Last time I asked about this, about a year ago IIRC, the answer seemed to be
"Well, don't do that then".
So I didn't - I'm still using the user-space server, which is a pity, as I
could have done with the improved performance and NFS locking in 2.2.
I believe the reasoning was that a malicious client could still manage to
guess the inode numbers for the root of the partition (/usr in this case) and
hence access it, so there was no point in specifying a subdirectory as a
possible mount point.
The fact that real clients may depend on being able to do so wasn't a
sufficient reason to introduce the extra complexity required to handle this
case, and I didn't have the time or knowledge to do it myself.
ISTR there was also a limit of a single entry in the export table per mounted
filesystem, so I couldn't even export a filesystem RW to local machines and RO
to the rest of the world. Perhaps I was just doing something wrong, though.
---- ---- ----
David Woodhouse David.Woodhouse@mvhi.com Office: (+44) 1223 812896
Project Leader, Process Information Systems Mobile: (+44) 976 658355
Axiom (Cambridge) Ltd., Swaffham Bulbeck, Cambridge, CB5 0NA, UK.
finger dwmw2@ferret.lmh.ox.ac.uk for PGP key.
-
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.altern.org/andrebalsa/doc/lkml-faq.html