[sheepdog] [PATCH] dog: add command to check and repair cluster

[Liu Yuan]

On Mon, Aug 19, 2013 at 05:56:19PM +0900, MORITA Kazutaka wrote:
> From: MORITA Kazutaka <morita.kazutaka at lab.ntt.co.jp>
> 
> This patch adds a 'dog cluster check' command to check all the
> objects' consistency.  While 'dog vdi check' is used when VM stops
> unexpectedly because of the gateway failure, 'dog cluster check' is
> expected to be used for fully checking of cluster status, e.g., after
> total node failure happens or sheepdog hangs because of bugs.
> 
> With the current implementation, running 'dog cluster check' is same
> as doing 'dog vdi check' against all the vdi images.  However, this is
> usually very wasteful because vdis can share the same objects
> especially when there are many snapshots.  I think of improving it to
> skip redundant checks.
> 
> I'm planning to add other check commands later:
> 
>  - 'collie cluster check reference' : fix reference counts for object reclaim
>  - 'collie cluster check epoch'     : fix a broken epoch file
> 
> Signed-off-by: MORITA Kazutaka <morita.kazutaka at lab.ntt.co.jp>

Applied thanks

Yuan