[Gluster-devel] split brain: how should it be cured?

[Pranith Kumar Karampuri]

They are not in split-brain, I responded to your previous mail just now giving examples of split-brain.

Avati,
   What should be the behavior when there is metadata split-brain?.

P.S. I think I forgot to CC avati in prev mail. I think he will respond this time.
Pranith.
----- Original Message -----
From: "Emmanuel Dreyfus" <manu at netbsd.org>
To: "Emmanuel Dreyfus" <manu at netbsd.org>
Cc: "Pranith Kumar Karampuri" <pkarampu at redhat.com>, gluster-devel at nongnu.org
Sent: Wednesday, June 20, 2012 3:10:57 PM
Subject: Re: [Gluster-devel] split brain: how should it be cured?

On Tue, Jun 19, 2012 at 04:01:16AM +0200, Emmanuel Dreyfus wrote:
> Um, now both bricks have the split brain flag. How removing the file on
> one brick will make it recreated on the other one, but the split brain
> flag is sticky!

I was about to resume investigation more than a day later, and the
offending file can now be readen without getting EIO. The split
brain flags are stil lthere, though (see below). Does that makes
sens, or did I overlooked something or got confused? Anyone can 
explain me how this thing is supposed to work?

On brick1, yesterday:
trusted.afr.pfs-client-0   00 00 00 00 00 00 00 03 00 00 00 00
trusted.afr.pfs-client-1   00 00 00 00 00 00 00 01 00 00 00 00
Today:
trusted.afr.pfs-client-0   00 00 00 00 00 00 00 00 00 00 00 00
trusted.afr.pfs-client-1   00 00 00 00 00 00 00 01 00 00 00 00

On brick2, yesterday:
trusted.afr.pfs-client-0   00 00 00 00 00 00 00 00 00 00 00 00
trusted.afr.pfs-client-1   00 00 00 00 00 00 00 01 00 00 00 00
Today:
trusted.afr.pfs-client-0   00 00 00 00 00 00 00 03 00 00 00 00
trusted.afr.pfs-client-1   00 00 00 00 00 00 00 00 00 00 00 00

-- 
Emmanuel Dreyfus
manu at netbsd.org