[Gluster-users] Migrating Hard Drives Containing Bricks to New Server

[Ken Schweigert]

I am learning GlusterFS in a test environment and trying out different
scenarios of when things go wrong and how to fix it.  Currently I have
4 VirtualBox instances running GlusterFS with each instance holding 2
attached virtual hard drives.  The instances are configured as
distributed-replicated and everything runs fine.  I can add new files
via a glusterfs mountpoint started from fstab and things look good.
Now I want to simulate a hardware failure, something like the CPU
fried but the drives are okay, and migrate the attached hard drives to
a new instance.

I've seen in the documentation on how to migrate bricks to another
node, but this requires the instance to still be online.  I'd like to
figure out how to move the hard drives out of one machine and put them
in to another and bring full functionality back to the cluster.

One thing that I tried, but failed, was cloning the existing instance,
bringing down the original instance, launching the cloned image with
the virtual hard drives attached, but gluster doesn't start and I
can't 'peer probe' in to the trusted pool.  I suspect it has something
to do with a unique identifier for each node in the cluster and
conflicting with that.

Many thanks for pushes and pointers in the right direction.

-ken
-- 
Have a nice day ... unless you've made other plans.