[Gluster-users] Gluster in a cluster

[Jerome]

Dear David

On 18/11/2012 21:21, David Duncan wrote:
> I have worked with NPACI and Platform rocks somet and it is typical to
> have nodes reinstall when they fail to boot, but your configuration
> seems to reinstall on any reboot. Have you worked with the rocks
> community to modify your configuration?

The node reinstall in case of severy fault, or because of some 
adminsttration tasks. And in this cases, any node include in a gluster 
volume lost all of the configuration, and put the gluster volume in a 
unstable state...

>
> Did you create a new appliance type for your storage nodes? How do you
> distinguish your storage nodes from your compute nodes? Is it possible
> that you might be able to maintain the storage nodes distinctly from
> your cluster and then add a script for mounting the volumes to your
> config.xml?

I'm now searching a such way for my cluster: to implement a special 
appliance to create some gluster nodes. This nodes will be dedicated to 
storage, and be accesible for computing uniquely for some testing.
The mounting is resolved by a automonting on all the cluster.

best regards


>
>
> On Thu, Nov 15, 2012 at 12:25 PM, Brian Candler <B.Candler at pobox.com
> <mailto:B.Candler at pobox.com>> wrote:
>
>     On Thu, Nov 15, 2012 at 09:30:52AM -0600, Jerome wrote:
>      > My problem is when a node reboot accidentaly, or for some
>      > administration task: the node reinstall itself, and the gluster
>      > volume begin to fail.... I detect taht the UUID of a machine is
>      > generated during the instalation, so i develop some script to get
>      > back the original UUID of the node. Despote this, the node could not
>      > get back in the volume. I miss some special task to do. So, it is
>      > possible to do a such system with gluster?
>
>     I think you're asking about this:
>     http://gluster.org/community/documentation/index.php/Gluster_3.2:_Brick_Restoration_-_Replace_Crashed_Server
>     (with minor modifications for gluster 3.3)
>
>     However the gluster nodes really should have persistent storage of
>     UUID and
>     volume configuration. If every time your machine reboots it loses its
>     /var/lib/glusterd/ directory then this is not good.
>     _______________________________________________
>     Gluster-users mailing list
>     Gluster-users at gluster.org <mailto:Gluster-users at gluster.org>
>     http://supercolony.gluster.org/mailman/listinfo/gluster-users
>
>
>
>
> --
> David Duncan
> http://about.me/davdunc
> Registered Linux User #279425
> http://counter.li.org


-- 
-- Jérôme
L'adulte ne croit pas au père Noël. Il vote.
	(Pierres Desproges)