[Sheepdog] [PATCH 1/2] deleting data objects of a vdi before deleting the inode

Christoph Hellwig hch at infradead.org
Tue May 1 13:54:41 CEST 2012


On Tue, May 01, 2012 at 07:49:57PM +0800, Liu Yuan wrote:
> 
> Nope, it fails because of *your* last wrong patch (cleanup nr_copies),
> that is why I came up with 'sheep: fix nr_copies in vdi.c', which aims
> to restore previous logic. [I haven't merged it yet, waiting for your
> agreement]

Can you please do the following from your checkout of the repository
and confirm the results for me?

git reset --hard dabb5200202f2225e69002c055afba97ee5cf73a
make

./sheep/sheep -f -c local -o -z 0 -p 7000 /mnt/sheepdog/7000 &
./sheep/sheep -f -c local -o -z 1 -p 7001 /mnt/sheepdog/7001 &
./sheep/sheep -f -c local -o -z 2 -p 7002 /mnt/sheepdog/7002 &

./collie/collie cluster format
./collie/collie vdi create test 10G
./collie/collie vdi write test
<enter something and ctrl+d>
./collie/collie vdi delete test
./collie/collie vdi list

For this trivial test case I still get the "test" vdi reported.  I'm
also pretty sure that I gave exactly these steps a few mails earlier.




More information about the sheepdog mailing list