Gluster users - Apr 2010 - Self-heal

Hello,

 

I am in the process of setting up a new 2 server cluster using
GlusterFS.  I have one box with roughly 8 TB of data that I will be
installing GlusterFS.  The secondary box will also have GLusterFS
installed and I would like the main box to self-heal/replicate all the
data over to the secondary box?  Is this feasible with this amount of
data?  Any pointers/suggestions on the most efficient way to handle this
would be greatly appreciated.

 

Thank you.

On Apr 7, 2010, at 7:02 PM, Brad Alexander wrote:
> Hello,
> 
> I am in the process of setting up a new 2 server cluster using
> GlusterFS.  I have one box with roughly 8 TB of data that I will be
> installing GlusterFS.  The secondary box will also have GLusterFS
> installed and I would like the main box to self-heal/replicate all the
> data over to the secondary box?  Is this feasible with this amount of
> data?  Any pointers/suggestions on the most efficient way to handle this
> would be greatly appreciated.

Self-heal would work, but you should probably just use rsync in this case
to sync the backends before starting GlusterFS. Self-heal will be overkill
because it holds locks and so on to ensure no one else writes to the same
file while the healing is happening.

------------------------------
Vikas Gorur
Engineer - Gluster, Inc.
+1 (408) 770 1894
------------------------------