I also noticed that at least sodomized-sheep.cow had been removed in
the latest update, breaking a login script on my system. Please
restore the previous cows. Thanks!
Package: less
Version: 487-0.1+b1
Severity: important
Tags: upstream
This is not a bug in any current Debian package, but an FYI about an
upstream regression.
As of version 520 the following test-case will wait a full 5 seconds
before printing "hello":
(echo "hello"; sleep 5; echo "bye bye")
Package: cowsay
Version: 3.03+dfsg1-16
Severity: wishlist
In 3.03+dfsg1-13 the cowsay-off package was created in response to
#769194. I used one of those cows in a login script and on upgrade had
an error about a nonexisting cow.
It took me longer than it otherwise would have to figure out what w
Package: apt-listchanges
Version: 2.85.13+nmu1
Severity: minor
I discovered this bug after leaving the message buffer of
apt-listchanges around for days in a screen session while the pager
was set to read the changelog:
Fetched 186 MB in 10s (17.8 MB/s)
Reading changelogs... Done
Exce
Package: pdns-server
Version: 3.4.1-3
Severity: important
I upgraded to 3.4.0 and my authoritative nameserver stopped working,
nothing major was noted in the Changelog or NEWS file, I think it's
worth a mention that the value for allow-axfr-ips changed from
0.0.0.0/0,::/0 to 127.0.0.0/8,::1, this
Package: pdns-server
Version: 3.4.1-3
Severity: grave
When I upgraded from 3.3.* I started getting this in error log:
Nov 17 19:28:30 u pdns[14671]: Loading '/usr/lib/powerdns/libbindbackend.so'
Nov 17 19:28:30 u pdns[14671]: Unable to load module
'/usr/lib/powerdns/libbindbackend.so': /
Package: iodine
Version: 0.6.0~rc1-19
Severity: serious
I couldn't find out how to re-open an existing bug but this issue that
I reported & was resolved a while ago is now happening again:
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=612723
Upgrading from 0.6.0~rc1-12 to 0.6.0~rc1-19 yields
Package: cronolog
Version: 1.6.2+rpk-1
Severity: wishlist
Tags: upstream
When you run cronolog successfully it exits with a non-zero exit code
due to this bit of code in main():
n_bytes_read = read(0, read_buf, sizeof read_buf);
if (n_bytes_read == 0)
{
exit(3);
}
I.e. if
Package: cronolog
Version: 1.6.2+rpk-1
Severity: normal
On my system:
$ cronolog --version
cronolog version 1.6.2
$ cronolog --vesion
Segmentation fault
A backtrace for this is:
Program received signal SIGSEGV, Segmentation fault.
0x77ab5f4c in ?? () from /lib/x86_64-linux-gnu/libc.so.6
Package: bash
Version: 4.2-1
Severity: important
Tags: upstream
The "&&" operator in bash seems to inhibit backgrounding when run over
ssh. You can try it with the following:
ssh localhost "cd /tmp && nohup sleep 10 >/dev/null 2>&1 &"
The above command will wait for sleep(1) to finish before
Package: debian-installer
Severity: wishlist
Tags: d-i
I use Dvorak and every time I install Debian I have to hunt down keys
on a completely different layout in order to complete the region
etc. settings.
Since a keyboard is a prerequisite to operate the rest of the
installer the very first quest
Package: po4a
Version: 0.41-1
Severity: important
The subject pretty much says it all. I have a custom perlbrew-based
perl installed in my $PATH and when I start po4a it gives:
$ po4a
Can't locate Locale/Po4a/Chooser.pm [...]
This is because the shebang specifies the perl from the
enviro
Package: rtorrent
Version: 0.8.9-2
Severity: important
Tags: patch
The log.execute option in ~/.rtorrentrc has no effect. This was patched
upstream months ago but a new version still isn't out. Can the patch be
included in Debian?
The upstream bug report: http://libtorrent.rakshasa.no/ticket/26
Package: apt-listchanges
Version: 2.85.7
Severity: important
I have a cronjob that runs "aptitude clean" daily, I left the
interactive apt-listchanges dialog open during the run of this
cronjob, this is what happened:
Get:236 http://ftp.de.debian.org/debian/ testing/main lm-sensors amd64
1:3
Package: iodine
Version: 0.6.0~rc1-4
Severity: serious
When I upgrade iodine it'll overwrite my changes in
/etc/default/iodine. This appears to be done by this bit in
debian/postinst:
db_get iodine/start_daemon
START_IODINED=$RET
db_get iodine/daemon_options
Package: bind9-host
Version: 1:9.7.2.dfsg.P3-1
Severity: wishlist
On a system with only bind9-host installed you'll get scary notices
about bind9 vulnerabilities on upgrade, even though the system doesn't
actually use bind9 at all, just its host(1) utility.
It would be much friendler to administ
Package: debconf
Version: 1.5.36
Severity: important
Note: The bug isn't being filed on the machine that has this issue,
but they have the same debconf version.
A dpkg --configure -a invocation of mine hung after I had a rude
shutdown of a system. When I traced it it turned out that debconf was
Package: wnpp
Severity: wishlist
* Package name: libdir-self-perl
Version : 0.10
Upstream Author : Lukas Mai
* URL : http://search.cpan.org/dist/Dir-Self/
* License : Artistic or GPL-1+
Programming Lang: Perl
Description : A Perl module for finding the
Package: rtorrent
Version: 0.8.6-1
Severity: important
rtorrent regularly crashes with a few std::* error messages, this is
one of them:
"rtorrent: std::bad_allocate"
After a crash it has to be manually restarted. I tried for a while to
run rtorrent under gdb, but these bugs are a failing as
On Sat, Oct 2, 2010 at 22:20, Timo Juhani Lindfors wrote:
> "Debian bug at v.nix.is" writes:
>> Being reminded >10 times in 4 hours I think is a bit too much.
>
> Since shutdown does not know who saw the message somebody could login
> 10 minutes before shutdown an
On Sat, Oct 2, 2010 at 20:41, Petter Reinholdtsen wrote:
>> This is way too noisy if you have a scheduled reboot and multiple
>> users on your system, they all have to C-l their terminals after
>> each message.
>
> Why is this too noisy? It seem just noisy enought for me.
Because it's enough to
Package: sysvinit
Version: 2.88dsf-12
Severity: wishlist
Tags: upstream
When I call shutdown like this:
sudo shutdown -r '+240' "We're rebooting in 4 hours for kernel and libc
updates"
It will proceed to wall() all consoles 17 times before shutting down,
or at T-minus:
240
150
Package: munin
Version: 1.4.5-2
Severity: normal
Tags: upstream patch
By default munin assumes that graph data is in ISO-8859-1, it should
use Encode::Guess to support data in other encodings, e.g. UTF-8.
I've filed this upstream with a patch, but it's been scheduled for
munin 1.6: http://munin-m
Package: iodine
Version: 0.6.0~rc1-2
Severity: minor
>From /usr/share/doc/iodine/README.Debian:
* Instruction for setting up the infrastructure (including delegating a new
subdomain) are included in /usr/share/doc/iodine/README and in iodined(8).
There's no such file, that should
Package: ptunnel
Version: 0.71-2
Severity: minor
Tags: upstream
This:
Contact details
---
You can the author, Daniel Stoedle, here:
Should say "You can contact the author [...]".
-- System Information:
Debian Release: squeeze/sid
APT prefers testing
APT policy: (
Package: iodine
Version: 0.6.0~rc1-1
Severity: important
iodine/iodined support spaces in passwords, e.g.:
iodine -P "a very long password with spaces"
But the Debian init.d/iodined script doesn't, because you do:
DAEMON_ARGS="-u iodine -t $CHROOTDIR -P $IODINED_PASSWORD $IODINED_ARGS"
Package: nstx
Version: 1.1-beta6-5
Severity: normal
Tags: upstream
>From the source code:
nstx_tuntap.c
169:sock.sin_port = htons(53);
190:sock.sin_port = htons(53);
198: "listening on UDP-Port 53 on %s\n",
204: fprintf(stderr, "Permission den
On Tue, Sep 14, 2010 at 20:30, Niko Tyni wrote:
> On Tue, Sep 14, 2010 at 02:18:15PM +0000, v.nix.is wrote:
>> Package: perl
>> Version: 5.10.1-14
>> Severity: normal
>> Tags: upstream
>>
>> Text::Tabs which is shipped with both upstream perl and the Debian
Package: wamerican-huge
Version: 6-3
Severity: minor
When I install this package I get:
$ sudo aptitude install wamerican-huge
The following NEW packages will be installed:
wamerican-huge
0 packages upgraded, 1 newly installed, 0 to remove and 17 not upgraded.
Need to ge
Package: perl
Version: 5.10.1-14
Severity: normal
Tags: upstream
Text::Tabs which is shipped with both upstream perl and the Debian
version has this license, (highlight by me):
Copyright (C) 1996-2002,2005,2006 David Muir Sharnoff. Copyright
(C) 2005 Aristotle Pagaltzis This module may be
On Mon, Aug 23, 2010 at 16:00, gregor herrmann wrote:
> I have no idea if this is enough or not; Nicholas and v.nix.is, what
> do you think?
I didn't track down what commits exactly were pertinent to fixing
this, I just did a log from 3.21 to the master at the time and picked
all th
Package: libtest-harness-perl
Version: 3.20-1
Severity: important
Test::Harness 3.22 is now available on the CPAN. This solves Debian
bug #592506.
Please push it to testing so that it'll be in the next stable.
-- System Information:
Debian Release: squeeze/sid
APT prefers testing
APT policy:
retitle 592506 Serious regressions in 3.20-1 solved in upstream Git,
probable blocker for stable
kthx
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Package: libtest-harness-perl
Version: 3.20-1
Severity: important
Tags: patch
Since upstream released 3.17_01 there have been two serious
regressions with the handling of:
* Passing arguments to executable non-Perl tests:
https://rt.cpan.org/Ticket/Display.html?id=59186#txn-801685
* Runn
Package: clang
Version: 2.7-2
Severity: wishlist
With clang installed:
$ sudo update-alternatives --config cc
There is only one alternative in link group cc: /usr/bin/gcc
Nothing to configure.
It would be great to be able to configure clang as the system
/usr/bin/cc. Then I could jus
Package: tomcat6
Version: 6.0.28-1
Severity: minor
When I do:
sudo service tomcat6 stop
sudo aptitude upgrade
tomcat6 will be started again as part of the upgrade:
Setting up tomcat6 (6.0.28-1) ...
Installing new version of config file /etc/tomcat6/web.xml ...
Starting Tomca
Package: rtorrent
Version: 0.8.6-1
Severity: important
I was running rtorrent and did:
C-x upload_ratio = 200 RET
At which point rtorrent crashed with:
rtorrent: Rate::insert(bytes) received out-of-bounds values..
When I start it again the crash isn't reproducable.
-- System Informat
Package: bash
Version: 4.1-3
Severity: normal
I'm not sure when this started occuring. \W which should report the
basename "boot" in /boot reports "bott":
$ export PS1="\W \w "
~ ~ cd /boot
bott /boot
I can't reproduce it on my Ubuntu machine which has "GNU bash, version
4.1.5(1)-rel
Package: logwatch
Version: 7.3.6.cvs20090906-1
Severity: minor
$ dpkg -L logwatch|grep '#'
/usr/share/logwatch/scripts/services/.#http.1.29
/usr/share/logwatch/scripts/services/.#http.1.31
/usr/share/logwatch/scripts/services/.#http.1.19
/usr/share/logwatch/scripts/service
On Wed, Jun 30, 2010 at 20:15, Christian Garbs wrote:
> On Mon, Jun 07, 2010 at 08:56:43PM +, Debian bug at v.nix.is wrote:
>
>> I looked at the source code for the 1.18 tarball and it doesn't work
>> as expected because you're using the --user option to ps:
>&g
Package: gnunet-server
Version: 0.8.1b-4
Severity: minor
The README.Debian for the server says this:
gnunet-server for Debian
By default, the gnunet-server starts automatically when the system is
bootet.
On Sun, Jun 6, 2010 at 11:19, Christian Garbs wrote:
> On Mon, May 31, 2010 at 09:00:01PM +0000, v.nix.is wrote:
>
>> The reniced program is really limited by the hardcoded use of a
>> `comm' argument to ps(1). It should optionally allow for using
>> `cmd'. Or
On Sun, Jun 6, 2010 at 10:20, Christian Garbs wrote:
> On Mon, May 31, 2010 at 09:02:09PM +0000, v.nix.is wrote:
>
>> This is hardcoded in the reniced source code:
>>
>> my $debug = 0;
>>
>> The debug output is really helpful to find if your rules ar
Package: libyaml-syck-perl
Version: 1.07-1
Severity: wishlist
There's now a 1.10 release of YAML::Syck. For the changes since then
see: http://cpansearch.perl.org/src/AVAR/YAML-Syck-1.10/Changes
-- System Information:
Debian Release: squeeze/sid
APT prefers testing
APT policy: (900, 'testing'
Package: reniced
Version: 1.17-1
Severity: wishlist
This is hardcoded in the reniced source code:
my $debug = 0;
The debug output is really helpful to find if your rules are taking
effect. I use this hacky wrapper to do this:
perl -0777 -pe 's[debug = \K0][1]' < /usr/bin/reniced | sudo
Package: reniced
Version: 1.17-1
Severity: wishlist
The reniced program is really limited by the hardcoded use of a `comm'
argument to ps(1). It should optionally allow for using `cmd'. Or any
other ps option with a command-line argument.
The use case is making the attached reniced.conf work. I'm
Package: tmpreaper
Version: 1.6.13
Severity: normal
The Sun JVM creates /tmp/hsperfdata_foo where foo is a username
directories. This is a performance feature of the JVM.
See: http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=5012932
-- System Information:
Debian Release: squeeze/sid
APT pr
Package: tmpreaper
Version: 1.6.13
Severity: normal
Just setting TMPREAPER_TIME= in /etc/tmpreaper.conf does not work
because the /etc/cron.daily/tmpreaper script looks at TMPTIME in
/etc/default/rcS first, and if it's set to 0 (which is the default)
uses a 7d TMPREAPER_TIME.
To override that I
On Mon, May 24, 2010 at 05:11, tony mancill wrote:
> I'm not sure what your minimal test case is meant to exhibit. The warnings
> you're seeing from log4j are normal warnings you see when there isn't a
> log4j.properties file. That is, log4j is simply letting us know that a
> component attempted
Package: libyaml-syck-perl
Version: 1.07-1
Severity: wishlist
A new upstream release of YAML::Syck is now available:
http://search.cpan.org/~avar/YAML-Syck-1.08/lib/YAML/Syck.pm
These are the changes since the last release:
http://github.com/avar/YAML-Syck/compare/1.07...1.08
The most notable o
Package: tmpreaper
Version: 1.6.13
Severity: normal
Emacs uses $TMPDIR/emacs$UID/server for its socket. tmpreaper will
delete it if there's a long-running Emacs instance.
Please add it to the default ignore list.
-- System Information:
Debian Release: squeeze/sid
APT prefers testing
APT poli
Package: libpoe-component-irc-perl
Severity: important
Tags: patch
IRC bots which do not take care of removing carriage returns and line
feeds from parameters they send to the IRC component are vulnerable to
this security hole. For example, passing an argument of "foo bar\rQUIT"
to the 'privmsg
severity 579752 serious
kthx
Setting severity to serious as the package is in violation of §10.7.3
of Debian policy, as documented above. Repeated for reference:
"Configuration file handling must conform to the following
behavior: local changes must be preserved during a package upgrade
lways added to /etc/hosts from now on".
This is an example of a change made by netbase.postinst:
diff --git a/hosts b/hosts
index 704eb6b..2e6b55b 100644
--- a/hosts
+++ b/hosts
@@ -1,2 +1,8 @@
127.0.0.1 localhost
109.74.193.250 v.nix.is v
Package: osmosis
Version: 0.34+ds1-1
Severity: minor
After upgrading osmosis I get output like this from cron jobs using it:
log4j:WARN No appenders could be found for logger
(org.java.plugin.ObjectFactory).
log4j:WARN Please initialize the log4j system properly.
log4j:WARN No a
Package: qpsmtpd
Version: 0.84-1
Severity: minor
I keep the files I modify from the Debian defaults in /etc in version
control, for most packages this works fine but qpsmtpd is a thorn in
my eye in this regard. Every time it upgrades it puts an autogenareted
string at the top of its config files:
On Sat, Apr 10, 2010 at 18:40, Noèl Köthe wrote:
> so its OK to close the bug because its not reproducible?
Yes. Not much that can be done about this now.
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.de
On Fri, Apr 9, 2010 at 22:49, Noèl Köthe wrote:
> tags 573324 + moreinfo
> thanks
>
> Hello ?,
>
> Am Mittwoch, den 10.03.2010, 16:01 +0000 schrieb v.nix.is:
>
>> Execute these commands:
>>
>> $ lftp http://www.asstr.org/files/Authors/
>&
Package: lftp
Version: 4.0.5-1
Severity: important
Execute these commands:
$ lftp http://www.asstr.org/files/Authors/
cd ok, cwd=/files/Authors
lftp www.asstr.org:/files/Authors> mirror .
On my system this makes lftp spiral nito 99% cpu use within minutes of
starting the download.
Package: emacs23-nox
Version: 23.1+1-5
Severity: normal
When emacs23-common-non-dfsg isn't install emacs should report a more
useful error than "Info file emacs does not exist".
This could be done by installing a stub info page when
emacs23-common-non-dfsg isn't installed explaining why Debian d
Package: mysql-server-5.1
Version: 5.1.41-3
Severity: minor
The default Debian security policy for MySQL is to set `bind-address =
127.0.0.1` for mysqld. However the examples in
/usr/share/doc/mysql-server-5.1/examples do not include this line.
Thus if a server administrator experiments with opt
Package: php5-cgi
Version: 5.2.12.dfsg.1-2
Severity: minor
Now the README.Debian documentaion for php5-cgi says you should put
this in /etc/apache2/conf.d/:
Action application/x-httpd-php /cgi-bin/php5
However that will simply yield errors from apache about /cgi-bin/php5
being missing. In
Package: libpg-java
Version: 8.4-701-1
Severity: important
When I try to connect to a valid database with a valid user/password
on localhost I get an error like this:
org.openstreetmap.osmosis.core.OsmosisRuntimeException: Unable to
establish a database connection.
at
Package: munin
Version: 1.4.3-2
Severity: important
Tags: patch
I set up munin-cgi-graph as documented on the munin wiki
(http://munin.projects.linpro.no/wiki/CgiHowto) but for certain graphs
I kept getting this error in my apache error log:
[error] [client 127.0.0.1] Premature end of script
debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20100214095948.13775.80538.report...@v.nix.is
Package: srg
Version: 1.3.6-1
Severity: normal
Running srg like this makes it segfault:
/usr/bin/srg -C /etc/srg/srg.conf -m 31 -f 2010-02-11 -t 2010-02-11 -o
/var/cache/srg_reports/daily /var/log/squid/access.log
Here's the GDB output:
(gdb) run -C /etc/srg/srg.conf -m 31 -f 2010-02-11 -
Package: srg
Version: 1.3.6-1
Severity: minor
I think the cron job for srg should mkdir the $REPORTBASE if it
doesn't exist:
$ sudo rm -rf /var/www/srg_reports/
$ sudo /etc/cron.daily/srg
srg: Error creating base output directory: /var/www/srg_reports/daily
$ sudo mkdir /var/
Package: squidview
Version: 0.73-1
Severity: minor
When I run squidview I get this:
v ~ (master) $ stat /var/log/squid/access.log
stat: cannot stat `/var/log/squid/access.log': Permission denied
v ~ (master) $ squidview
Making .squidview directory and links
The squid
Package: munin-plugins-extra
Version: 1.4.3-2
Severity: normal
The mailman plugin at /usr/share/munin/plugins/mailman uses an
undefined variable at this line:
$statefile = "$MUNIN_PLUGSTATE/munin-mailman-log.state";
I.e. it hasn't done C or
something similar previously.
This causes it to t
Package: cron
Version: 3.0pl1-106
Severity: important
Given the following to jobs in a user's crontab:
*/1 * * * * echo `date +%Y-%m-%d`
*/1 * * * * echo $(date +%Y-%m-%d)
Cron will execute this:
Jan 19 21:30:01 v /USR/SBIN/CRON[8641]: (avar) CMD (echo $(date +)
Jan 19 21:30:01
Package: dash
Version: 0.5.5.1-3
Severity: important
dash doesn't understand the ** exponent operator. This caused some
breakage when moving from bash to dash as suggested by the package
manager as several (non-debian) cronjobs used the feature on their
argument list.
$ bash -c 'echo $((2**1
Package: munin
Version: 1.4.2-1
Severity: normal
The apache example in /etc/munin/apache.conf uses .
That doesn't work, it should use .
-- System Information:
Debian Release: squeeze/sid
APT prefers testing
APT policy: (650, 'testing'), (500, 'unstable')
Architecture: amd64 (x86_64)
Kerne
72 matches
Mail list logo
