xfs: validate inode numbers in file handles correctly--NFS Stale File Handle Again

[Christoph Hellwig]

On Thu, Dec 29, 2011 at 05:19:55PM +0800, yangguoquan wrote:
> 
> NFS Stale File Handle Again, But this happens when mount operation not during accessing file or directory.
>  
> My /etc/exports file:
>  
> /mnt/storage_pool/testnfs00     *(rw,sync,no_root_squash)
> /mnt/storage_pool/testnfs01     *(rw,sync,no_root_squash)
> /mnt/storage_pool/testnfs02     *(rw,sync,no_root_squash)
> /mnt/storage_pool/testnfs03     *(rw,sync,no_root_squash)
> /mnt/storage_pool/testnfs04     *(rw,sync,no_root_squash)
> /mnt/storage_pool/testnfs05     *(rw,sync,no_root_squash)
> /mnt/storage_pool/testnfs06     *(rw,sync,no_root_squash)
> /mnt/storage_pool/testnfs07     *(rw,sync,no_root_squash)
> /mnt/storage_pool/testnfs08     *(rw,sync,no_root_squash)
> /mnt/storage_pool/testnfs09     *(rw,sync,no_root_squash)

Are these mount points?  If not take a look at

	http://xfs.org/index.php/XFS_FAQ#Q:_Why_doesn.27t_NFS-exporting_subdirectories_of_inode64-mounted_filesystem_work.3F