[Gluster-devel] Glusterd: A New Hope

[Ian Latter]

> How does it handle online reconfiguration (e.g. our "volume set")? 

Not sure which feature you're referring too;
  http://gluster.org/community/documentation/index.php/Gluster_3.1:_Setting_Volume_Options


> Also, how does a client find a backup DIR if the primary is down?  DNS aliases?

A quick search indicates that this should have been made available in version 1.4 (I assume it was) when the clients were given the means to discover additional DIR nodes from the original;
  http://code.google.com/p/xtreemfs/issues/detail?id=224


If this thing compiles to a binary via gcj then I'll be able to tell you when I get it running and start embedding it.  


> I think we can do better than that.

I'm also keen to understand if their replication is disk level (no use to me) or volume level (the reason why I started using GlusterFS, right before it was taken out again).  

If I have to choose, I will take integrated volume replication over dynamic reconfiguration.  Everything Gluster does for me at the moment can be replaced with hardware (sub $100) and an NFS daemon.


----- Original Message -----
>From: "Jeff Darcy" <jdarcy at redhat.com>
>To: "Ian Latter" <ian.latter at midnightcode.org>
>Subject:  Re: [Gluster-devel] Glusterd: A New Hope
>Date: Tue, 26 Mar 2013 17:51:57 -0400
>
> On 03/26/2013 05:36 PM, Ian Latter wrote:
> > From a user perspective the "cluster" establishment is done via text file configuration to direct nodes to network services;
> >   http://www.xtreemfs.org/quickstart_distr.php
> 
> Looks like an alternative worth considering.  How does it handle online
> reconfiguration (e.g. our "volume set")?  Also, how does a client find a
> backup DIR if the primary is down?  DNS aliases?  I think we can do
> better than that.
> 
> 
> 


--
Ian Latter
Late night coder ..
http://midnightcode.org/