On Mon March 13, jens@pinguin.conetix.de wrote:
> On Mon, Mar 13, 2000 at 10:15:56AM +1100, Neil F. Brown wrote:
>
> > Two problems. 1/ nfs doesn't see mount points unless you export both the
> > parent and the child filesystems, and flags the child filesystem as
> > "nohide". 2/ knfsd won't export filesystems that don't live on a local
> > disc - at least, not yet. You will have to use unfsd - the Universal
> > Usermode NFS daemon to re-export SMB filesystems.
>
> Thank you. I don't really need this feature, it would only be "nice to
> have". Is this behaviour intentional (wrt security or something) or just
> not (cleanly) implementable?
I don't know enough about SMB to be certain.
The important thing is that there needs to be a way to map between a
"filehandle" of limited size (32 bytes for NFSv2) and a file. With
ext2fs and similar, the inode number is used. I don't know if SMB
supports a similar sort of handle.
The UNFSD does it by using a hash of the path name and keeping a
cache. It works 98% of the time.
NeilBrown
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Wed Mar 15 2000 - 21:00:27 EST