[Gluster-users] Some questions about theoretical gluster failures.

Joe Landman landman at scalableinformatics.com
Wed Oct 26 02:09:43 UTC 2011

On 10/25/2011 10:01 PM, Harry Mangalam wrote:

> - what happens in a distributed system if a node goes down? Does the
> rest of the system keep working with the files on that brick unavailable
> until it comes back or is the filesystem corrupted? In my testing, it
> seemed that the system indeed kept working and added files to the
> remaining systems, but that files that were hashed to the failed volume
> were unavailable (of course).

This is basically it.

> - is there a head node? the system is distributed but you're mounting a

Only if you mount via nfs, though technically you can mount it from any 
server.  If you mount via gluster client, just point it at any of the 
servers.  In the nfs case, if the mount server goes away, so does access 
unless you remount.  In the glusterfs case, if the mount server goes 
away, the other servers can continue talking with the client.

> specific node for the glusterfs mount - if that node goes down, is the
> whole filesystem hosed or is that node reference really a group
> reference and the gluster filesystem continues with the loss of that
> node's files? ie can any gluster node replace a mountpoint node and does
> that happen transparently? (I haven't tested this).

You can mount from any node, but the mount target has to be specifically 
unmounted/remounted under nfs (umount -l is your friend).  With 
GlusterFS client its less of an issue.

This said, I don't know many people using the nfs client version.  I 
haven't tested 3.2.4's server, but through 3.2.3, we can crash the NFS 
server with a moderate load.

> - can you intermix distributed and mirrored volumes? This is of

Not sure what you mean by intermix ... but yes, you can have multiple 
(many) volumes of all different types coming from the same units on 
different volume names.

> particular interest since some of our users want to have replicated data
> and some don't care.
> Many thanks
> hjm
> --
> Harry Mangalam - Research Computing, OIT, Rm 225 MSTB, UC Irvine
> [ZOT 2225] / 92697 Google Voice Multiplexer: (949) 478-4487
> MSTB Lat/Long: (33.642025,-117.844414) (paste into Google Maps)
> --
> This signature has been OCCUPIED!
> _______________________________________________
> Gluster-users mailing list
> Gluster-users at gluster.org
> http://gluster.org/cgi-bin/mailman/listinfo/gluster-users

Joseph Landman, Ph.D
Founder and CEO
Scalable Informatics, Inc.
email: landman at scalableinformatics.com
web  : http://scalableinformatics.com
phone: +1 734 786 8423 x121
fax  : +1 866 888 3112
cell : +1 734 612 4615

More information about the Gluster-users mailing list