[Gluster-users] Replicated Volume (mirror) on 17 nodes.

Ravishankar N ravishankar at redhat.com
Thu Feb 25 05:07:33 UTC 2016


On 02/25/2016 03:09 AM, Simone Taliercio wrote:
> Hi all :)
>
> I would need soon to create a pool of 17 nodes. Each node requires a 
> copy of the same file "locally" so that can be accessed from the 
> deployed application.
>
> * Do you see any performance issue in creating a Replica Set on 17 
> nodes ? Any best practice that I should follow ?
>
> * An other question: in case there's no problem then on this line
>
> gluster volume create gv0 replica* 2* server1:/data/brick1/gv0 
> server2:/data/brick1/gv0
>
> do i need to provide 17 instead of 2 ?

I don't know what your use case is but I don't think you want to create 
so many replicas. Why not just create a glusterfs volume  of replica-2 
or replica-3 or even an arbiter volume, and mount them on all 17 nodes?
I think replica 4 is what some people in the community have used at the 
max but even that is an overkill IMO.

Regards,
Ravi
>
> Sorry for my basic questions.
>
> Thanks a lot for your support!
>
> Simone
>
>
> _______________________________________________
> Gluster-users mailing list
> Gluster-users at gluster.org
> http://www.gluster.org/mailman/listinfo/gluster-users


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.gluster.org/pipermail/gluster-users/attachments/20160225/72eeb777/attachment.html>


More information about the Gluster-users mailing list