Package: firmware-realtek
Version: 0.44
Severity: serious
Justification: 12.5, stretch release policy 1
The RTL8192SU/rtl8192sfw.bin file is not present upstream. Its actual
upstream source (presumably Realtek?) is not documented in the
copyright file. Please document it.
$ cd git/linux-firmware
found 590521 1.8.15-3
thanks
Boris Pek wrote:
>> >* Build with -DQTC_MODIFY_MOZILLA=false:
>> > do not modify Iceweasel and/or Firefox settings. (Closes: #590521)
>> Again?
>> http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=590521#24
>> http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=
found 590521 1.8.15-2
thanks
The bug is, indeed, not fixed. Steps to reproduce:
* Under XFCE 4.8, run xfce4-appearance-settings.
* Select the QtCurve style. This previews the style, which in turn
modifies userChrome.css.
* Select the style used previously.
* Close xfce4-appearance-settings.
--
Package: unhide
Version: 20110113-3
Severity: serious
Justification: wheezy RC policy 3
The unhide postinst script switches the unhide alternative to manual
mode, which is a violation of section 3 of the wheezy RC policy. The
manual mode is provided for the system administrator.
The use of the al
Thorsten Glaser wrote:
> I actually like the idea of making editorrc a slave to editor…
And this solution would indeed probably be the best, if it is possible.
--
To UNSUBSCRIBE, email to debian-bugs-rc-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@list
Thorsten Glaser wrote:
> I think that’s a lot more silly. Besides, wrapping
> would slow down and bloat.
It avoids the problem of editor and editorrc being out of sync.
Slowdown is imperceptible (which is the only thing that matters, since
joe is always used interactively). There is no memory c
Piotr Engelking wrote:
> #!/bin/bash
> if [ "x$(basename -- $0)" = xeditor ]; then
Sorry, insufficient quoting, the above should be of course:
if [ "x$(basename -- "$0")" = xeditor ]; then
--
To UNSUBSCRIBE, email to debian-bugs-rc-requ...@lists.debia
I think that all of this silliness could be avoided, while perfectly
reproducing current behavior, if jstar, joe, etc were wrappers, not
symlinks:
#!/bin/bash
if [ "x$(basename -- $0)" = xeditor ]; then
progname=jstar
else
progname="$0"
fi
exec -a "$progname" /usr/lib/joe/joe.rea
Thorsten Glaser wrote:
>>There seems to be no point of jupp sharing the editorrc alternative
>>with joe, especially considering that neither program is compatible
>>with the shipped rc files of the other anymore.
>
> The problem is that joe/jupp use argv[0] + "rc" as name to look
> for, so it’s e
Thorsten Glaser wrote:
> Do you think it would work if I just add
> /etc/jupp/editorrc@ -> ../joe/editorrc
> and then manage /etc/joe/editorrc with update-alternatives
> just like joe does?
There seems to be no point of jupp sharing the editorrc alternative
with joe, especially considerin
Package: jupp
Version: 3.1.17-2
Severity: serious
Justification: wheezy RC policy §3
Jupp hijacks the editorrc alternative from joe, moving it from
/etc/joe/editorrc to /etc/jupp/editorrc. This breaks joe if it is set
as the default system editor:
# dpkg-query -W joe
joe 3.7-2
# ls -F /etc/jo
Package: binutils-multiarch
Version: 2.21-1
Severity: grave
Justification: renders package unusable
Attempting to install binutils-multiarch 2.21-1 results in the
following error:
dpkg: error processing
/var/cache/apt/archives/binutils-multiarch_2.21-1_i386.deb (--unpack):
trying to overwrite `
Package: gtk2-engines-qtcurve
Version: 1.5.1-1
Severity: serious
Justification: squeeze RC policy section 3
The ~/.mozilla/firefox/*.default/chrome/userChrome.css file contains
the following snippet:
menubar > menu { color: #141312 !important; } menubar >
menu[_moz-menuactive="true"][open="false"
Package: binutils-multiarch
Version: 2.20.51.20100518-1
Severity: grave
Justification: renders package unusable
Binutils-multiarch breaks ld by diverting ld.bfd away without providing
a replacement:
$ ld
bash: /usr/bin/ld: No such file or directory
$ whereis ld
ld: /usr/bin/ld /usr/bin/ld.gold /u
Package: scummvm-data
Version: 1.0.0-3
Severity: serious
Justification: squeeze RC policy §2
With scummvm 1.0.0-2 installed, attempting to install scummvm-data
results in the following error:
Unpacking scummvm-data (from .../scummvm-data_1.0.0-3_all.deb) ...
dpkg: error processing
/var/cache/apt/
Package: otf-yozvox-yozfont-antique
Version: 12.18~beta28-dfsg-1
Severity: serious
Justification: squeeze RC policy §2
With otf-yozvox-yozfont 12.18~beta26-dfsg1-1 installed, attempting to
install otf-yozvox-yozfont-antique results in the following error:
Unpacking otf-yozvox-yozfont-antique (fro
Thorsten Glaser :
> Zack says it’s okay for me to close this. For your reference,
> the commit diff can be retrieved using the following command:
>
> % cvs -d :pserver:c...@cvs.fefe.de:/cvs -z9 rdiff -ur1.7 -r1.8 dietlibc/README
Excellent, thanks.
It may be useful to update the license on http:/
Gerrit Pape :
> The README states that the dietlibc is licensed under GPL-2, which
> actually allows to apply any later version. I do not see anything wrong
> here.
It does not allow that, licensing software under the "or later"
version of the GPL is just a practice strongly encouraged by FSF, s
Package: dietlibc
Version: 0.32-1
Severity: serious
Justification: squeeze RC policy §1
The debian/copyright file claims that dietlibc is licensed under GPL
version 2 or later. According to the README file, it is actually
licensed under GPL version 2.
Please refer to the correct license in debian
Package: traceroute-nanog
Version: 6.4.2-1
Severity: serious
Justification: squeeze RC policy §5.c
FHS specifies that /bin, /usr/bin, and /usr/local/bin contains programs
for use by all users. In particular, root-only programs are placed in
/sbin, /usr/sbin, and /usr/local/sbin, instead.
As trace
reopen 552010
thanks
After upgrade from 2.9-25 to 2.10.1-5, name resolution breaks if the
'nameserver' option is not present in /etc/resolv.conf:
# ping debian.org
ping: unknown host debian.org
# cat /etc/resolv.conf
options ndots:0
# echo nameserver 127.0.0.1 >> /etc/resolv.conf
# ping -c 1 debi
reopen 545366
thanks
dselect 1.15.4.1 (with dpkg 1.15.4.1) still does not distinguish
between new and old packages. This makes dselect mostly useless on
upgrades and completely useless on adding new repositories.
The only difference between 1.15.4 and 1.15.4.1 is that former marks
new and old pac
Package: firmware-linux-nonfree
Version: 0.19
Severity: serious
Justification: squeeze RC policy §2
With firmware-linux 0.18 installed, attempting to install
firmware-linux-nonfree results in the following error:
Unpacking firmware-linux-nonfree (from
.../firmware-linux-nonfree_0.19_all.deb) ...
Package: xz-utils
Version: 4.999.8beta-1
Severity: serious
Justification: squeeze RC policy §2
With lzip 1.7-1 installed, attempting to install xz-utils results in the
following error:
Unpacking xz-utils (from .../xz-utils_4.999.8beta-1_i386.deb) ...
dpkg: error processing
/var/cache/apt/archives
Package: slim
Version: 1.3.0-2
Severity: grave
Tags: security
Justification: user security hole
X-Debbugs-Cc: t...@security.debian.org
If scrot is installed, users are able to overwrite arbitrary files in the
filesystem.
Pressing F11 on the slim login screen runs scrot with the root rights to
sav
2009/7/16 Nico Golde :
> This is not really a bug but a feature, you can disable it
> editing /etc/slim.conf. However I agree this is not really a
> nice feature in a default configuration. I think a big fat
> note to README.Debian should be added to warn users of the
> possible implications.
The
Package: slim
Version: 1.3.0-2
Severity: grave
Tags: security
Justification: user security hole
X-Debbugs-Cc: t...@security.debian.org
Typing 'console' in the SLiM login prompt opens a gnome terminal running
login(1). Opening a new terminal from the terminal menu (File -> Open
Terminal) opens a gn
block 420129 by 481240
severity 420129 important
thanks
The behaviour you observe is caused by the fact that libnids by default
filters packets with incorrect checksum, and that, if checksum calculation
is done in hardware, libpcap is only able to capture outgoing packets before
the checksum is co
Package: keytouch
Version: 2.3.2-2
Severity: serious
Justification: Policy 12.5, 12.7
The keytouch binary package doesn't ship debian/changelog nor
debian/copyright, which is a violation of section 1 of the RC policy.
-- System Information:
Debian Release: lenny/sid
APT prefers testing
APT p
Clint Adams <[EMAIL PROTECTED]> wrote:
> Googling for lenny rc policy, I only find mails from you. Where can I
> find this document?
http://release.debian.org/etch_rc_policy.txt
(Despite the name, it applies to lenny as well, for now. Sorry for confusion.)
--
To UNSUBSCRIBE, email to [EMAIL P
Package: gnome-applets
Version: 2.14.3-4
Severity: serious
Justification: Policy 2.2.1
The gweather/Locations.xml file seems to be autogenerated from
gweather/Locations.xml.in, which isn't included in the Debian source
package. Please include it.
(While this is a violation of section 1 of the et
reopen 396949
thanks
Version 2.8.7dev2 attempts to fix the bug by checking if the user owns
the .mime.types and .mailcap files before opening them. The assumption
that user trusts the files he owns, is, however, flawed. Consider, for
an example:
* files downloaded by the user
* files unpacked by
maximilian attems <[EMAIL PROTECTED]> wrote:
forcemerge 397139 397207
stop and over
can you please look for duplicates, before reporting!
Hmm... For some strange reason http://bugs.debian.org/src:linux-2.6
redirects to
http://bugs.debian.org/cgi-bin/pkgreport.cgi?src=linux-2.6;dist=unstable,
Package: linux-source-2.6.18
Version: 2.6.18-4
Severity: serious
Justification: no longer builds from source
debian/patches/bugfix/copy-user-highpage.patch adds fourth argument to
copy_user_highpage(), but fails to update function call in mm/hugetlb.c
accordingly:
CC mm/hugetlb.o
mm/hugetl
Package: lynx-cur
Version: 2.8.7dev1-1
Severity: grave
Tags: security
Justification: user security hole
Lynx attempts to use the .mime.types and .mailcap files located in the
current directory:
$ strace lynx -dump 2>&1 | grep '^open("[^/]'
open(".mailcap", O_RDONLY) = -1 ENOENT (No
On 08/04/06, Justin Pryzby <[EMAIL PROTECTED]> wrote:
> The etch release policy states that:
>
> | http://release.debian.org/etch_rc_policy.txt
> | If two packages cannot be installed together, one must list
> | the other in its "Conflicts:" field.
>
> At least one of these packages mus
Package: cdrtools
Severity: serious
Justification: Policy 2.3
In cdrtools 2.01.01a03 license of several makefiles have been changed to a
custom version of CDDL, which is a non-GPL-compatible license. These
makefiles are used to build GPL-licensed binaries, which is a violation of
paragraph 3 of th
37 matches
Mail list logo
