Bug 102593 – following symlinks in mount (util-linux) should be an USE-flag

• Status: RESOLVED FIXED
• Product: Gentoo Linux

[Description Honza 2005-08-15 04:44:45 UTC]

I don't care how much you don't like your df output.
Not following symlinks make necessary to write unnecessary long paths as
argument to mount in my case, doing symlink resolving by hand (or memory).
If there is no security bug in it, I'd like to choose following symlinks,
especially because I don't use devfs.
On the other hand, I understand you don't have same problem, so I think simple
use flag for application of that patch will be best solution.
(Note that application default is to follow symlinks - it seems there are more
users like me).

Reproducible: Always
Steps to Reproduce:
1. mount /mnt/sketch/hda-45
2. mount /unix/mnt/sketch/sda-2


Actual Results:  
Only second works.

Expected Results:  
With recomended use flag, both should works.


Portage 2.0.51.22-r2 (default-linux/x86/2004.0, gcc-3.3.2, glibc-2.3.2-r9,
2.6.12-gentoo-r8-32 i686)
=================================================================
System uname: 2.6.12-gentoo-r8-32 i686 AMD Athlon(tm) 64 Processor 3000+
Gentoo Base System version 1.4.16
distcc 2.16 i686-pc-linux-gnu (protocols 1 and 2) (default port 3632) [disabled]
dev-lang/python:     2.3.3
sys-devel/autoconf:  2.13, 2.59-r6
sys-devel/automake:  1.4_p6, 1.5, 1.6.3, 1.7.9-r1, 1.8.5-r3, 1.9.4
sys-devel/binutils:  2.15.92.0.2-r7
sys-devel/libtool:   1.4.3-r3, 1.5.2-r7
virtual/os-headers:  2.6.11-r2
ACCEPT_KEYWORDS="x86"
AUTOCLEAN="yes"
CFLAGS="-O2 -mcpu=athlon -march=i686 -fomit-frame-pointer -pipe"
CHOST="i686-pc-linux-gnu"
CONFIG_PROTECT="/etc /usr/kde/2/share/config /usr/kde/3.2/share/config
/usr/kde/3/share/config /usr/lib/X11/xkb /usr/share/config
/usr/share/texmf/dvipdfm/config/ /usr/share/texmf/dvips/config/
/usr/share/texmf/tex/generic/config/ /usr/share/texmf/tex/platex/config/
/usr/share/texmf/xdvi/ /var/qmail/control"
CONFIG_PROTECT_MASK="/etc/gconf /etc/terminfo /etc/env.d"
CXXFLAGS="-O2 -mcpu=athlon -march=i686 -fomit-frame-pointer -pipe"
DISTDIR="/usr/portage/distfiles"
FEATURES="autoconfig distlocks sandbox sfperms"
GENTOO_MIRRORS="http://www.mirror.ac.uk/sites/www.ibiblio.org/gentoo/
http://gentoo.oregonstate.edu http://www.ibiblio.org/pub/Linux/distributions/gentoo"
MAKEOPTS="-j2"
PKGDIR="/usr/portage/packages"
PORTAGE_TMPDIR="/var/tmp"
PORTDIR="/usr/portage"
PORTDIR_OVERLAY="/usr/local/portage"
SYNC="rsync://rsync.europe.gentoo.org/gentoo-portage"
USE="3dnow 3dnowext X Xaw3d aalib alsa apache2 apm arts avi bash-completion
berkdb bitmap-fonts caps cdr crypt cups curl dga directfb doc dvd emboss encode
esd f77 fbcon flac font-server foomaticdb fortran gd gdbm ggi gif gnome gpm gtk
gtk2 imagemagick imlib innodb ipv6 java jpeg lcms lesstif libcaca libg++ libwww
lirc mad mailwrapper mbox mcal memlimit mikmod mmx mmx2 mmxext mng motif mozilla
mp3 mpeg multislot mysql ncurses nls ogg oggvorbis opengl oss pam pdflib perl
png python qt quicktime readline real samba sdl slang snmp spell sqlite sse sse2
ssl svga tcpd tetex theora tiff truetype truetype-fonts type1-fonts unicode usb
v4l v4l2 vhosts videos vorbis wmf x86 xgetdefault xml xml2 xmms xosd xv xvid
zlib userland_GNU kernel_linux elibc_glibc"
Unset:  LDFLAGS

[Comment 1 Honza 2005-08-15 04:46:28 UTC]

By "that patch" I mean
/usr/portage/sys-apps/util-linux/files/no-symlink-resolve.patch of course.

[Comment 2 SpanKY 2005-08-15 19:56:50 UTC]

the patch is only applied if you have /dev/.devfsd

if you arent using devfs, then why do you have that on your system ?

[Comment 3 Honza 2005-08-16 00:57:51 UTC]

First: thats not true. Or to be more precise, thats not true for
sys-apps/util-linux-2.12i-r1. I see in later ebuilds - now marked unstable -
there are such change.

Second: Because I use it as reference. I have mounted it do /dev/DEVfs and if I
miss some device, I simply copy it from devfs instead of studying driver and
looking for major and minor. Also, I have /dev/snd and svga_helper symlinked
from devfs. But I don't mount disc from it.

Third: In fact my problem will be solved by that test, because I don't have
/dev/.devfsd (I have /dev/DEVfs/.devfsd, but ebuild doesn't test for it). But
are you sure there's no-one using devfs with some problem ? Also, there should
be at least some information that patch is used or is not used, some users can
be confused that simply start or stop using devfs and recompiling util-linux
will change mount behaviour ... 

[Comment 4 SpanKY 2005-08-16 15:10:05 UTC]

very few people notice this change in behavior

a lot more notice when using devfs and `df` output is weird

backported the -e /dev/.devfs check to the older ebuild