping to use quota just so we could keep tabs on the usage:
>Using OST pools, do we have to rule out the possibility of using quota
>feature all together ? I was hoping to accomplish the following, would
>quota's work at any level of the hierarchy?
>
> /lustre/{sas,sata}/<user>
You can still use quotas with ost pools, but the quotas will apply to the
aggregate usage across both the sas and sata pools. If you want to
enforce separate quotas for sas and sata, you would need different file
systems.
--
Rick Mohr
Senior HPC System Administrator
National Institute for Computational Sciences
http://www.nics.tennessee.edu
_______________________________________________
HPDD-discuss mailing list
HPDD-discuss(a)lists.01.org
https://lists.01.org/mailman/listinfo/hpdd-discuss
Cheers, Andreas
--
Andreas Dilger
Lustre Software Architect
Intel High Performance Data Division