Subject: Re: [vserver] cluster file system and vserver
From: Martin Fick <mogulguy@yahoo.com>
Date: Tue, 15 Jul 2008 15:40:03 -0700 (PDT)

--- On Tue, 7/15/08, Edward Capriolo <edlinuxguru@gmail.com> wrote:
> 
> GlusterFS seems really impressive. I wish I had more time
> for it. Seems like you can do just about anything with it. 
> Network mirrors/ network stripes/ file replication. I 
> never benchmarked it. Once I figured out the install it 
> seems to be pretty easy.

Funny, I found this install very easy, one of the most 
impressive parts of glusterfs.

 
> DRBD has its place as a more ACTIVE/PASSIVE ACTIVE/ACTIVE
> fail over solution with very low level replication. It is 
> not going to scale past three servers.

Well, with DRBD under glusterfs volumes, it will scale as
well as any non AFR glusterfs setup (since it is glusterfs).

Think like this:


            Gluster
             UNIFY
           /  |    \
      Vol1   Vol2... Voln
        |      |       |
      DRBD   DRBD...  DRBD
      |  |   |  |     |  |
      1A 1B  2A 2B... nA nB

The real beauty of glusterfs with DRBD and vservers is the 
simple ability to mount a glusterfs volume on the same 
machine that exports it.  This means that you can have a
two node cluster and simply failover with DRBD from one
node to the other but you can run you vservers on either
node.  This cannot be done safely with NFS (I tried, it
resulted in intermittent lockups), although I do no know
one (the) major telco provider which does exactly this 
(DRBD+NFS mount share on sever, no vservers), on one of 
its primary platforms, go figure! :)

-Martin