Hi, I've recently had dataloss on an XFS (inode64) glusterfs (3.2.5) NFS exported file system. I was using the gluster NFS server. On the XFS FAQ page, they have this: Q: Why doesn't NFS-exporting subdirectories of inode64-mounted filesystem work? The default fsid type encodes only 32-bit of the inode number for subdirectory exports. However, exporting the root of the filesystem works, or using one of the non-default fsid types (fsid=uuid in /etc/exports with recent nfs-utils) should work as well. (Thanks, Christoph!) Is this still a factor under Gluster NFS server? Gerald
