Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!uunet!snorkelwacker!ira.uka.de!fauern!tumuc!lan!charly.bl.physik.tu-muenchen.de!k2
From: k2@charly.bl.physik.tu-muenchen.de (Klaus Steinberger)
Newsgroups: comp.unix.ultrix
Subject: Re: vcc creates setuid executables on HP NFS mounted disks
Message-ID: <1567@tuminfo1.lan.informatik.tu-muenchen.dbp.de>
Date: 29 Mar 90 07:10:13 GMT
References: <4622@hemuli.tik.vtt.fi>
Sender: news@lan.informatik.tu-muenchen.dbp.de
Lines: 27

tml@hemuli.tik.vtt.fi (Tor Lillqvist) writes:


>I have noticed a strange thing with the VAX Ultrix 3.1 vcc compiler
>(actually the lk linker) when you use it on a directory NFS-mounted
>from an hp9000s800 machine running HP-UX 3.1).  It sets the setuid and
>setgid bits on the executables it generates!!!  Very interesting...
>Who is to blame, DEC or HP?

I've seen the same problem with a file-system located on a PCS-Cadmus
and mounted on a CONVEX. I've talked with a guy at PCS, and he told
me that there is maybe a problem with short/long in the NFS protocol.
But we don't get a closer look to the problem, because we mainly
exported Convex Filesystems to the Cadmus and not vice-versa.
So we don't know which side is buggy, the client or the server.

I don't know details of the HP NFS Implementation, but the NFS
for the PCS-Cadmus was originally from LAI. The Cadmus is a SYS-V
machine. The Convex is BSD-ish, and I think its NFS is derived from
the BSD Implementation like Ultrix.

Sincerely,
Klaus Steinberger

Klaus Steinberger               Beschleunigerlabor der TU und LMU Muenchen
Phone: (+49 89)3209 4287        Hochschulgelaende, D-8046 Garching, West Germany
BITNET:  K2@DGABLG5P            Internet: k2@charly.bl.physik.tu-muenchen.de