Bug#1026072: firefox: fails to start (cannot load libnssutil3.so)

[Martin Dickopp]

Package: firefox
Version: 108.0-1
Severity: grave
Justification: renders package unusable

When starting firefox, it displays the following message and exits immediately 
with exit code 255:

XPCOMGlueLoad error for file /usr/lib/firefox/libnssutil3.so:
/usr/lib/firefox/libnssutil3.so: cannot open shared object file: No such file 
or directory
Couldn't load XPCOM.

-- Package-specific info:


-- Addons package information

-- System Information:
Debian Release: bookworm/sid
  APT prefers testing
  APT policy: (500, 'testing'), (200, 'unstable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 6.0.0-5-amd64 (SMP w/16 CPU threads; PREEMPT)
Locale: LANG=C, LC_CTYPE=C.UTF-8 (charmap=UTF-8), LANGUAGE not set
Shell: /bin/sh linked to /usr/bin/dash
Init: systemd (via /run/systemd/system)
LSM: AppArmor: enabled

Versions of packages firefox depends on:
ii  debianutils  5.7-0.4
ii  fontconfig   2.13.1-4.5
ii  libasound2   1.2.8-1
ii  libatk1.0-0  2.46.0-4
ii  libc62.36-6
ii  libcairo-gobject21.16.0-6
ii  libcairo21.16.0-6
ii  libdbus-1-3  1.14.4-1
ii  libdbus-glib-1-2 0.112-3
ii  libevent-2.1-7   2.1.12-stable-5+b1
ii  libffi8  3.4.4-1
ii  libfontconfig1   2.13.1-4.5
ii  libfreetype6 2.12.1+dfsg-3
ii  libgcc-s112.2.0-9
ii  libgdk-pixbuf-2.0-0  2.42.10+dfsg-1
ii  libglib2.0-0 2.74.2-1
ii  libgtk-3-0   3.24.35-3
ii  libnspr4 2:4.35-1
ii  libnss3  2:3.85-1
ii  libpango-1.0-0   1.50.10+ds-1
ii  libstdc++6   12.2.0-9
ii  libvpx7  1.12.0-1
ii  libx11-6 2:1.8.1-2
ii  libx11-xcb1  2:1.8.1-2
ii  libxcb-shm0  1.15-1
ii  libxcb1  1.15-1
ii  libxcomposite1   1:0.4.5-1
ii  libxdamage1  1:1.1.5-2
ii  libxext6 2:1.3.4-1+b1
ii  libxfixes3   1:6.0.0-2
ii  libxrandr2   2:1.5.2-2+b1
ii  libxtst6 2:1.2.3-1.1
ii  procps   2:3.3.17-7.1
ii  zlib1g   1:1.2.13.dfsg-1

Versions of packages firefox recommends:
ii  libavcodec-extra58 [libavcodec58]  7:4.4.2-1+b3
ii  libavcodec59   7:5.1.2-1

Versions of packages firefox suggests:
pn  fonts-lmodern  
ii  fonts-stix [otf-stix]  1.1.1-4.1
ii  libcanberra0   0.30-10
ii  libgssapi-krb5-2   1.20.1-1
ii  pulseaudio 16.1+dfsg1-2+b1

-- no debconf information

Bug#1005884: linux-image-5.16.0-1-amd64: Kernel oops (unable to handle page fault) during boot

[Martin Dickopp]

On Sat, Feb 19, 2022 at 03:55:20PM +0100, Salvatore Bonaccorso wrote:
> This is possibly the same as reported in
> https://lore.kernel.org/stable/05b11936073c8d6b7a28c07cc...@stwm.de/
> where the reporter found that the culprit is ab07506b0454 ("iwlwifi:
> fix leaks/bad data after failed firmware load")
> 
> Does the upstream commit bea2662e7818 ("iwlwifi: fix
> use-after-free")[0] fixes the issue for you?
> 
>  [0] https;//git.kernel.org/linus/bea2662e7818e15d7607d17d57912ac984275d94

Yes, this commit fixes the issue for me. Thanks!

Best regards,
Martin

Bug#885061: systemd-timesyncd.service fails to start if static user is removed, but group is retained

[Martin Dickopp]

On Sat, Dec 23, 2017 at 11:35:44AM +0100, Michael Biebl wrote:
> Am 23.12.2017 um 09:26 schrieb Martin Dickopp:
> > Following the instructions in /usr/share/doc/systemd/NEWS.Debian.gz
> > 
> >   DynamicUser=yes has been enabled for systemd-journal-upload.service,
> >   systemd-journal-gatewayd.service and systemd-timesyncd.service.
> >   This means we no longer need to statically allocate a 
> > systemd-journal-upload,
> >   systemd-journal-gateway and systemd-timesync user and you can now safely
> >   remove those system users.
> > 
> > I removed the static systemd-timesync user and rebooted the system, after
> > which systemd-timesyncd.service failed to start.
> > 
> > When I removed the static systemd-timesync group as well, and rebooted
> > again, systemd-timesyncd.service worked fine.
> 
> Out of interest: which tool did you use to remove the system user?
> deluser by default will remove the group along with the user automatically.

I removed the entry manually from /etc/passwd and /etc/shadow (with vipw).

> That said, I'd be happy to provide a text which clarifies the above.
> Would ".. and you can now safely remove those system users along with
> their associated groups." be sufficient in your opinion?

Yes, this clarification would have helped me. Thanks!

Best regards,
Martin

Bug#885061: systemd-timesyncd.service fails to start if static user is removed, but group is retained

[Martin Dickopp]

Package: systemd
Version: 236-1
Severity: normal

Hi,

Following the instructions in /usr/share/doc/systemd/NEWS.Debian.gz

  DynamicUser=yes has been enabled for systemd-journal-upload.service,
  systemd-journal-gatewayd.service and systemd-timesyncd.service.
  This means we no longer need to statically allocate a systemd-journal-upload,
  systemd-journal-gateway and systemd-timesync user and you can now safely
  remove those system users.

I removed the static systemd-timesync user and rebooted the system, after
which systemd-timesyncd.service failed to start.

When I removed the static systemd-timesync group as well, and rebooted
again, systemd-timesyncd.service worked fine.

(I don't use systemd-journal-{upload,gatewayd}.service, so I don't know if
this applies to these services as well.)

Best regards,
Martin


-- Package-specific info:

-- System Information:
Debian Release: buster/sid
  APT prefers testing
  APT policy: (500, 'testing'), (200, 'unstable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 4.13.0-1-amd64 (SMP w/12 CPU cores)
Locale: LANG=C, LC_CTYPE=C.UTF-8 (charmap=UTF-8), LANGUAGE=C (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)

Versions of packages systemd depends on:
ii  adduser 3.116
ii  libacl1 2.2.52-3+b1
ii  libapparmor12.11.1-4
ii  libaudit1   1:2.8.2-1
ii  libblkid1   2.30.2-0.1
ii  libc6   2.25-3
ii  libcap2 1:2.25-1.2
ii  libcryptsetup4  2:1.7.5-1
ii  libgcrypt20 1.8.1-4
ii  libgpg-error0   1.27-5
ii  libidn111.33-2.1
ii  libip4tc0   1.6.1-2+b1
ii  libkmod224-1
ii  liblz4-10.0~r131-2+b1
ii  liblzma55.2.2-1.3
ii  libmount1   2.30.2-0.1
ii  libpam0g1.1.8-3.6
ii  libseccomp2 2.3.1-2.1
ii  libselinux1 2.7-2
ii  libsystemd0 236-1
ii  mount   2.30.2-0.1
ii  procps  2:3.3.12-3
ii  util-linux  2.30.2-0.1

Versions of packages systemd recommends:
ii  dbus1.12.2-1
ii  libpam-systemd  236-1

Versions of packages systemd suggests:
ii  policykit-10.105-18
ii  systemd-container  236-1

Versions of packages systemd is related to:
pn  dracut   
ii  initramfs-tools  0.130
ii  udev 236-1

-- Configuration Files:
/etc/systemd/logind.conf changed [not included]

-- no debconf information

Bug#866902: nftables: systemd unit loads /etc/nftables.conf too late in the boot process

[Martin Dickopp]

Package: nftables
Version: 0.7-1
Severity: normal

Dear Maintainer,

The systemd unit file /lib/systemd/system/nftables.service should contain
a Before=network-pre.target statement, so that firewall rules are
established before network interfaces are up. Otherwise, there may be a
small time window during which the host is not protected by the firewall.
This is also recommended in the systemd documentation; see
https://www.freedesktop.org/wiki/Software/systemd/NetworkTarget/

Furthermore, the unit should be WantedBy=network.target instead of
WantedBy=multi-user.target, so that it is started when the network is
started, even if the system is not in multi-user mode.

Best regards,
Martin


--- /lib/systemd/system/nftables.service.ORIG   2017-07-02 13:27:00.310036693 
+0200
+++ /lib/systemd/system/nftables.service2017-07-02 13:31:24.827267347 
+0200
@@ -1,6 +1,7 @@
 [Unit]
 Description=nftables
 Documentation=man:nft(8) http://wiki.nftables.org
+Before=network-pre.target
 
 [Service]
 Type=oneshot
@@ -13,4 +14,4 @@
 ExecStop=/usr/sbin/nft flush ruleset
 
 [Install]
-WantedBy=multi-user.target
+WantedBy=network.target



-- System Information:
Debian Release: buster/sid
  APT prefers testing
  APT policy: (500, 'testing'), (200, 'unstable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 4.9.0-3-amd64 (SMP w/12 CPU cores)
Locale: LANG=C, LC_CTYPE=C.UTF-8 (charmap=UTF-8), LANGUAGE=C (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)

Versions of packages nftables depends on:
ii  dpkg 1.18.24
ii  init-system-helpers  1.48
ii  libc62.24-12
ii  libgmp10 2:6.1.2+dfsg-1
ii  libmnl0  1.0.4-2
ii  libnftnl41.0.7-1
ii  libreadline7 7.0-3
ii  libxtables12 1.6.0+snapshot20161117-6

Bug#823097: blender: please link with OpenVDB library

[Martin Dickopp]

Package: blender
Version: 2.77.a+dfsg0-2
Severity: wishlist
Tags: patch

Dear Maintainer,

Please consider linking with OpenVDB, a library that Blender (since 2.77)
can use to cache smoke simulations more efficiently.

Suggested patch attached.

Best regards,
Martin


-- System Information:
Debian Release: stretch/sid
  APT prefers testing
  APT policy: (500, 'testing'), (200, 'unstable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 4.5.0-1-amd64 (SMP w/12 CPU cores)
Locale: LANG=C, LC_CTYPE=C.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)

Versions of packages blender depends on:
ii  blender-data  2.77.a+dfsg0-2
ii  fonts-dejavu  2.35-1
ii  libavcodec-extra577:3.0.1-3
ii  libavdevice57 7:3.0.1-3
ii  libavformat57 7:3.0.1-3
ii  libavutil55   7:3.0.1-3
ii  libboost-atomic1.58.0 1.58.0+dfsg-5+b1
ii  libboost-chrono1.58.0 1.58.0+dfsg-5+b1
ii  libboost-date-time1.58.0  1.58.0+dfsg-5+b1
ii  libboost-filesystem1.58.0 1.58.0+dfsg-5+b1
ii  libboost-locale1.58.0 1.58.0+dfsg-5+b1
ii  libboost-regex1.58.0  1.58.0+dfsg-5+b1
ii  libboost-system1.58.0 1.58.0+dfsg-5+b1
ii  libboost-thread1.58.0 1.58.0+dfsg-5+b1
ii  libc6 2.22-7
ii  libfftw3-double3  3.3.4-2+b1
ii  libfontconfig12.11.0-6.4
ii  libfreetype6  2.6.3-3+b1
ii  libgcc1   1:5.3.1-14
ii  libgl1-mesa-glx [libgl1]  11.1.3-1
ii  libglew1.13   1.13.0-2
ii  libglu1-mesa [libglu1]9.0.0-2.1
ii  libgomp1  5.3.1-14
ii  libilmbase12  2.2.0-11
ii  libjack-jackd2-0 [libjack-0.116]  1.9.10+20150825git1ed50c92~dfsg-1
ii  libjemalloc1  3.6.0-9
ii  libjpeg62-turbo   1:1.4.2-2
ii  libopenal11:1.17.2-1
ii  libopencolorio1v5 1.0.9~dfsg0-5
ii  libopenexr22  2.2.0-10
ii  libopenimageio1.6 1.6.12~dfsg0-1+b1
ii  libopenjpeg5  1:1.5.2-3.1
ii  libpng16-16   1.6.21-4
ii  libpython3.5  3.5.1-11
ii  libsndfile1   1.0.25-10
ii  libspnav0 0.2.3-1
ii  libstdc++65.3.1-14
ii  libswscale4   7:3.0.1-3
ii  libtiff5  4.0.6-1
ii  libx11-6  2:1.6.3-1
ii  libxi62:1.7.6-1
ii  libxxf86vm1   1:1.1.4-1
ii  zlib1g1:1.2.8.dfsg-2+b1

blender recommends no packages.

blender suggests no packages.

-- no debconf information
From: Martin Dickopp <mar...@zero-based.org>
Date: Sat, 30 Apr 2016 17:45:53 +0200
Subject: Link with OpenVDB library

---
 debian/control | 3 +++
 debian/rules   | 1 +
 2 files changed, 4 insertions(+)

diff --git a/debian/control b/debian/control
index 3f6d122..414a41d 100644
--- a/debian/control
+++ b/debian/control
@@ -14,6 +14,7 @@ Build-Depends: autotools-dev,
libavfilter-dev,
libboost-dev,
libboost-filesystem-dev,
+   libboost-iostreams-dev,
libboost-locale-dev,
libboost-regex-dev,
libboost-system-dev,
@@ -34,11 +35,13 @@ Build-Depends: autotools-dev,
libopenexr-dev (>= 2.2.0),
libopenimageio-dev,
libopenjpeg-dev,
+   libopenvdb-dev,
libpng-dev,
libsdl-dev,
libsndfile-dev,
libspnav-dev,
libswscale-dev,
+   libtbb-dev,
libtiff-dev,
libvorbis-dev,
libxi-dev,
diff --git a/debian/rules b/debian/rules
index 18b64ba..4bcfdc4 100755
--- a/debian/rules
+++ b/debian/rules
@@ -37,6 +37,7 @@ override_dh_auto_configure:
 		-DWITH_JACK=ON \
 		-DWITH_MOD_OCEANSIM=ON \
 		-DWITH_OPENCOLORIO=ON \
+		-DWITH_OPENVDB=ON \
 		-DWITH_PLAYER=ON \
 		-DWITH_PYTHON_INSTALL=OFF \
 		-DWITH_SYSTEM_OPENJPEG=ON

Bug#818179: This issue is resolved upstream in xorg

[Martin Dickopp]

On Mon, Mar 21, 2016 at 11:58:38AM +0200, Laszlo KERTESZ wrote:
> After a bit of digging i found this issue was indeed caused by glamor in x
> server 1.18.2 and it has a fixed bug in xorg:
> 
> https://bugs.freedesktop.org/show_bug.cgi?id=94554

I recompiled xserver-xorg-core 2:1.18.2-1 with this patch

https://cgit.freedesktop.org/xorg/xserver/commit/?id=a288cf58a0dc0f965a6f964c76bb86bb1989d797

applied, and it does indeed fix the issue for me.

This bug should probably be reassigned to xserver-xorg-core.

Best regards,
Martin

Bug#818179: firefox: rgba(0,0,0,0.99) becomes read instead of black

[Martin Dickopp]

Hello,

I see the same bug on my PC with Radeon graphics card

$ lspci | grep -i radeon
01:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] 
Curacao XT [Radeon R7 370 / R9 270X/370 OEM]
01:00.1 Audio device: Advanced Micro Devices, Inc. [AMD/ATI] Cape 
Verde/Pitcairn HDMI Audio [Radeon HD 7700/7800 Series]

but not on my laptop with Intel graphics.

Best regards,
Martin

Bug#815126: linux-image-4.4.0-1-amd64 hangs after 'Loading initial ramdisk ...'

[Martin Dickopp]

On Fri, Feb 19, 2016 at 10:50:34AM +0100, Vincent Bernat wrote:
>  ❦ 19 février 2016 07:34 GMT, Jim Barber 
>  :
> 
> > I upgraded from linux-image-4.3.0-1-amd64 (version 4.3.5-1) to 
> > linux-image-4.4.0-1-amd64.
> > On rebooting the system I am presented with grub, and then start booting 
> > into Linux.
> > I only get to see 2 lines with the cursor below them on the lefthand side.
> >
> > Loading Linux 4.4.0-1 ...
> > Loading initial ramdisk ...
> > _
> >
> > The system freezes at this point and is unresponsive.
> > I have to hold my power button down for approx 10s to power off the laptop.
> > Booting back into linux-4.3.0-1-amd64 works fine.
> 
> I am in the same situation. I am also on the same system (Thinkpad X1
> Carbon, but 2nd).

I observe exactly the same behavior on my Toshiba Kira 10D.

Best regards,
Martin

Bug#809494: network-manager: NetworkManager 1.0.10-1 does not add default route for VPN

[Martin Dickopp]

On Thu, Dec 31, 2015 at 11:32:20AM +0100, Marcel Lippmann wrote:
> I am using NetworkManager and the OpenVPN plugin
> (network-manager-openvpn). After upgrading NetworkManager 1.0.8-1 to
> 1.0.10-1, no default routes were set anymore.

Hi,

I had the same issue. Adding the line

  topology subnet

to the OpenVPN server configuration fixed it for me.

Best regards,
Martin

Bug#733261: [Pkg-xfce-devel] Bug#733261: lightdm lost ability not to set LANG

[Martin Dickopp]

On Sat, Dec 28, 2013 at 11:45:30PM +0100, Yves-Alexis Perez wrote:
 On Fri, Dec 27, 2013 at 10:38:33PM +0100, Martin Dickopp wrote:
  Package: lightdm
  Version: 1.8.5-2
  Severity: normal
  
  Hello,
  
  It seems that LightDM lost the ability not to set the LANG environment
  variable with the 1.6.3-1 - 1.8.5-2 upgrade.
  
  My system is configured as follows:
  
  $ cat /etc/default/locale
  LC_CTYPE=C.UTF-8
  $ grep -v ^# /etc/locale.gen
  
  de_DE ISO-8859-1
  de_DE.UTF-8 UTF-8
  en_US ISO-8859-1
  en_US.UTF-8 UTF-8
  
  I have disabled the language selector in
  /etc/lightdm/lightdm-gtk-greeter.conf (show-language-selector=false).
  
  Observed behavior: After logging in, the LANG environment variable is set
  to de_DE.utf8, which appears to be based on one of the entries (maybe
  the first) in /etc/locale.gen.
  
  Expected behavior: LANG unset, or set to something neutral like C or
  POSIX.
 
 Do you have a .dmrc? Can you paste the content?

$ cat ~/.dmrc

[Desktop]
Session=lightdm-xsession
Language=de_DE.utf8

If I remove the file and restart lightdm, it is recreated with the same
content. After some experimentation, I found that lightdm keeps another
copy of the file in /var/cache/lightdm/dmrc. Removing both ~/.dmrc and
/var/cache/lightdm/dmrc/martin.dmrc resolves the issue!

While my immediate issue is resolved, I still find this behavior
unfortunate for three reasons:

- Recreating a file a user has deleted from a shadow copy is probably as
  surprising to many users as it has been to me. :-)

- While I'm also root on the machine in question, most users don't have
  write permission in /var/cache/lightdm/dmrc.

- According to my understanding, programs must handle files stored under
  /var/cache going away (e.g. /var/cache can be omitted from backups).
  It's therefore surprising that removal of a file under /var/cache
  changes the behavior.

Thanks,
Martin


-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#733261: lightdm lost ability not to set LANG

[Martin Dickopp]

Package: lightdm
Version: 1.8.5-2
Severity: normal

Hello,

It seems that LightDM lost the ability not to set the LANG environment
variable with the 1.6.3-1 - 1.8.5-2 upgrade.

My system is configured as follows:

$ cat /etc/default/locale
LC_CTYPE=C.UTF-8
$ grep -v ^# /etc/locale.gen

de_DE ISO-8859-1
de_DE.UTF-8 UTF-8
en_US ISO-8859-1
en_US.UTF-8 UTF-8

I have disabled the language selector in
/etc/lightdm/lightdm-gtk-greeter.conf (show-language-selector=false).

Observed behavior: After logging in, the LANG environment variable is set
to de_DE.utf8, which appears to be based on one of the entries (maybe
the first) in /etc/locale.gen.

Expected behavior: LANG unset, or set to something neutral like C or
POSIX.

If I enable the language selector in the greeter application, I can choose
between German or English, but there is still no way not to set LANG at
all.

Best regards,
Martin


-- System Information:
Debian Release: jessie/sid
  APT prefers testing
  APT policy: (500, 'testing'), (200, 'unstable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 3.11-2-amd64 (SMP w/4 CPU cores)
Locale: LANG=C, LC_CTYPE=C.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages lightdm depends on:
ii  adduser3.113+nmu3
ii  consolekit 0.4.6-3+b1
ii  dbus   1.6.18-2
ii  debconf [debconf-2.0]  1.5.52
ii  libc6  2.17-97
ii  libgcrypt111.5.3-2
ii  libglib2.0-0   2.36.4-1
ii  libpam0g   1.1.3-9
ii  libxcb11.9.1-3.1
ii  libxdmcp6  1:1.1.1-1
ii  lightdm-gtk-greeter [lightdm-greeter]  1.6.1-4

Versions of packages lightdm recommends:
ii  xserver-xorg  1:7.7+4

Versions of packages lightdm suggests:
pn  accountsservice  none
ii  upower   0.9.23-2+b1

-- Configuration Files:
/etc/lightdm/lightdm.conf changed:
[LightDM]
[SeatDefaults]
display-setup-script=/etc/X11/Xsetup
[XDMCPServer]
[VNCServer]

/etc/lightdm/users.conf changed:
[UserAccounts]
minimum-uid=1000
hidden-users=nobody nobody4 noaccess
hidden-shells=/bin/false /usr/sbin/nologin

/etc/pam.d/lightdm changed:
auth  requisite pam_nologin.so
auth  required pam_env.so envfile=/etc/default/locale
@include common-auth
auth  optional pam_gnome_keyring.so
@include common-account
session  [success=ok ignore=ignore module_unknown=ignore default=bad] 
pam_selinux.so close
session  requiredpam_limits.so
session  optionalpam_mail.so dir=~/Maildir nopen
session  requiredpam_loginuid.so
@include common-session
session [success=ok ignore=ignore module_unknown=ignore default=bad] 
pam_selinux.so open
session optionalpam_gnome_keyring.so auto_start
@include common-password


-- debconf information:
* shared/default-x-display-manager: lightdm
  lightdm/daemon_name: /usr/sbin/lightdm


-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#720433: [php-maint] Bug#720433: libapache2-mod-php5: Segmentation fault in 5.5.2+dfsg-1

[Martin Dickopp]

It seems the bug is fixed in upstream 5.5.4, and therefore in 5.5.4+dfsg-1.
Inspecting the upstream git repository shows that the fix is included,
although the upstream changelog doesn't mention it.

Furthermore, the problem no longer occurs for me with 5.5.4+dfsg-1.

Best regards,
Martin


-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#717423: libfontconfig1:amd64: C.UTF-8 locale causes warning to be printed on standard error

[Martin Dickopp]

Package: libfontconfig1
Version: 2.10.2-2
Severity: normal
Tags: patch

When the C.UTF-8 locale is used (i.e. LC_CTYPE=C.UTF-8 or LANG=C.UTF-8),
programs using fontconfig print the message

  Fontconfig warning: ignoring C.UTF-8: not a valid language tag

on standard error serveral times. For example, Emacs prints the warning
ten times during startup.

The attached patch fixes the issue for me. It is based on the assumption
that C.UTF-8 should be treated like the other locale names built into the
C library (C and POSIX).

Best regards,
Martin


-- System Information:
Debian Release: jessie/sid
  APT prefers testing
  APT policy: (500, 'testing'), (200, 'unstable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 3.9-1-amd64 (SMP w/4 CPU cores)
Locale: LANG=C, LC_CTYPE=C.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages libfontconfig1:amd64 depends on:
ii  fontconfig-config  2.10.2-2
ii  libc6  2.17-7
ii  libexpat1  2.1.0-4
ii  libfreetype6   2.4.9-1.1
ii  multiarch-support  2.17-7

libfontconfig1:amd64 recommends no packages.

libfontconfig1:amd64 suggests no packages.

-- no debconf information
diff -Nur a/src/fclang.c b/src/fclang.c
--- a/src/fclang.c	2012-10-30 01:53:37.0 +0100
+++ b/src/fclang.c	2013-07-20 19:49:14.352195363 +0200
@@ -188,6 +188,7 @@
 	return NULL;
 
 if (FcStrCmpIgnoreCase (lang, (const FcChar8 *)C) == 0 ||
+	FcStrCmpIgnoreCase (lang, (const FcChar8 *)C.UTF-8) == 0 ||
 	FcStrCmpIgnoreCase (lang, (const FcChar8 *)POSIX) == 0)
 {
 	result = FcStrCopy ((const FcChar8 *)en);

Bug#712434: lightdm-gtk-greeter.conf: discrepancy between comment and default value

[Martin Dickopp]

Package: lightdm-gtk-greeter
Version: 1.6.0-1
Severity: minor

Hi,

/etc/lightdm/lightdm-gtk-greeter.conf contains these two lines:

# xft-hintstyle = What degree of hinting to use (none, slight, medium, or full)

xft-hintstyle=hintfull

According to the comment, hintfull is not one of the allowed values.
(I'm not sure if the comment or the value is wrong, though.)

Thanks,
Martin


-- System Information:
Debian Release: jessie/sid
  APT prefers testing
  APT policy: (500, 'testing'), (200, 'unstable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 3.9-1-amd64 (SMP w/4 CPU cores)
Locale: LANG=C, LC_CTYPE=C.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages lightdm-gtk-greeter depends on:
ii  libc6   2.17-3
ii  libcairo2   1.12.14-4
ii  libgdk-pixbuf2.0-0  2.28.1-1
ii  libglib2.0-02.36.1-2build1
ii  libgtk-3-0  3.4.2-6
ii  liblightdm-gobject-1-0  1.6.0-3
ii  libx11-62:1.5.0-1+deb7u1

Versions of packages lightdm-gtk-greeter recommends:
ii  desktop-base   7.0.3
ii  gnome-icon-theme   3.4.0-2
ii  gnome-themes-standard  3.4.2-2.1
ii  policykit-10.105-3

lightdm-gtk-greeter suggests no packages.

-- no debconf information


-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#712436: lightdm-gtk-greeter: Disabling language selector leaves strange artifact on the screen

[Martin Dickopp]

Package: lightdm-gtk-greeter
Version: 1.6.0-1
Severity: normal

Hi,

When I disable the language selector (by setting
show-language-selector=false in /etc/lightdm/lightdm-gtk-greeter.conf), it
is replaced by a dark, almost black area that is two pixels wide and one
pixel high. Please see the attached screenshot of the upper right screen
area.

I would expect the language selector to be removed entirely (which is what
earlier lightdm-gtk-greeter versions did).

Thanks,
Martin


-- System Information:
Debian Release: jessie/sid
  APT prefers testing
  APT policy: (500, 'testing'), (200, 'unstable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 3.9-1-amd64 (SMP w/4 CPU cores)
Locale: LANG=C, LC_CTYPE=C.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages lightdm-gtk-greeter depends on:
ii  libc6   2.17-3
ii  libcairo2   1.12.14-4
ii  libgdk-pixbuf2.0-0  2.28.1-1
ii  libglib2.0-02.36.1-2build1
ii  libgtk-3-0  3.4.2-6
ii  liblightdm-gobject-1-0  1.6.0-3
ii  libx11-62:1.5.0-1+deb7u1

Versions of packages lightdm-gtk-greeter recommends:
ii  desktop-base   7.0.3
ii  gnome-icon-theme   3.4.0-2
ii  gnome-themes-standard  3.4.2-2.1
ii  policykit-10.105-3

lightdm-gtk-greeter suggests no packages.

-- Configuration Files:
/etc/lightdm/lightdm-gtk-greeter.conf changed:
[greeter]
background=/usr/share/images/desktop-base/login-background.svg
theme-name=Adwaita
xft-antialias=true
xft-hintstyle=hintfull
xft-rgba=rgb
show-language-selector=false


-- no debconf information
attachment: greeter.png

Bug#679753: calc: prime number test for large numbers results in error

[Martin Dickopp]

Package: emacs24-common
Version: 24.1+1-1
Severity: normal

The prime number test in calc results in an error if the tested number is
large. It works fine for small numbers.

Steps to reproduce:

- Start calc (M-x calc RET)
- Enter the number 123456789 (123456789 RET)
- Perform prime number test (k p)

Expected result: The message Non-prime (or something similar) is
displayed.

Actual result: The message Wrong type argument: listp, 123456789 is
displayed.

If small numbers are tested, calc behaves as expected. E.g. typing
123 k p displays the message Non-prime.


-- System Information:
Debian Release: wheezy/sid
  APT prefers testing
  APT policy: (500, 'testing'), (200, 'unstable')
Architecture: amd64 (x86_64)

Kernel: Linux 3.2.0-2-amd64 (SMP w/4 CPU cores)
Locale: LANG=C, LC_CTYPE=C.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages emacs24-common depends on:
ii  dpkg1.16.4.3
ii  emacsen-common  2.0.3
ii  install-info4.13a.dfsg.1-10

emacs24-common recommends no packages.

Versions of packages emacs24-common suggests:
ii  emacs24-common-non-dfsg  24.1+1-1
pn  emacs24-el   none

-- no debconf information



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#655861: unknown-horizons: crashes when clicking Settings

[Martin Dickopp]

Package: unknown-horizons
Version: 2011.3+dfsg1-1
Severity: normal

Hi,

When I start unknown-horizons and click on the Settings button,
it crashes with the following backtrace:

 File /usr/lib/python2.7/dist-packages/fife/extensions/fife_timer.py, line 
125, in updateEvent
self._callback()
 File /usr/lib/python2.7/dist-packages/fife/extensions/pychan/tools.py, line 
83, in real_callback
callback(*args,**kwargs)
 File /usr/lib/python2.7/dist-packages/fife/extensions/pychan/events.py, line 
164, in delayed_f
f( event )
 File /usr/lib/python2.7/dist-packages/fife/extensions/pychan/events.py, line 
327, in captured_f

tools.applyOnlySuitable(self_ref().callbacks[group_name][event_name],event=event,widget=self_ref().widget_ref())
 File /usr/lib/python2.7/dist-packages/fife/extensions/pychan/tools.py, line 
60, in applyOnlySuitable
return func(*args,**kwargs)
 File /usr/lib/python2.7/dist-packages/horizons/gui/gui.py, line 159, in 
show_settings
horizons.main.fife._setting.onOptionsPress()
 File /usr/lib/python2.7/dist-packages/fife/extensions/fife_settings.py, line 
503, in onOptionsPress
self.fillWidgets()
 File /usr/lib/python2.7/dist-packages/fife/extensions/fife_settings.py, line 
546, in fillWidgets
raise ValueError(\ + value + \ is not a valid value for  + entry.name 
+ . Valid options:  + str(entry.initialdata))
 ValueError: 1024x768 is not a valid value for ScreenResolution. Valid 
options: ['800x600']

Traceback (most recent call last):
  File /usr/games/unknown-horizons, line 490, in module
main()
  File /usr/games/unknown-horizons, line 217, in main
ret = horizons.main.start(options)
  File /usr/lib/python2.7/dist-packages/horizons/main.py, line 176, in start
fife.run()
  File /usr/lib/python2.7/dist-packages/horizons/engine.py, line 572, in run
self.loop()
  File /usr/lib/python2.7/dist-packages/horizons/engine.py, line 581, in loop
self.engine.pump()
  File /usr/lib/python2.7/dist-packages/fife/fife.py, line 2150, in pump
return _fife.Engine_pump(self)
RuntimeError: Catched director exception


My screen is an LCD with a physical resolution of 1024x600.

Cheers,
Martin


-- System Information:
Debian Release: wheezy/sid
  APT prefers testing
  APT policy: (500, 'testing'), (200, 'unstable')
Architecture: amd64 (x86_64)

Kernel: Linux 3.1.0-1-amd64 (SMP w/4 CPU cores)
Locale: LANG=C, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages unknown-horizons depends on:
ii  dpkg 1.16.1.2
ii  python   2.7.2-9
ii  python-enet  0.0~svn24-1
ii  python-fife  0.3.3+r2+dfsg1-3
ii  python-yaml  3.10-2
ii  python2.62.6.7-4
ii  python2.72.7.2-8

unknown-horizons recommends no packages.

unknown-horizons suggests no packages.

-- no debconf information



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#652839: eclipse: Please package Eclipse 3.7.1

[Martin Dickopp]

Package: eclipse
Version: 3.7.0-1
Severity: wishlist

Please package Eclipse version 3.7.1. This version supports Java 7
development and is therefore useful for everyone switching from Java 6
to Java 7.

Thank you for your work on Eclipse packaging!

Best regards,
Martin


-- System Information:
Debian Release: wheezy/sid
  APT prefers testing
  APT policy: (500, 'testing'), (200, 'unstable')
Architecture: amd64 (x86_64)

Kernel: Linux 3.1.0-1-amd64 (SMP w/4 CPU cores)
Locale: LANG=C, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages eclipse depends on:
ii  eclipse-jdt  3.7.0-1
ii  eclipse-pde  3.7.0-1

eclipse recommends no packages.

eclipse suggests no packages.

Versions of packages eclipse-platform depends on:
ii  ant1.8.2-4
ii  ant-optional   1.8.2-4
ii  eclipse-platform-data  3.7.0-1
ii  eclipse-rcp3.7.0-1
ii  java-common0.45
ii  libc6  2.13-21
ii  libcommons-codec-java  1.5-1
ii  libcommons-el-java 1.0-7
ii  libcommons-httpclient-java 3.1-10
ii  libcommons-logging-java1.1.1-8
ii  libjasper-java 5.5.33-2
ii  libjetty-java  6.1.24-6
ii  libjsch-java   0.1.42-2
ii  liblucene2-java2.9.4+ds1-4
ii  libservlet2.5-java 6.0.33-1
ii  openjdk-7-jre [java6-runtime]  7~b147-2.0-1
ii  perl   5.14.2-6
ii  sat4j  2.3.0-3

Versions of packages eclipse-platform recommends:
ii  eclipse-pde  3.7.0-1

Versions of packages eclipse-platform suggests:
ii  eclipse-jdt  3.7.0-1

Versions of packages eclipse-pde depends on:
ii  eclipse-jdt3.7.0-1
ii  eclipse-platform   3.7.0-1
ii  libasm3-java   3.3.2-1
ii  openjdk-7-jre [java6-runtime]  7~b147-2.0-1

eclipse-pde suggests no packages.

Versions of packages eclipse-jdt depends on:
ii  eclipse-platform   3.7.0-1
ii  junit  3.8.2-8
ii  junit4 4.8.2-2
ii  libhamcrest-java   1.1-8
ii  openjdk-7-jre [java6-runtime]  7~b147-2.0-1

Versions of packages eclipse-jdt recommends:
pn  default-jdk | sun-java6-jdk  none

Versions of packages eclipse-jdt suggests:
pn  eclipse-plugin-cvs  none

-- no debconf information



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#597937: consolekit

[Martin Dickopp]

Hi,

I use xfce4 4.8.0.2, xdm 1:1.1.10-3, and consolekit 0.4.5-1. What I
observe is that the session is local, but not active:

Session2:
unix-user = '1000'
realname = 'Martin Dickopp'
seat = 'Seat2'
session-type = ''
active = FALSE
x11-display = ':0'
x11-display-device = ''
display-device = ''
remote-host-name = ''
is-local = TRUE
on-since = '2011-06-26T10:56:05.101604Z'
login-session-id = '2'

My /etc/pam.d/xdm file:

authrequisite   pam_nologin.so
authrequiredpam_group.so
authrequiredpam_env.so
authrequiredpam_env.so envfile=/etc/default/locale
session requiredpam_limits.so
session optionalpam_mail.so dir=~/Maildir nopen
session optionalpam_loginuid.so
session optionalpam_ck_connector.so
@include common-auth
@include common-account
@include common-session
@include common-password
authoptionalpam_gnome_keyring.so
session optionalpam_gnome_keyring.so auto_start

The issue seems to be related to the fact that x11-display-device is
empty. If I force pam_ck_connector to set it by setting the environment
variable CKCON_X11_DISPLAY_DEVICE=/dev/tty7 manually, all is well:

Session3:
unix-user = '1000'
realname = 'Martin Dickopp'
seat = 'Seat1'
session-type = ''
active = TRUE
x11-display = ':0'
x11-display-device = '/dev/tty7'
display-device = ''
remote-host-name = ''
is-local = TRUE
on-since = '2011-06-26T11:17:36.448291Z'
login-session-id = '3'

This setup (without the pam_loginuid.so line and without having to set
CKCON_X11_DISPLAY_DEVICE) used to work fine with the squeeze versions of
xfce4, xdm, and consolekit. Sorry, I don't remember which package update
broke it.

Martin



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#605313: postgresql-client-9.0: Del key stopped working in psql

[Martin Dickopp]

Package: postgresql-client-9.0
Version: 9.0.1-2
Severity: minor

When I hit the Del key on the psql command line, a ~ character is
inserted. I would expect the character at the cursor position to be
deleted instead.

I observe this behavior when I run psql in xfce4-terminal as well as on
the Linux console. Please let me know if you need any further information.

Thanks,
Martin


-- System Information:
Debian Release: squeeze/sid
  APT prefers testing
  APT policy: (500, 'testing'), (200, 'unstable')
Architecture: amd64 (x86_64)

Kernel: Linux 2.6.32-5-amd64 (SMP w/4 CPU cores)
Locale: LANG=C, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages postgresql-client-9.0 depends on:
ii  libc6   2.11.2-7 Embedded GNU C Library: Shared lib
ii  libedit22.11-20080614-2  BSD editline and history libraries
ii  libpq5  9.0.1-2  PostgreSQL C client library
ii  libssl0.9.8 0.9.8o-3 SSL shared libraries
ii  postgresql-client-commo 112  manager for multiple PostgreSQL cl
ii  zlib1g  1:1.2.3.4.dfsg-3 compression library - runtime

postgresql-client-9.0 recommends no packages.

Versions of packages postgresql-client-9.0 suggests:
ii  postgresql-9.09.0.1-2object-relational SQL database, ve
ii  postgresql-doc-9.09.0.1-2documentation for the PostgreSQL d

-- no debconf information



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#601093: grub-common: wrong pathnames in linux menu entry in grub.cfg if kernel on separate filesystem

[Martin Dickopp]

grub-common 1.98+20100804-7 works fine for me, so as far as I'm
concerned, this bug can be closed.

Thanks,
Martin



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#601093: grub-common: wrong pathnames in linux menu entry in grub.cfg if kernel on separate filesystem

[Martin Dickopp]

Package: grub-common
Version: 1.98+20100804-6
Severity: important

Severity justification: Fails to boot the linux kernel.


/boot is an ext3 filesystem on /dev/sda1

/ is an ext3 filesystem on an LVM logical volume (/dev/mapper/vg-root)

When update-grub creates /boot/grub/grub.cfg, it writes the following
section for the linux kernel:

### BEGIN /etc/grub.d/10_linux ###
menuentry 'Debian GNU/Linux, with Linux 2.6.32-5-amd64' --class debian --class 
gnu-linux --class gnu --class os {
insmod part_msdos
insmod ext2
set root='(hd0,msdos1)'
search --no-floppy --fs-uuid --set 9d53cdf2-13b4-4bc5-beb2-c0e681f1af13
echo'Loading Linux 2.6.32-5-amd64 ...'
linux   /boot/vmlinuz-2.6.32-5-amd64 root=/dev/mapper/vg-root ro  
echo'Loading initial ramdisk ...'
initrd  /boot/initrd.img-2.6.32-5-amd64
}
### END /etc/grub.d/10_linux ###

Instead of /boot/vmlinuz-2.6.32-5-amd64, the pathname
/vmlinuz-2.6.32-5-amd64 should be used; likewise for the initrd image.

It seems the issue lies either in how the grub-mkrelpath program
behaves or how it is invoked in /etc/grub.d/10_linux. 10_linux calls
'grub-mkrelpath /boot', which outputs '/boot'. I would have expected
an empty string.

The commands

  grub-mkrelpath /boot
  grub-mkrelpath /boot/
  grub-mkrelpath /boot/.

all output '/boot' while e.g.

  grub-mkrelpath /boot/vmlinuz-2.6.32-5-amd64

outputs '/vmlinuz-2.6.32-5-amd64'.

Thanks,
Martin


-- System Information:
Debian Release: squeeze/sid
  APT prefers testing
  APT policy: (500, 'testing'), (200, 'unstable')
Architecture: amd64 (x86_64)

Kernel: Linux 2.6.32-5-amd64 (SMP w/4 CPU cores)
Locale: LANG=C, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages grub-common depends on:
ii  base-files  5.9  Debian base system miscellaneous f
ii  dpkg1.15.8.5 Debian package management system
ii  gettext-base0.18.1.1-3   GNU Internationalization utilities
ii  install-info4.13a.dfsg.1-5   Manage installed documentation in
ii  libc6   2.11.2-6 Embedded GNU C Library: Shared lib
ii  libdevmapper1.02.1  2:1.02.48-3  The Linux Kernel Device Mapper use
ii  libfreetype62.4.2-1  FreeType 2 font engine, shared lib
ii  zlib1g  1:1.2.3.4.dfsg-3 compression library - runtime

Versions of packages grub-common recommends:
pn  os-prober none (no description available)

Versions of packages grub-common suggests:
pn  grub-emu  none (no description available)
pn  multiboot-doc none (no description available)
pn  xorriso   none (no description available)

-- no debconf information



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#525596: xdm: Message Login incorrect only partially cleared on next login

[Martin Dickopp]

Package: xdm
Version: 1:1.1.8-5
Severity: minor
Tags: patch

Hi,

These steps reproduce the issue:

  - Enter a username and a _wrong_ password into xdm's login window.
  - A message Login incorrect appears.
  - After a few seconds, the Login field appears again.
(So far, everything works as expected.)
  - Type the first character of a username.

This clears the part of the Login incorrect message below the baseline
(i.e. the descent part of the letter 'g').

Looking at the source code, greeter/Login.c in particular, the intention
seems to be to clear the message by means of the XClearArea function.
However, the code uses the baseline of the text as the y coordinate of
the top edge of the area to clear, therfore clearing only the part below
the baseline.

The attached patch fixes this issue.

Best regards,
Martin


-- System Information:
Debian Release: squeeze/sid
  APT prefers testing
  APT policy: (500, 'testing'), (200, 'unstable')
Architecture: amd64 (x86_64)

Kernel: Linux 2.6.29-1-amd64 (SMP w/4 CPU cores)
Locale: LANG=C, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages xdm depends on:
ii  cpp   4:4.3.3-2  The GNU C preprocessor (cpp)
ii  debconf [debconf-2.0] 1.5.26 Debian configuration management sy
ii  libc6 2.9-4  GNU C Library: Shared libraries
ii  libpam0g  1.0.1-9Pluggable Authentication Modules l
ii  libselinux1   2.0.71-1   SELinux shared libraries
ii  libsm62:1.1.0-2  X11 Session Management library
ii  libx11-6  2:1.2.1-1  X11 client-side library
ii  libxau6   1:1.0.4-2  X11 authorisation library
ii  libxaw7   2:1.0.5-2  X11 Athena Widget library
ii  libxdmcp6 1:1.0.2-3  X11 Display Manager Control Protoc
ii  libxext6  2:1.0.4-1  X11 miscellaneous extension librar
ii  libxft2   2.1.13-3   FreeType-based font drawing librar
ii  libxinerama1  2:1.0.3-2  X11 Xinerama extension library
ii  libxmu6   2:1.0.4-1  X11 miscellaneous utility library
ii  libxpm4   1:3.5.7-1  X11 pixmap library
ii  libxrender1   1:0.9.4-2  X Rendering Extension client libra
ii  libxt61:1.0.5-3  X11 toolkit intrinsics library
ii  lsb-base  3.2-22 Linux Standard Base 3.2 init scrip
ii  x11-common1:7.3+18   X Window System (X.Org) infrastruc
ii  x11-utils 7.4+1  X11 utilities

xdm recommends no packages.

Versions of packages xdm suggests:
ii  xutils1:7.3+18   X Window System utility programs m

-- debconf information:
  xdm/stop_running_server_with_children: false
  xdm/daemon_name: /usr/bin/xdm
* shared/default-x-display-manager: xdm
diff -ur xdm-1.1.8.orig/greeter/Login.c xdm-1.1.8/greeter/Login.c
--- xdm-1.1.8.orig/greeter/Login.c	2008-05-21 20:08:45.0 +0200
+++ xdm-1.1.8/greeter/Login.c	2009-04-25 21:23:50.0 +0200
@@ -629,7 +629,7 @@
 		x = ERROR_X(w, start);
 #ifdef USE_XFT
 		if (w-login.failUp == 0) {
-			XClearArea(XtDisplay(w), XtWindow(w), x, y,
+			XClearArea(XtDisplay(w), XtWindow(w), x, y - F_ASCENT(fail),
    ERROR_W(w, start), FAIL_Y_INC(w), False);
 		} else
 #endif		
@@ -650,7 +650,7 @@
 
 #ifdef USE_XFT
 	if (w-login.failUp == 0) {
-	XClearArea(XtDisplay(w), XtWindow(w), x, y,
+	XClearArea(XtDisplay(w), XtWindow(w), x, y - F_ASCENT(fail),
 		   ERROR_W(w, w-login.fail), FAIL_Y_INC(w), False);
 	} else
 #endif		

Bug#412311: magicor doesn't start (undefined symbol: mouse_seteventhandler)

[Martin Dickopp]

I could solve the issue by deinstalling svgalib1-libggi2 and
installing libsvga1 instead. Apparently the latter provides the
mouse_seteventhandler symbol, but the former does not.

Best regards,
Martin



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#471244: simutrans: Segmentation fault immediately after program start

[Martin Dickopp]

Package: simutrans
Version: 99.18~0.svn1664-1
Severity: normal

Hello,

When starting simutrans (with the command simutrans) from the command
line, it prints the following messages and dies with a segmentation
fault:

Reading low level config data ...
Initializing tombstones ...
Reading simuconf.tab successful!
Preparing display ...
Screen Flags: requested=10, actual=10
Loading font 'font/prop.fnt'
font/prop.fnt sucessfully loaded as old format prop font!
Init done.
Segmentation fault

Before it dies, it appears to open a window - there is quick rectangular
flash on the screen.

Cheers,
Martin


-- System Information:
Debian Release: lenny/sid
  APT prefers testing
  APT policy: (500, 'testing'), (200, 'unstable')
Architecture: amd64 (x86_64)

Kernel: Linux 2.6.21-5
Locale: LANG=C, LC_CTYPE=de_DE.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages simutrans depends on:
ii  libc6  2.7-6 GNU C Library: Shared libraries
ii  libgcc11:4.3.0~rc2-1 GCC support library
ii  libsdl1.2debian1.2.13-2  Simple DirectMedia Layer
ii  libstdc++6 4.3.0~rc2-1   The GNU Standard C++ Library v3
ii  simutrans-data 99.18~0.svn1664-1 transportation simulator (base dat
ii  simutrans-pak6499.18~0.svn26-1   transportation simulator (data fil
ii  zlib1g 1:1.2.3.3.dfsg-11 compression library - runtime

simutrans recommends no packages.

-- no debconf information



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#471244: simutrans: Segmentation fault immediately after program start

[Martin Dickopp]

On Mon, Mar 17, 2008 at 12:16:55AM +0100, Ansgar Burchardt wrote:
 Hi,
 
 Martin Dickopp wrote:
  When starting simutrans (with the command simutrans) from the
  command line, it prints the following messages and dies with a
  segmentation fault:
 
 Can you run Simutrans under a debugger and check weather the crash
 occurs somewhere in strlen()? If so, it should be the same problem I
 had on PowerPC [1].

Yes, it appears to be the same issue (see below). I'll apply the patch
and report what happens.

Cheers,
Martin


 gdb /usr/games/simutrans
GNU gdb 6.7.1-debian
Copyright (C) 2007 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later http://gnu.org/licenses/gpl.html
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.  Type show copying
and show warranty for details.
This GDB was configured as x86_64-linux-gnu...
Using host libthread_db library /lib/libthread_db.so.1.
(gdb) run
Starting program: /usr/games/simutrans
[Thread debugging using libthread_db enabled]
Reading low level config data ...
Initializing tombstones ...
Reading simuconf.tab successful!
Preparing display ...
[New Thread 0x2b4698554830 (LWP 22712)]
Screen Flags: requested=10, actual=10
Loading font 'font/prop.fnt'
font/prop.fnt sucessfully loaded as old format prop font!
Init done.

Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x2b4698554830 (LWP 22712)]
0x2b4695c23020 in strlen () from /lib/libc.so.6
(gdb) bt
#0  0x2b4695c23020 in strlen () from /lib/libc.so.6
#1  0x2b4695befdba in vfprintf () from /lib/libc.so.6
#2  0x2b4695c0db69 in vsprintf () from /lib/libc.so.6
#3  0x004cf734 in cstring_t::vprintf (this=0x7ba668, format=0x55d00e 
%s/, args=0x7fff15d90160)
at utils/cstring_t.cc:284
#4  0x004cf902 in cstring_t::printf (this=0xffa8, 
format=0x55d00e %s/)
at utils/cstring_t.cc:315
#5  0x005105dd in simu_main (argc=1, argv=0x7fff15d91658) at 
simmain.cc:223
#6  0x005468c5 in main (argc=1, argv=0x7fff15d91658) at simsys_s.c:602


signature.asc
Description: Digital signature

Bug#471244: simutrans: Segmentation fault immediately after program start

[Martin Dickopp]

On Mon, Mar 17, 2008 at 12:45:15AM +0100, Martin Dickopp wrote:
 I'll apply the patch and report what happens.

The patch solves the problem for me.

Thanks,
Martin


signature.asc
Description: Digital signature

Bug#412311: magicor doesn't start (undefined symbol: mouse_seteventhandler)

[Martin Dickopp]

Package: magicor
Version: 1.0-1
Severity: normal

Hi,

When typing magicor at the command line, the following messages are
shown and the program immediately exits with a non-zero exit status.

Traceback (most recent call last):
  File /usr/games/magicor, line 19, in ?
from magicor import GameEngine, getConfig
  File /usr/lib/python2.4/site-packages/magicor/__init__.py, line 12, in ?
from pygame.locals import *
  File /usr/lib/python2.4/site-packages/pygame/__init__.py, line 75, in ?
from pygame.base import *
ImportError: /usr/lib/libSDL-1.2.so.0: undefined symbol: mouse_seteventhandler

Cheers,
Martin


-- System Information:
Debian Release: 4.0
  APT prefers testing
  APT policy: (500, 'testing'), (200, 'unstable')
Architecture: amd64 (x86_64)
Shell:  /bin/sh linked to /bin/dash
Kernel: Linux 2.6.18-4
Locale: LANG=C, LC_CTYPE=de_DE.UTF-8 (charmap=UTF-8)

Versions of packages magicor depends on:
ii  libglade2-0 1:2.6.0-4library to load .glade files at ru
ii  libsdl-image1.2 1.2.5-2+b1   image loading library for Simple D
ii  libsdl-mixer1.2 1.2.6-1.1+b1 mixer library for Simple DirectMed
ii  magicor-data1.0-1data files for the magicor puzzle 
ii  python  2.4.4-2  An interactive high-level object-o
ii  python-gtk2 2.8.6-8  Python bindings for the GTK+ widge
ii  python-pygame   1.7.1release-4.1 SDL bindings for games development

magicor recommends no packages.

-- no debconf information


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#408621: O: usbmount -- automatically mount and unmount USB mass storage devices

[Martin Dickopp]

Package: wnpp
Severity: normal

I don't have enough time any more to maintain the usbmount package,
therefore I'm orphaning it.

Martin


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#192365: xserver-xfree86: [s3] accelerated solid Bresenham line drawing buggy on Trio32/64 rev 84

[Martin Dickopp]

On Sat, Jan 13, 2007 at 11:47:29PM +0100, Brice Goglin wrote:
 About 2 years ago, you reported a bug to the Debian BTS regarding a
 buggy drawing of accelerated solid Bresenham line on Trio32/64 board. A
 patch apparently fixing the problem was supposed to be in xfree86 cvs.
 Did you reproduce this problem with Xorg recently? If not, I will close
 this bug in the next weeks.

Hi Brice,

I have replaced the Trio graphics card with a different model after
a hardware failure. Therefore, I'm unfortunately unable to try to
reproduce the problem.

Cheers,
Martin


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#395915: I can't reproduce this

[Martin Dickopp]

reassign 395915 python-selinux 1.32-2
retitle 395915 python-selinux: package almost empty (except on i386)
severity 395915 grave
thanks


On Sun, Oct 29, 2006 at 02:08:47AM -0600, Manoj Srivastava wrote:
 It seems to work ok here. The _selinux.so module seems to be
  present: 
 
 __ ll /var/lib/python-support/*/_selinux.so
 4 lrwxrwxrwx 1 root root 60 2006-10-20 14:49 
 /var/lib/python-support/python2.3/_selinux.so - 
 /usr/lib/python-support/python-selinux/python2.3/_selinux.so
 4 lrwxrwxrwx 1 root root 60 2006-10-20 14:49 
 /var/lib/python-support/python2.4/_selinux.so - 
 /usr/lib/python-support/python-selinux/python2.4/_selinux.so

Hi,

Thank you for the quick followup. It seems that the python-selinux
package is the culprit. On i386 it contains a file selinux.py and two
files _selinux.so (for Python 2.3 and 2.4), but on all other archs,
these files are missing. In other words, python-selinux just consists
of empty directories and documentation on all archs except for i386.

(I'm setting the severity to grave since the package is unusable for
all non-i386 users.)

The /var/lib/python-support/python2.4/_selinux.so symlink is present
(but dangling) on my system.

Cheers,
Martin


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#395915: I can't reproduce this

[Martin Dickopp]

Hi,

When I tried to build the package in a pbuilder (sid) environment on my
amd64 machine, I noticed that the build scripts attempted to run the
swig command. Adding swig to the Build-Depends and rebuilding yielded
a correctly working package.

Cheers,
Martin


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#395915: policycoreutils: semanage doesn't work: ImportError: No module named _selinux

[Martin Dickopp]

Package: policycoreutils
Version: 1.32-1
Severity: normal

Hi,

When the semanage command is run, it prints the following error message
and exits:

  Traceback (most recent call last):
File /usr/sbin/semanage, line 24, in ?
  import seobject
File /var/lib/python-support/python2.4/seobject.py, line 24, in ?
  import pwd, string, selinux, tempfile, os, re, sys
File /var/lib/python-support/python2.4/selinux.py, line 5, in ?
  ImportError: No module named _selinux

Cheers,
Martin


-- System Information:
Debian Release: testing/unstable
  APT prefers testing
  APT policy: (500, 'testing'), (200, 'unstable')
Architecture: amd64 (x86_64)
Shell:  /bin/sh linked to /bin/dash
Kernel: Linux 2.6.18-4
Locale: LANG=C, LC_CTYPE=de_DE.UTF-8 (charmap=UTF-8)

Versions of packages policycoreutils depends on:
ii  libc62.3.6.ds1-4 GNU C Library: Shared libraries
ii  libpam0g 0.79-3.2Pluggable Authentication Modules l
ii  libselinux1  1.32-2  SELinux shared libraries
ii  libsemanage1 1.8-1   shared libraries used by SELinux p
ii  libsepol11.14-1  Security Enhanced Linux policy lib
ii  python   2.4.3-11An interactive high-level object-o
ii  python-selinux   1.32-2  Python bindings to SELinux shared 
ii  python-semanage  1.8-1   Python bindings  for SELinux polic

Versions of packages policycoreutils recommends:
pn  selinux-policy-refpolicy-targ none (no description available)

-- no debconf information


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#390425: gij-4.1: dangling symlink rmiregistry.1.gz in /etc/alternatives

[Martin Dickopp]

reopen 390425
thanks


Hi,

The issue is persistent in version 4.1.1-15.

The 'postinst' script in the package contains the following code:

update-alternatives --quiet \
--install /usr/bin/rmiregistry rmiregistry /usr/bin/grmiregistry-4.1 
$prio \
--slave /usr/share/man/man1/rmiregistry.1.gz rmiregistry.1.gz 
/usr/share/man/man1/grmiregistry-4.1.1.gz

However, the slave target 'grmiregistry-4.1.1.gz' doesn't exist.

Cheers,
Martin


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#390425: gij-4.1: dangling symlink rmiregistry.1.gz in /etc/alternatives

[Martin Dickopp]

Package: gij-4.1
Version: 4.1.1-14j1
Severity: minor

Hi,

The package sets up the symlinks /usr/share/man/man1/rmiregistry.1.gz -
/etc/alternatives/rmiregistry.1.gz -
/usr/share/man/man1/grmiregistry-4.1.1.gz, but grmiregistry-4.1.1.gz
is missing.

Cheers,
Martin


-- System Information:
Debian Release: testing/unstable
  APT prefers testing
  APT policy: (500, 'testing'), (200, 'unstable')
Architecture: amd64 (x86_64)
Shell:  /bin/sh linked to /bin/dash
Kernel: Linux 2.6.18-2
Locale: LANG=C, LC_CTYPE=de_DE.UTF-8 (charmap=UTF-8)

Versions of packages gij-4.1 depends on:
ii  gcj-4.1-base 4.1.1-14j1  The GNU Compiler Collection (gcj b
ii  libc62.3.6.ds1-4 GNU C Library: Shared libraries
ii  libgcc1  1:4.1.1-13  GCC support library
ii  libgcj7-04.1.1-14j1  Java runtime library for use with 
ii  zlib1g   1:1.2.3-13  compression library - runtime

gij-4.1 recommends no packages.

-- no debconf information


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#390433: dash: does not execute EXIT trap if shell is killed

[Martin Dickopp]

Package: dash
Version: 0.5.3-3
Severity: normal

Hi,

If I run the this two line script:

  trap 'echo EXIT trap executed' EXIT
  kill -HUP $$

I would expect it to display EXIT trap executed, but it does not
display anything.

(If I add a HUP trap, both the HUP trap and the EXIT trap are executed.)

Cheers,
Martin


-- System Information:
Debian Release: testing/unstable
  APT prefers testing
  APT policy: (500, 'testing'), (200, 'unstable')
Architecture: amd64 (x86_64)
Shell:  /bin/sh linked to /bin/dash
Kernel: Linux 2.6.18-2
Locale: LANG=C, LC_CTYPE=de_DE.UTF-8 (charmap=UTF-8)

Versions of packages dash depends on:
ii  libc62.3.6.ds1-4 GNU C Library: Shared libraries

dash recommends no packages.

-- debconf information:
* dash/sh: true


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#374441: cobex: Mistake in description: handy

[Martin Dickopp]

The word Nokie in the description might also be a mistake.
Do you mean Nokia?

Cheers,
Martin


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#365337: gs-gpl: gdi driver missing

[Martin Dickopp]

Package: gs-gpl
Version: 8.50-1.1
Severity: normal

Hi,

gs-gpl no longer contains the gdi driver used for some Samsung
printers. This driver was contained in earlier versions on the package.

Cheers,
Martin


-- System Information:
Debian Release: testing/unstable
  APT prefers testing
  APT policy: (500, 'testing'), (200, 'unstable')
Architecture: amd64 (x86_64)
Shell:  /bin/sh linked to /bin/dash
Kernel: Linux 2.6.16-1-amd64-k8
Locale: LANG=C, LC_CTYPE=de_DE.UTF-8 (charmap=UTF-8)

Versions of packages gs-gpl depends on:
ii  gs-common 0.3.9  Common files for different Ghostsc
ii  libc6 2.3.6-7GNU C Library: Shared libraries
ii  libice6   1:1.0.0-3  X11 Inter-Client Exchange library
ii  libpaper1 1.1.14-7   Library for handling paper charact
ii  libsm61:1.0.0-4  X11 Session Management library
ii  libx11-6  2:1.0.0-6  X11 client-side library
ii  libxext6  1:1.0.0-4  X11 miscellaneous extension librar
ii  libxt61:1.0.0-4  X11 toolkit intrinsics library

Versions of packages gs-gpl recommends:
ii  gsfonts   8.14+v8.11+urw-0.2 Fonts for the Ghostscript interpre
ii  psfontmgr 0.11.8-0.1 PostScript font manager -- part of

-- no debconf information


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#364513: apt: Buffer overflow in pkgDPkgPM::Go.

[Martin Dickopp]

Package: apt
Version: 0.6.43.3
Severity: normal
Tags: patch

Hello,

The method pkgDPkgPM::Go in apt-pkg/deb/dpkgpm.cc contains the following
code:

  char* list[4];
  TokSplitString(':', line, list, 5);

Since the last argument to TokSplitString is the number of elements in
the list, the code can cause a buffer overflow. I suggest the attached
patch.

The bug has caused aptitude to segfault on my system under some rare
circumstances which I cannot quite pin down, unfortunately. I don't know
a reliable way to reproduce the segfault, but hope to convince you that
the current code is incorrect anyway. ;-)

Cheers,
Martin


-- System Information:
Debian Release: testing/unstable
  APT prefers testing
  APT policy: (500, 'testing'), (200, 'unstable')
Architecture: amd64 (x86_64)
Shell:  /bin/sh linked to /bin/dash
Kernel: Linux 2.6.16-1-amd64-k8
Locale: LANG=C, LC_CTYPE=de_DE.UTF-8 (charmap=UTF-8)

Versions of packages apt depends on:
ii  libc6 2.3.6-7GNU C Library: Shared libraries
ii  libgcc1   1:4.1.0-1  GCC support library
ii  libstdc++64.1.0-1The GNU Standard C++ Library v3

Versions of packages apt recommends:
ii  debian-archive-keyring2006.01.18 GnuPG archive keys of the Debian a

-- no debconf information
--- apt-pkg/deb/dpkgpm.cc~	2005-10-19 21:19:08.0 +0200
+++ apt-pkg/deb/dpkgpm.cc	2006-04-24 00:09:46.0 +0200
@@ -623,8 +623,8 @@
 	'status: conffile-prompt: conffile : 'current-conffile' 'new-conffile' useredited distedited
 	
 	 */
-	 char* list[4];
-	 TokSplitString(':', line, list, 5);
+	 char* list[5];
+	 TokSplitString(':', line, list, sizeof(list)/sizeof(list[0]));
 	 char *pkg = list[1];
 	 char *action = _strstrip(list[2]);
 

Bug#353477: tmpreaper: --force command line option cannot be combined with --ctime or --mtime

[Martin Dickopp]

Package: tmpreaper
Version: 1.6.6
Severity: normal
Tags: patch

Hi,

If the command line options --force and --ctime are both specified, the
message error: --ctime conflicts with --mtime option is displayed and
tmpreaper exits.

The problem is the line (in the code for parsing the --ctime option)
if (flags  FLAGS_MTIME == FLAGS_MTIME)
which is equivalent to
if (flags  (FLAGS_MTIME == FLAGS_MTIME))
which is equivalent to
if (flags  1)
which is equivalent to
if (flags  FLAGS_FORCE)
so it doesn't check if the FLAGS_MTIME bit is set in flags, but if the
FLAGS_FORCE bit is set.

The same applies to the combination of --force and --mtime.

I propose the following patch:


--- tmpreaper-1.6.6/tmpreaper.c~2006-02-03 10:26:26.0 +0100
+++ tmpreaper-1.6.6/tmpreaper.c 2006-02-18 21:24:47.0 +0100
@@ -751,7 +751,7 @@
break;
 
case 'c':
-if (flags  FLAGS_MTIME == FLAGS_MTIME) {
+if ((flags  FLAGS_MTIME) == FLAGS_MTIME) {
 message(LOG_FATAL, --ctime conflicts with --mtime option\n);
 /*NOTREACHED*/
 }
@@ -759,7 +759,7 @@
break;
 
case 'm':
-if (flags  FLAGS_CTIME == FLAGS_CTIME) {
+if ((flags  FLAGS_CTIME) == FLAGS_CTIME) {
 message(LOG_FATAL, --mtime conflicts with --ctime option\n);
 /*NOTREACHED*/
 }


Cheers,
Martin


-- System Information:
Debian Release: testing/unstable
  APT prefers testing
  APT policy: (500, 'testing'), (200, 'unstable')
Architecture: amd64 (x86_64)
Shell:  /bin/sh linked to /bin/dash
Kernel: Linux 2.6.15-1-amd64-k8
Locale: LANG=C, LC_CTYPE=de_DE.UTF-8 (charmap=UTF-8)

Versions of packages tmpreaper depends on:
ii  debconf [debconf-2.0] 1.4.70 Debian configuration management sy
ii  libc6 2.3.5-13   GNU C Library: Shared libraries an

tmpreaper recommends no packages.

-- debconf information excluded


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#317516: dash: does not handle NUL characters gracefully

[Martin Dickopp]

tags 317516 + patch
thanks


Hi Gerrit,

When analysing the bug I came to the following conclusion: The loop
beginning at input.c:302 overwrites a single input line with itself,
skipping NUL characters. Therefore, after the loop the line buffer
has less characters than originally read if and only if the input line
contains NUL characters.

The pointer that is used to read from the buffer, 'parsenextc' (cf.
input.h:66), is also used as the beginning of the next line in
input.c:296. This fails if the buffer contains less characters
than originally read into it due to NUL characters.

The proposed patch (attached) keeps track of the number of skipped
characters and advances 'parsenextc' accordingly before processing
the next input line.

Cheers,
Martin
--- dash-0.5.2.ORIG/src/input.c 2003-03-08 03:43:47.0 +0100
+++ dash-0.5.2/src/input.c  2005-09-03 14:22:23.0 +0200
@@ -266,6 +266,7 @@
int something;
 #endif
char savec;
+   static size_t skipped_characters = 0;
 
while (parsefile-strpush) {
if (
@@ -285,6 +286,9 @@
flushout(errout);
 #endif
 
+   parsenextc += skipped_characters;
+   skipped_characters = 0;
+
 again:
if (parselleft = 0) {
if ((parselleft = preadfd()) = 0) {
@@ -303,6 +307,7 @@
switch (*p) {
case '\0':
p++;/* Skip nul */
+   skipped_characters++;
goto check;
 
 #ifndef SMALL

Bug#317844: usbmount: My pendrive isn't mounted anymore since version = 0.0.13

[Martin Dickopp]

On Mon, Jul 11, 2005 at 11:31:58PM +0100, Rafael Rodríguez wrote:
 Usbmount won't do any mount/umount command here since version =
 0.0.13. In fact it worked with upstream svn revision 30, but not with
 31. Syslog shows it isn't mounting/umounting anything (i.e. i don't
 have any usbmount output in syslog, just the usual kernel detection
 and usb.agent's).

Hi Rafael,

Thank you for the bug report. Could you please provide the output
of the following command?

  udevinfo -a -p $( udevinfo -q path -n /dev/sdb )

Please type the command while the USB device is plugged in, and please
replace /dev/sdb with the device node that corresponds to your device.

Cheers,
Martin


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#317516: dash: does not handle NUL characters gracefully

[Martin Dickopp]

Package: dash
Version: 0.5.2-5
Severity: minor

Hi,

dash does not handle NUL characters in scripts gracefully. To reproduce
this, create the following script:

  echo 'foo^@'
  echo 'bar^@'
  echo 'baz^@'

The characters depicted here as ^@ should be single NUL characters.

When run in dash, this script generates the following output:

  foo
  bar
  t.sh: 5: Syntax error: Unterminated quoted string

(Please note that the third line is referred to as line 5, and the error
message is also somewhat confusing.)

The code in input.c around line 304 as well as various comments indicate
that the intention is that NUL characters are silently ignored, but it
seems that this doesn't always work.

Cheers,
Martin


-- System Information:
Debian Release: testing/unstable
  APT prefers unstable
  APT policy: (200, 'unstable')
Architecture: amd64 (x86_64)
Shell:  /bin/sh linked to /bin/dash
Kernel: Linux 2.6.11-9-amd64-k8
Locale: LANG=C, LC_CTYPE=C (charmap=ANSI_X3.4-1968)

Versions of packages dash depends on:
ii  libc6   2.3.2.ds1-22 GNU C Library: Shared libraries an

dash recommends no packages.

-- debconf information:
* dash/sh: true


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#316758: Should ask whether devices should be mounted g:rw for plugdev at install

[Martin Dickopp]

severity 316758 wishlist
thanks


On Sun, Jul 03, 2005 at 06:16:10PM +0200, Ulrich Eckhardt wrote:
 I think it would be incredibly useful if usbmount asked at install
 time

Hi,

That would be against Debian's best packaging practices. Package
installs should only be interactive if no useful default configuration
is possible.

 if devices should be mounted read/write for group plugdev.

In the default configuration, USBmount considers only ext2 and ext3
filesystems. With these filesystems, owner, group, and access
permissions are not mount options, but are stored in the filesystem
itself.

On the other hand, a comment in the main configuration file
(/etc/usbmount/usbmount.conf) has an example of how to make a device
containing a vfat filesystem accessible to a specific group. The example
uses the group floppy, but of course it works equally well with the
plugdev group. I'll copy the example from the configuration file to
the README file.

 Another thing I had to change in my setup was to allow vfat as
 additional filesystem type,

There's a comment in the configration file which explains why vfat isn't
considered by default:

#
# WARNING!  The vfat filesystem does not yet fully implement sync-mounting. #
# If you include vfat in the list of filesystem types, you *MUST* make#
# sure all data is written to the medium before you remove it (e.g. run the #
# sync command in a terminal window).  Otherwise, you *WILL* lose data!   #
#

Again, I agree that this explanation should be repeated in the README
file.

Cheers,
Martin


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#317022: new udev in sid breaks usbmount(cannnot execute /sbin/udev_volume_id)

[Martin Dickopp]

tags 317022 + patch
thanks

On Wed, Jul 06, 2005 at 12:28:30AM +0800, WANG WenRui wrote:
 There is no /sbin/udev_volume_id in the new udev package, so usbmount
 failed to mount, syslog reads: 
 
 cannnot execute /sbin/udev_volume_id

Hi,

Thank you for the bug report.

I'll have a new usbmount version ready in two or three days. Another
change which would break usbmount is planned for the next udev version
(support for /etc/dev.d will be dropped), so I want to fix both issues
at the same time.

In the meantime, you can make usbmount work again by applying the
attached patch to /usr/share/usbmount/usbmount.

Cheers,
Martin

Index: usbmount
===
--- usbmount(revision 28)
+++ usbmount(revision 29)
@@ -40,9 +40,8 @@
 test -n $FILESYSTEMS || { log err FILESYSTEMS not set; exit 1; }
 test -n $MOUNTOPTIONS || { log err MOUNTOPTIONS not set; exit 1; }
 
-# Test if /sbin/udev_volume_id is executable.
-test -x /sbin/udev_volume_id \
-|| { log err cannnot execute /sbin/udev_volume_id; exit 1; }
+# Test if /sbin/vol_id is executable.
+test -x /sbin/vol_id || { log err cannnot execute /sbin/vol_id; exit 1; }
 
 
 umask 022
@@ -82,13 +81,12 @@
fi
 
# Test if the device contains a filesystem.  If it doesn't, no
-   # further action is required, but calling udev_volume_id has the
-   # side effect that the partition table is read and partition
-   # devices are created.
-   if /sbin/udev_volume_id $DEVNAME | egrep -q 
'^F:(filesystem|disklabel)$'; then
+   # further action is required, but calling vol_id has the side effect
+   # that the partition table is read and partition devices are created.
+   if /sbin/vol_id $DEVNAME | egrep -q 
'^ID_FS_USAGE=(filesystem|disklabel)$'; then
log debug $DEVNAME contains a filesystem or disklabel
 
-   fstype=`/sbin/udev_volume_id -t \$DEVNAME\`
+   fstype=`/sbin/vol_id -t \$DEVNAME\`
log debug $DEVNAME contains filesystem type $fstype
 
# Test if the filesystem type is in the list of filesystem

Bug#315083: libgtk2.0-0: breaks mozilla-firefox on amd64

[Martin Dickopp]

On Sat, Jul 02, 2005 at 05:41:41PM +0200, Sebastien Bacher wrote:
 Can you get a backtrace of the crash with libgtk2.0-0-dbg installed?

There's no crash; the program keeps running, but never opens a window.
If I force a coredump by sending it a QUIT signal in this state, I get
the following backtrace:

#0  0x2b299d25 in raise () from /lib/libpthread.so.0
#1  0x00ceb164 in nsPRUint32Key::~nsPRUint32Key ()
#2  signal handler called
#3  0x2d825a55 in poll () from /lib/libc.so.6
#4  0x2c231497 in g_main_loop_get_context () from 
/usr/lib/libglib-2.0.so.0
#5  0x2c230b80 in g_main_context_dispatch () from 
/usr/lib/libglib-2.0.so.0
#6  0x2c2310d5 in g_main_loop_run () from /usr/lib/libglib-2.0.so.0
#7  0x2b5bdc81 in IA__gtk_main () at gtkmain.c:974
#8  0x00654c85 in ?? ()
#9  0x00ce67c0 in nsPRUint32Key::~nsPRUint32Key ()
#10 0x2d778441 in __libc_start_main () from /lib/libc.so.6
#11 0x0043b0aa in ?? ()
#12 0x7fffe8a8 in ?? ()
#13 0x in ?? ()
[...]
#245 0x657269662f786f66 in ?? ()
#246 0x006e69622d786f66 in ?? ()
#247 0x in ?? ()
Cannot access memory at address 0x7000


Martin


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#315192: usbmount doesn't find vendor and model for my usb memory stick

[Martin Dickopp]

tags 315192 + pending
thanks


Hello Dan,

Thank you for the patch and sorry that it has taken me such a long time
to reply.  I have prepared a new version which my sponsor will upload
shortly.

On Mon, Jun 20, 2005 at 10:21:29PM -0400, Dan Christensen wrote:
 The patch below makes usbmount find these files.  But it's a bit of a
 hack.

Yes, finding the vendor and model name *is* a hack (with or without your
patch).

 Maybe this is a bug in udev, or a bug in my device?

I don't think so.  The problem apparently is that different drivers
provide different interfaces to the vendor and model name.  This is
probably the case because the /sys filesystem is still relatively
new.

 I wonder if usbmount should just use udevinfo instead of
 searching through /sys?

udevinfo just displays what it finds in the /sys filesystem, so that
wouldn't make much difference.  USBmount would have to search through
the udevinfo output in similar way to how it now searches through /sys.

Cheers,
Martin


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#315083: libgtk2.0-0: breaks mozilla-firefox on amd64

[Martin Dickopp]

Package: libgtk2.0-0
Version: 2.6.8-1
Severity: minor

Hello,

Since libgtk2.0-0 version 2.6.8-1, the following message is printed when
I run mozilla-firefox:

(firefox-bin:6091): Gdk-WARNING **: gdk_property_get(): length value has 
wrapped in calculation (did you pass G_MAXLONG?)

This happens on my x86 machine as well as on my amd64 machine. On the
x86 machine, firefox starts nevertheless, but on the amd64 machine, no
browser window opens, so that firefox is unusable.

I am reporting this as a minor bug since the amd64 port is not official.

Cheers,
Martin



-- System Information:
Debian Release: testing/unstable
  APT prefers unstable
  APT policy: (200, 'unstable')
Architecture: amd64 (x86_64)
Shell:  /bin/sh linked to /bin/bash
Kernel: Linux 2.6.11-9-amd64-k8
Locale: LANG=C, LC_CTYPE=de_DE.UTF-8 (charmap=UTF-8)

Versions of packages libgtk2.0-0 depends on:
ii  libatk1.0-0  1.10.1-2The ATK accessibility toolkit
ii  libc62.3.2.ds1-22GNU C Library: Shared libraries an
ii  libfontconfig1   2.3.2-1 generic font configuration library
ii  libfreetype6 2.1.10-1FreeType 2 font engine, shared lib
ii  libglib2.0-0 2.6.5-1 The GLib library of C routines
ii  libgtk2.0-0  2.6.8-1 The GTK+ graphical user interface 
ii  libgtk2.0-bin2.6.8-1 The programs for the GTK+ graphica
ii  libgtk2.0-common 2.6.8-1 Common files for the GTK+ graphica
ii  libjpeg626b-10   The Independent JPEG Group's JPEG 
ii  libpango1.0-01.8.1-1 Layout and rendering of internatio
ii  libpng12-0   1.2.8rel-1  PNG library - runtime
ii  libtiff4 3.7.2-3 Tag Image File Format (TIFF) libra
ii  libx11-6 4.3.0.dfsg.1-14 X Window System protocol client li
ii  libxcursor1  1.1.3-1 X cursor management library
ii  libxext6 4.3.0.dfsg.1-14 X Window System miscellaneous exte
ii  libxft2  2.1.7-1 FreeType-based font drawing librar
ii  libxi6   4.3.0.dfsg.1-14 X Window System Input extension li
ii  libxrandr2   4.3.0.dfsg.1-14 X Window System Resize, Rotate and
ii  libxrender1  1:0.8.3-1   X Rendering Extension client libra
ii  xlibs4.3.0.dfsg.1-14 X Keyboard Extension (XKB) configu
ii  zlib1g   1:1.2.2-4   compression library - runtime

Versions of packages libgtk2.0-0 recommends:
ii  hicolor-icon-theme0.8-1  default fallback theme for FreeDes

-- no debconf information


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#305672: usbmount: flashdisk not mounted

[Martin Dickopp]

On Mon, May 02, 2005 at 11:13:13AM +0200, Peter Stelmachovic wrote:
 I am sending the output of user.log after applying the patch.

Hi,

Thank you for the quick reply. I have once again a patch for you. :-)
This should fix the issue, so it is hopefully the last patch I have to
ask you to try. Please let me know if it really fixes the issue.

 P.S. I have noticed that the same problem occurs if the device is 
 plugged into USB1 port
 (different pc with debian stable - no usbmount, no udev, kernel 2.4 but 
 a simple script to automatically mount the flashdisk when plugged in. 
 Again the disk cannot be mounted immediately but the script needs to 
 wait for some time before  mounting a partition on the disk).

It appears that the device reports itself ready (i.e. 'cat
/sys/block/sdc/device/state' returns 'running') before it really is.
My proposed fix therefore tries to read from the device and only
continues if this succeeds (or gives up after 20 seconds).

Cheers,
Martin


--- usbmount.orig   2005-05-02 12:02:25.0 +0200
+++ usbmount2005-05-02 12:02:40.0 +0200
@@ -62,6 +62,25 @@
trap '( lockfile-remove /var/run/usbmount/.mount )' 0
log debug acquired lock /var/run/usbmount/.mount.lock
 
+   # Try to read from the device.  Some devices need a few seconds
+   # initialization time before they can be accessed.  Give up after
+   # 20 seconds.  Thanks to Peter Stelmachovic for his help with
+   # debugging this.
+   log debug testing whether $DEVNAME is readable
+   read_success=no
+   for t in 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19; do
+   if dd if=$DEVNAME of=/dev/null bs=512 count=1; then
+   read_success=yes
+   break
+   fi
+   log debug attempt $t to read from $DEVNAME failed
+   sleep 1
+   done
+   if test $read_success != yes; then
+   log err cannot read from $DEVNAME
+   exit 1
+   fi
+
# Test if the device contains a filesystem.  If it doesn't, no
# further action is required, but calling udev_volume_id has the
# side effect that the partition table is read and partition


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#305672: usbmount: flashdisk not mounted

[Martin Dickopp]

tags 305672 + pending
thanks

On Mon, May 02, 2005 at 01:24:02PM +0200, Peter Stelmachovic wrote:
 Perfect.
 Now it worked flawlessly.:-))

Good! I've prepared a new package version which my sponsor will upload
shortly.

Thank you for your help with resolving this issue!
Martin


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#305672: usbmount: flashdisk not mounted

[Martin Dickopp]

Hi,

I hope you can help me again with this bug: Please apply the patch below
to the original /usr/share/usbmount/usbmount file and report what is
logged in /var/log/user.log when you plug the device in.

Thanks!
Martin


--- usbmount.orig   2005-05-01 22:28:49.0 +0200
+++ usbmount2005-05-01 22:47:15.0 +0200
@@ -62,6 +62,21 @@
trap '( lockfile-remove /var/run/usbmount/.mount )' 0
log debug acquired lock /var/run/usbmount/.mount.lock
 
+   log info  BEGIN DEBUGGING SECTION ---
+   log info ls -l $DEVNAME: `ls -l $DEVNAME 21 || :`
+   log info /sbin/udev_volume_id $DEVNAME: `/sbin/udev_volume_id $DEVNAME 
21 || :`
+   log info fdisk -l $DEVNAME: `fdisk -l $DEVNAME 21 || :`
+   log info ls /sys$DEVPATH/device: `ls /sys$DEVPATH/device 21 || :`
+   log info cat /sys$DEVPATH/device/state: `cat /sys$DEVPATH/device/state 
21 || :`
+   log info sleeping 6 seconds
+   sleep 6
+   log info ls -l $DEVNAME: `ls -l $DEVNAME 21 || :`
+   log info /sbin/udev_volume_id $DEVNAME: `/sbin/udev_volume_id $DEVNAME 
21 || :`
+   log info fdisk -l $DEVNAME: `fdisk -l $DEVNAME 21 || :`
+   log info ls /sys$DEVPATH/device: `ls /sys$DEVPATH/device 21 || :`
+   log info cat /sys$DEVPATH/device/state: `cat /sys$DEVPATH/device/state 
21 || :`
+   log info  END DEBUGGING SECTION ---
+
# Test if the device contains a filesystem.  If it doesn't, no
# further action is required, but calling udev_volume_id has the
# side effect that the partition table is read and partition


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#305672: usbmount: flashdisk not mounted

[Martin Dickopp]

retitle 305672 usbmount: mounts only if there is a few seconds delay before 
udev_volume_id invocation
severity 305672 normal
thanks

On Thu, Apr 21, 2005 at 11:48:47AM +0200, Peter Stelmachovic wrote:
 usbmount does not mount my usb flashdisk
 the problem occurs when a hi-speed sub2 flashdisk is plugged into
 USB2 port 
 
 It seems that the drive is not ready imediately after it 
 has been plugged in.
 So I added two lines in /usr/share/usbmount/usbmount to wait for a while
 The time is set to 6s bud 5s worked as well. On other hand 4s turned 
 out to be not sufficient.

Thank you for the bug report and /var/log/user.log excerpt.

A six seconds delay has the major disadvantage that everybody has to
wait six seconds, even users with devices that don't require it.
Therefore, I will first of all try to find a solution which determines
when the device is ready and only waits as long as needed.

Cheers,
Martin


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#305672: usbmount: flashdisk not mounted

[Martin Dickopp]

Hi,

Can you please test if the problem also goes away if you apply the
patch below to the original /usr/share/usbmount/usbmount file (i.e.
the file in the package, not your modified version)?

Thanks!
Martin


--- usbmount.orig   2005-04-21 13:33:18.0 +0200
+++ usbmount2005-04-21 13:34:37.0 +0200
@@ -135,6 +135,7 @@
fi
else
log debug $DEVNAME does not contain a filesystem or disklabel
+   fdisk -l $DEVNAME
fi
 fi
 



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#303735: usbmount: mounting in specific directory

[Martin Dickopp]

On Fri, Apr 08, 2005 at 03:06:12PM +0200, Jakub Turski wrote:
 How about such feature: when new device is found, vendor and model are
 checked against a table in /etc, and device is mounted in a directory
 specified in this table. If not found, device is mounted in some
 generic directory, just like it is now.

 Having this, I could have my camera mounted on /mnt/minolta, my palm on
 /mnt/palm, and everything else on /mnt/usbstorage. I hope it is not
 hard to implement, as there's already symlink created in /var...

Hi Jakub,

Thank you for your suggestion.

The existing symlink feature is supposed to solve this problem, i.e. if
you create a symlink /mnt/minolta pointing to /var/run/usbmount/Minolta
(or whatever the model name of your camera is), you can use /mnt/minolta
to access your camera while it is mounted.

The current version (0.0.10) also runs hook scripts when a device is
mounted or unmounted, so if you prefer, you can create a symlink in
/mnt pointing to to mountpoint directly. Please have a look at
/etc/usbmount/mount.d/00_create_model_symlink and
/etc/usbmount/umount.d/00_remove_model_symlink to see how they manage
the symlink in /var/run/usbmount. The vendor and model name are
available to the scripts in /etc/usbmount/mount.d in environment
variables; see the README file for details.

Please let me know if this does what you want. Otherwise, please tell me
why a symlink to the mountpoint (as opposed to specifying the mountpoint
directly) doesn't do what you want.

Thanks,
Martin


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#290984: yafray doesn't work with blender

[Martin Dickopp]

On Sun, Apr 10, 2005 at 01:15:55PM +1000, Brian May wrote:
 Please consider this bug a wishlist:
 
 a) error message more informative (I guess it makes sense now, but it
 didn't make sense before...)
 
 b) default settings should work out-of-the-box.
 
 Thanks.

(Please note that I'm not the maintainer of yafray or blender, but
merely a user interested in the development of these programs.)

 Also, can you please explain to me the difference between XML export
 and blender plugin?

In XML export mode, blender creates a yafray input file (the XML file)
and then calls the yafray program in the same way as you would from the
command line. In plugin mode, blender loads yafray as a plugin, so that
it doesn't have to export the scene to an intermittent file.

 When would you want to use one or the other?

XML export was the only option before yafray was usable as a plugin.
Nowadays, I'd usually recommend plugin mode, which has the advantage
that you can watch the image build up in the output window while it is
being rendered. I'd only recommend XML export if you need the exported
file for some reason (e.g. you want to run yafray on a different
machine).

Martin


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#303354: fontconfig: /etc/fonts/conf.d/30-debconf-no-bitmaps.conf should be relative symlink

[Martin Dickopp]

Package: fontconfig
Version: 2.3.1-2
Severity: minor

Hi,

The package creates /etc/fonts/conf.d/30-debconf-no-bitmaps.conf as an
absolute symbolic link to /etc/fonts/conf.d/no-bitmaps.conf. According
to policy 10.5, the link should be relative.

Cheers,
Martin


-- System Information:
Debian Release: 3.1
Architecture: amd64 (x86_64)
Kernel: Linux 2.6.11-9-amd64-k8
Locale: LANG=C, LC_CTYPE=de_DE.UTF-8 (charmap=UTF-8)

Versions of packages fontconfig depends on:
ii  debconf [debco 1.4.47Debian configuration management sy
ii  defoma 0.11.8-0.1Debian Font Manager -- automatic f
ii  gsfonts-x110.17  Make Ghostscript fonts available t
ii  libc6  2.3.2.ds1-20.0.0.1.pure64 GNU C Library: Shared libraries an
ii  libfontconfig1 2.3.1-2   generic font configuration library
ii  ttf-bitstream- 1.10-3The Bitstream Vera family of free 
ii  ttf-freefont   20031008-1.1  Freefont Serif, Sans and Mono True
ii  ucf1.17  Update Configuration File: preserv

-- debconf information:
* fontconfig/subpixel_rendering: Automatic
* fontconfig/enable_bitmaps: false
* fontconfig/rendering_type: Subpixel rendering (LCD screens)
* fontconfig/hinting_type: Native
* fontconfig/enable_autohinter: true


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#266681: blender: yafray does not work as a plugin

[Martin Dickopp]

Hi,

This bug has been fixed a long time ago.
May I close the bug report?

Cheers,
Martin


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#290984: yafray doesn't work with blender

[Martin Dickopp]

On Tue, Jan 18, 2005 at 01:16:10PM +1100, Brian May wrote:
 When trying to render an image using yafray, I get the following
 error:
 
 Starting scene conversion.
 Scene conversion done.
 No export directory set in user defaults!

Do you use yafray as a blender plugin or via XML export? To find out,
click on the YafRay panel in the Render panel and check if the xml
button is selected (XML export) or not (plugin).

If you use XML export, you have to set a valid (i.e. existing) directory
where blender can write the XML file. To verify this, go to the User
Preferences (the window type button that looks like an i), click on
File Paths, and check what YFexport is set to. It should have a
trailing slash. I personally set it to ./, so that the XML file is
created in the current working directory.

 This is with blender in sid:
 ii  blender2.35-1 Very fast and versatile 3D modeller/renderer
 
 As far as I can tell, this is a different problem to bug #266677 and
 #266681, as I don't get any undefined symbol errors.

I agree. First of all, #266677 and #266681 are fixed in blender 2.35,
and secondly, if you use XML export (it sounds as if you do), these bugs
don't affect you anyway.

Cheers,
Martin


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#303376: usbmount removes all model-name based symlinks when one usb drive is removed

[Martin Dickopp]

On Wed, Apr 06, 2005 at 01:44:53PM +0200, Martin Steigerwald wrote:
 I found that usbmount does remove the model-name based symlinks (in
 /var/run/usbmount) of all USB drives when only one of them is removed.
 It seems that the test whether the symlinks points to a mounted
 directory does not work.

Thanks for the bug report, I will investigate.

 I did not investigate the cause for this but simply reverted to the link
 creating behaviour I did myself for a previous version of usbmount:
 
 # Teamix: Determine user of the current X session
 XUSER=`who | awk '$2 == :0 {print $1}'`
 
 # Teamix: Create a symlink in the Desktop directory of that user
 if test -n $XUSER; then
   log info creating symlink /home/$XUSER/Desktop/`basename $mountpoint` - 
 $mountpoint
   ln -sf $mountpoint /home/$XUSER/Desktop/`basename $mountpoint`
 fi
 
 # Teamix: Remove symlink in Desktop directory of that user again
 if test -n $XUSER  test -L /home/$XUSER/Desktop/`basename 
 $mountpoint`; then
   log info removing symlink /home/$XUSER/Desktop/`basename $mountpoint`
   rm /home/$XUSER/Desktop/`basename $mountpoint`
 fi

I my opinion, an automatically run script should not try to modify a
user's home directory.

My solution to the problem of making resources available to the current
desktop user is to make them available to the members of a Unix group.
I then use PAM to ensure that exactly the current desktop user is a
member of the groups. (Incidently, the user can circumvent this if she
or he can create setgid executables anywhere on the system.)

Cheers,
Martin


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#303434: /var/run/usbmount directory doesn't exist

[Martin Dickopp]

On Wed, Apr 06, 2005 at 07:33:27PM +0200, Christian Marillat wrote:
 The /var/run/usbmount directory doesn't exist and usbmount is unable
 to write the lock file :

Thanks for the bug report. It seems the directory is created when the
package is freshly installed, but not when it is updated from a previous
version (which didn't have /var/run/usbmount). Have you updated to 0.0.8
from an earlier version of usbmount?

Thanks,
Martin


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#303434: /var/run/usbmount directory doesn't exist

[Martin Dickopp]

tags 303434 pending
thanks

On Wed, Apr 06, 2005 at 08:33:07PM +0200, Christian Marillat wrote:
 Martin Dickopp [EMAIL PROTECTED] writes:
 
  On Wed, Apr 06, 2005 at 07:33:27PM +0200, Christian Marillat wrote:
  The /var/run/usbmount directory doesn't exist and usbmount is unable
  to write the lock file :
 
  Thanks for the bug report. It seems the directory is created when the
  package is freshly installed, but not when it is updated from a previous
  version (which didn't have /var/run/usbmount). Have you updated to 0.0.8
  from an earlier version of usbmount?
 
 Yes, from your unofficial package (I don't remember the version).

Okay, then my understanding of what the problem is was correct.

 I think the best way is to package this empty directory in the package
 and let dpkg to add or remove this directory.

I have done so in usbmount 0.0.9, which my sponsor will upload shortly.

Cheers,
Martin


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#303376: usbmount removes all model-name based symlinks when one usb drive is removed

[Martin Dickopp]

tags 303376 pending
thanks

On Wed, Apr 06, 2005 at 01:44:53PM +0200, Martin Steigerwald wrote:
 I found that usbmount does remove the model-name based symlinks (in
 /var/run/usbmount) of all USB drives when only one of them is removed.

The bug is fixed in usbmount 0.0.9, which my sponsor will upload
shortly.

Cheers,
Martin


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#298167: blender: insecure writing to /tmp/quit.blender

[Martin Dickopp]

reopen 298167
tags 298167 - security
severity 298167 wishlist
retitle 298167 blender: please revert writing quit.blend with O_EXCL and make 
U.tempdir default to a userdir instead
thanks


Hi,

Creating quit.blend with the O_EXCL flag has a major drawback: after the
file has been written for the first time, all further attempts to write
it again will fail. This makes the feature quite useless, IMHO.

A much better solution, IMHO, would be to make U.tempdir default to a
directory that is not world-writable, e.g. the user's home directory.
Note that the user can configure the value of U.tempdir within Blender,
so she or he can change it to a directory of her or his liking.

This solution would also be more robust with respect to future security
bugs, should future Blender versions use U.tempdir for other purposes
than just creating quit.blend there.

I therefore suggest replacing this code

if (BLI_streq(U.tempdir, /)) {
char *tmp= getenv(TEMP);
strcpy(U.tempdir, tmp?tmp:/tmp/);
}

with

if (BLI_streq(U.tempdir, /)) {
strcpy(U.tempdir, BLI_gethome());
}

in source/blender/src/usiblender.c.

Thanks,
Martin


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]