yes this is what I meant although did not convey in the best terms. This is working fine in testing but my main concern was having multiple volumes i/o hitting the same disks as a potential bottleneck. What I think I have settled on is not to subpartition the LVM/Physical disks but create seperate bricks from seperate RAID volumes on each server such that I can isolate I/O for many clients without losing the ability to scale the filesystem--one volume per brick, multiple bricks.
thanks guys
On Mon, Sep 21, 2015 at 10:59 PM, Dave Warren <davew@xxxxxxxxxxxx> wrote:
On 2015-09-21 21:17, Atin Mukherjee wrote:
On 09/22/2015 02:07 AM, Gluster Admin wrote:
>Gluster users,This is not possible in Gluster. A brick (partition) is considered to be
>
>We have a multiple node setup where each server has a single XFS brick
>(underlying storage is hardware battery backed raid6). Are there any
>issues creating multiple gluster volumes using the same underlying
>bricks from a performance or management standpoint?
the smallest unit of the break down of a volume.
While true, I wonder if the original poster meant to ask if they can have multiple bricks belonging to different gluster volumes on the same file system volume?
In this case, the answer is yes, one could mount a file system, and create separate bricks which belong to different gluster volumes.
--
Dave Warren
http://www.hireahit.com/
http://ca.linkedin.com/in/davejwarren
_______________________________________________ Gluster-users mailing list Gluster-users@xxxxxxxxxxx http://www.gluster.org/mailman/listinfo/gluster-users