[Gluster-users] XFS and MD RAID

[Brian Candler]

On Mon, Sep 10, 2012 at 06:43:44PM +0100, Brian Candler wrote:
> It has been running fine for the last 7 hours or so. I have purposely sent
> some dd reads to the two failed drives in the server - I see the errors on
> those drives in dmesg, but activity on the remaining drives has not been
> affected.  (The remaining drives are in an md raid0 array loaded by running
> 4 instances of bonnie++)
>  
> Tomorrow, when I have physical access to the machine, I'll try hot-plugging
> the two failed drives as well.

Good news:

- hot plugging the two bad drives while I/O access is going on to the other
drives is fine.  The other drives continue to work without a hitch.

- Even removing the two bad drives while dd'ing from them was fine. I/O to
the other drives was also unaffected.

So this patch looks good to me. I hope it can find its way to the production
Ubuntu kernel soon.

Regards,

Brian.