On 12/07/2015 01:30 PM, Yves-Alexis Perez wrote:
> On sam., 2015-12-05 at 17:59 +0100, Daniel Reichelt wrote:
>> Up until now I thought, lightdm failed to correctly register a
>> user-session, when really it failed to register one at all...
>
> It might very well be beca
> Yes it's a wrapper, but it's a needed one, it'll especially filter stuff in
> the environment. It would help to show the logind session before and after
> restart. I seem to remember that when you restart like that, the lightdm
> process will inherit the console session or something like that,
> Can you try to restart lightdm using: service lightdm restart and report back?
Same faulty behaviour, since `service lightdm restart` only acts as a
wrapper for `/etc/init.d/lightdm restart`, see attached console.log.
Daniel
# service lightdm stop; sleep 10; bash -x service lightdm start
Package: lightdm
Version: 1.10.3-3
Severity: important
Hi,
this system is Jessie with lightdm, sysvinit-core and systemd-shim installed,
running sysvinit-core's /sbin/init.
After booting, lightdm works fine. After doing a `/etc/init.d/lightdm restart`,
whenever I log on with the username user,
The patch suggested by me has been accepted upstream. [1], [2]
Maybe this could be included in the next point release...
Thanks
Daniel
[1]
https://webdav.seas.upenn.edu/viewvc/unison/trunk/src/uigtk2.ml?r1=551=550=551
[2]
Package: unison-gtk
Version: 2.40.102-2
Severity: normal
Tags: patch
Hi,
shortcut keys defined in profiles for the GTK UI ceased working since 2.40
(Jessie). Until 2.32.x it was working fine.
Finally I found some time to dig into this and found, that now the profiles are
scanned only after the
Package: xfwm4
Version: 4.10.1-3
Severity: normal
Hi,
running xfwm4 and vncviewer, since the Jessie release I'm no longer able to do
multi-selections using CTRL+click or SHIFT+click in an application within the
vncviewer window.
Expected behaviour: multiple items selected (contiguous in case
correct link to git commit:
http://git.xfce.org/xfce/xfwm4/commit/?id=d30cd2fb8a2303ed93dd4d0e5d73bfb337541f59
PS: dpkg --force-depends --remove roundcube-core; apt-get install
roundcube-core
resolved the situation for me.
Package: roundcube
Version: 1.1.2+dfsg.1-1
Severity: important
Hi,
during the upgrade, the package tinymce got removed, complaining about the
directory /usr/share/tinymce/[...] not being empty.
Turns out, when roundcube-core got extracted, there still was the symplink
Package: base-files
Version: 9.2
Severity: normal
Tags: patch
In testing /usr/bin/mesg is provided by util-linux instead of sysvinit-utils.
Its changed behaviour with respect to exit codes needs to be accounted for in
/usr/share/base-files/dot.profile.
When displaying of messages is set to no
this is already fixed in git
Not quite, see [1]
Cheers
Daniel
[1]
http://live.debian.net/gitweb/?p=live-build.git;a=commitdiff;h=2a400dc63ec7305c80dd072984a14839aed74501
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact
Hi,
Commit a1d5ddea5b fixed this.
http://anonscm.debian.org/cgit/d-i/debootstrap.git/commit/?id=a1d5ddea5b7db00548173cee8236aaad61cf5961
cu
Daniel
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact
Package: exim4-config
Version: 4.80-7+deb7u1
Severity: grave
Tags: security
Justification: user security hole
Hi folks,
suppose you have set up an exim4 which provides virtual mailing, managing
domains/accounts in a DB, say mysql.
Just adding mysql queries and DB-*authentication data* to the
Hi *,
How about now? :)
thanks for your patience :)
I can confirm Didier's steps to work under jessie and sid, however it
takes ages for lpstat, lpadmin, cupsenable et al to timeout (the overal
postinst run took several minutes). I didn't dig very deep into this
this time, but at least the
Martin-Eric,
thanks for bumping this. I'll try to get on this between the holidays.
Daniel
On 12/20/2014 11:56 PM, Martin-Éric Racine wrote:
2014-10-27 18:16 GMT+02:00 Daniel Reichelt deb...@nachtgeist.net:
Have you had time to reproduce this by following Didier's steps?
Not so far
Package: php5-common
Version: 5.4.35-0+deb7u1
Severity: serious
Justification: fails to build from source (but built successfully in the past)
Hi,
on a wheezy system I just updated php5-common (and other stuff) via
security.d.o and since then php5-common's cron script fails. Invoked from the
Sorry for the misleading justification - must have misused reportbug...
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
I seem to be unable to reproduce it with 1.0.60~bpo70+1 but I can with
1.0.48+deb7u1; I didn't spot anything obvious from a very quick look,
but I'll try to track this down a bit later (probably after the release
sprint).
This bug actually is a duplicate of #702861 which was fixed in 1.0.51
Have you had time to reproduce this by following Didier's steps?
Not so far. For personal reasons I'll be busy for at least the next 2 to
3 weeks :/
Daniel
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact
Hi guys,
Thanks Didier. I'll try to work in this on the weekend (not sure
though). One more question: which debian version were you running on the
chroot host while you were trying to reproduce this?
Cheers,
Daniel
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a
I cannot confirm this: CUPS 1.7 is installable and working inside a
chroot, as long as the 631 port is available. Is this a bug with CUPS
1.4 only? Has it been reproduced with later versions ?
Thanks, Didier.
Which steps exactly did you take in trying to reproduce this?
Thanks
Daniel
--
Also, Daniel, did you try to bootstrap wheezy or sid? And are you running
systemd?
Like I saide: of wheezy/jessie only wheezy is affected. The commands I
ran are
# debootstrap wheezy /tmp/wheezy http://ftp.de.debian.org/debian
and
# debootstrap jessie /tmp/jessie
And are you running systemd?
Oh I forgot to mention: the host used to build the chroots is plain
wheezy, not running systemd:
# dpkg -l | grep systemd
ii libsystemd-login0:amd64 44-11+deb7u4 amd64 systemd login utility library
# aptitude why libsystemd-login0
i rhythmbox Depends dbus
i A
Hi Bastian,
is there anything I can do in the meantime to mitigate this, apart from
running `service mountdevsubfs.sh start` manually after e.g.
xen-create-image (which invokes deboostrap)?
Daniel
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of
Package: debootstrap
Version: 1.0.48+deb7u1
Severity: critical
Justification: breaks unrelated software
Hi,
after running debootstrap the host's /run/shm tmpfs mount gets unmounted. Since
the mount point /run/shm itself is set to root:root/750, this breaks any other
package which relies on being
Hi there,
I managed to track down what's going wrong with compiling
jitsi/2.4.4.997-1 on wheezy, although I didn't solve it in the sense
that it is now compilable on plain Wheezy but on a Wheezy-based system,
pulling in some build-depends from Jessie. Here are my steps:
1) minimal Wheezy
I just found out this only happens if /bin/sh points to dash. No
problems at all with bash.
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Hi,
somehow I messed up the PATH on my system and I was testing the wrong
script all along.
The patch from my previous mail does NOT work. Sry for the confusion.
Daniel
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact
Hi,
the attached patch fixes the issue for me, with -j/--jobs taking
precedence over an exported envvar CONCURRENCY_LEVEL.
Cheers
Daniel
--- /usr/bin/make-kpkg.orig 2012-08-15 17:57:29.0 +0200
+++ /usr/bin/make-kpkg 2013-11-01 22:12:25.421575757 +0100
@@ -714,10 +714,6 @@
of .
Package: libnss-ldap
Version: 264-2.5
Severity: normal
Hi,
libnss-ldap fails to install in a chroot (like with e.g. live-build)
Excerpt from the live-bild logfile:
-8--
Setting up libnss-ldap:amd64 (264-2.5) ...^M
invoke-rc.d: unknown initscript, /etc/init.d/nscd not found.^M
dpkg:
Just now I noticed the also alternating change of the server signature
(apache 2.2.x vs 2.4.x) - maybe a remnant process from before the upgrade?
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Package: git-el
Version: 1:1.7.10.4-1+wheezy1
Severity: normal
Hi,
I just tried to do a simple aptitude install git-all with no git or emacs
packages installed (no unpurged config data present as well) but it failed
during postinst of git-el. Below is the console output. Please let me know if
Package: debian-cd
Severity: normal
Hi *
using [1] I end up downloading stuff from [2] which ATM obviously should be
[3]. Moreover, I suggest the creation of an additional stable symlink.
Thanks
Daniel
[1] http://cdimage.debian.org/cdimage/unofficial/non-free/firmware/testing/
[2]
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Hi Moritz
Please test/report, whether the packages located at
http://people.debian.org/~jmm/ fix the problem for you.
Could you please publish the source package as well?
And is this going to go into squeeze-updates eventually?
Cheers
Daniel
Package: devscripts
Version: 2.10.69+squeeze2
Severity: normal
Hi *,
when lintian exits with a non-zero code, that fact is totally ignored by
debuild which essentially renders debuild useless for automated building and
testing with lintian.
I suggest one of these modifications:
* at least pass
Package: live-config
Version: 2.0.15-1
Severity: important
Tags: squeeze sid
Hi Live-Team,
I spotted another incompatibility between a .deb and how the Live system is
handling its behaviour (similar to live vs. dwww).
Usually I boot my live-systems so that LIVE_TIMEZONE=Europe/Berlin gets set
Package: smbldap-tools
Version: 0.9.5-1
Severity: important
Tags: patch squeeze sid
Hi,
I'm using slapd via ldaps://. smbldap-tools work fine as far as operation from
an account with access to smbldap_bind.conf is concerned. However users can't
change their own passwords since the bind operation
Hi everybody,
1. Linux, by its design, is a multi-task system, and there's nothing in
the Policy that disallows spawning background tasks from maintainer
scripts, so I don't understand why this bug was marked serious.
Actually, I have to agree on that one. (Although I'm glad it got this
much
Package: live-build
Version: 2.0.12-2
Severity: normal
Tags: squeeze
Hi live team,
I'm building quite large live systems which I employ in daily production use,
so they are BIG (around 4GB). Recently I added the package dwww, which causes
the build to fail occasionally. I was able to track down
Daniel: to cercumvent that in live-build, you can add a local preseed file.
gr8 ptr, thx so much! ...and for the quick reply!
cu
D
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
just wanted to tell 2.5.1-3 works fine here and thanks for the quick action.
Daniel
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Package: squid3
Version: 3.1.6-1.2
Severity: normal
Tags: patch
Hi,
I have configured a custom cachedir in squid.conf and noticed it doesn't get
created properly - the init script tried to create the default path
/var/spool/squid3
Changing /etc/init.d/squid3:67 from
$DAEMON -z
to
Hi Guys,
I did some strace'ing of the troubling lpadmin invocation both on the
host rootfs and within the chroot and after the connect() call I noticed
some numeric named files in /var/run/cups/certs get accessed:
Host:
8-
connect(4, {sa_family=AF_INET, sin_port=htons(631),
Package: dpkg-www
Version: 2.54+nmu1
Severity: normal
Hi *,
I just stumbled over an inconsitency in handling casing. When I accidentally
searched for mC instead of mc, I was presented with this search result:
-8--
Package: mC
Status: install ok installed
Priority: optional
Section:
Package: squid
Version: 2.7.STABLE9-2.1
Severity: normal
Tags: patch
Hi *,
I just experienced problems with my DNS which actually were caused by
/etc/resolvconf/update-libc.d/squid
What happened? The reload of the squid config caused a core-dump which got
saved to CWD which was
Package: w3c-markup-validator
Version: 0.7.4-5.2
Severity: normal
Hi all,
suppose someone has an apache2 config block like
Directory /
order allow,deny
deny from all
/Directory
Now the symlinked config of this package doesn't override that. Could you
please add something like
Directory
Package: manpages-dev
Version: 3.25-1
Severity: normal
Hi,
a)
when assigning
int i = pclose(pipeFD);
OR assigning
long i = pclose(pipeFD);
I get a compiler warning like this:
warning: assignment makes integer from pointer without a cast
b)
when assigning
int i = (int) pclose(pipeFD);
I get a
Package: gitweb
Version: 1:1.7.2.3-2
Severity: normal
Hi,
I just noticed that integrating gitweb via reverse proxy is impossible when
path_info is enabled in gitweb.conf. The base href sent on gitweb.cgi:3427
contains the internal URL called by the reverse proxy mechanism, not the
original
I'm hijacking this bug to request a better manual page. :)
Jonathan
Fine with me, thx. Cya upstream...
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Hi Goswin,
just tested with 20101117, works fine now. Thanks!
Daniel
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Package: ia32-libs
Version: 20090808
Severity: serious
Tags: patch
Justification: Appendix G
Hi,
on my system I have installed a 3rd party package which adds a diversion like
this:
$ dpkg-divert --list /usr/bin/ldd
diversion of /usr/bin/ldd to /usr/bin/ldd.real by myutils-lddwrapper
When
Hi Julien,
If it only happens with 3rd party packages I don't think this counts as
serious.
Short answer: if the 3rd party package is perfectly fine, that's no
justification, ia32-libs postinstall script still is broken.
Long answer: It may very well be disputable if my bug report is to be
Package: munin
Version: 1.4.5-2
Severity: normal
Tags: upstream patch
Hi,
I just stumbled over the following nasty inconvenience:
- cgi-graphing enabled
- www-data is member of group munin, so for cgi graphing to work the cache
dirs need to be group writeable
- added new node to be monitored
Good point! In that case 775 would be a more reasonable default for cgi
and non-cgi.
Well, in that case my patch was in vain. Might have just talked to you
guys first :)
Since Holger asked me to write s.th. for the README, are we talking
about an upstream default or Debian-specific?
--
To
I think 775 by default would make the default installation (non-cgi
mode) insecure.
Depends. Since the htmldir is still owned by group munin, only a member
of that group could go rampage on those files. IMHO the rrd files are
far more valueable than the dynamically generated files within the
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
+1, thank you!
Cheers,
Daniel
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/
iQIcBAEBAgAGBQJMfCf1AAoJEIWTgWPaKFdzMDsP/jgIDEKGJkx/78Fz6oCP6gwD
Hi again,
I totally missed the fact that with the previous patch on multigraph-pages
within every device-specific graph *all* devices were drawn.
The attached patch (incorporates the previous one) fixes that:
- first of all, munin-cgi-graph prints the device-specific image url instead of
the
Package: munin
Version: 1.4.4-1
Severity: normal
Tags: patch
Hi,
when watchin a multigraph page usign cgi-graphing, my apache2 errorlog gets
clobbed with lines like this:
---snip--
[Sun May 23 17:45:15 2010] [error] [client 10.0.50.112] Use of uninitialized
value $scale in
Hi,
I just could reproduce the described behaviour like this:
when having a sources.list like this
**
deb http://ftp.de.debian.org/debian experimental main contrib non-free
deb http://ftp.de.debian.org/debian stable main contrib non-free
deb
Package: munin-node
Version: 1.4.4-1
Severity: normal
Tags: patch
Hi,
within mktempfile() in amavis plugin the executed variable $cmd looks like this:
/bin/mktemp -p /tmp/ munin-amavis.XX
where the quotes are part of the string itself. I inserted a set -x in the
script and the
Package: munin-node
Version: 1.4.4-1
Severity: normal
Tags: patch
Hi,
currently cron reports errors on munin-cron like this:
Malformed configuration line $domain;$hostname:amavis.graph_vlabel. at
/usr/share/perl5/Munin/Master/Utils.pm line 337.
I didn't look
Package: munin-node
Version: 1.4.3-2
Severity: wishlist
Tags: patch
Hi!
I just noticed the new total graph in the df_abs diagrams and find it mostly
usefull but in some cases also irritating. The attached patch makes df_abs
mind a new env var $nototal and disables the total graph when it's set
Package: munin-node
Version: 1.4.3-2
Severity: normal
Tags: patch
Hi,
in /usr/share/munin/plugins/df_abs there's an error within the sed call: it
fails to
correctly prepend each element of a space-separated list of file systems to be
ignored
in the graphs to be prepended by '-x '. The
by Daniel Reichelt deb...@itamservices.de
##
## All lines beginning with `## DP:' are a description of the patch.
## DP: No description.
@DPATCH@
diff -urNad rdiff-backup-1.2.8~/rdiff_backup/selection.py rdiff-backup-1.2.8/rdiff_backup/selection.py
--- rdiff-backup-1.2.8~/rdiff_backup/selection.py
Package: live-initramfs
Version: 1.157.4-1
Severity: important
Hi,
since udev stops shipping /lib/udev/vol_id initramfs-tools needs updating.
According to [1], all calls to
vol_id -t
should be changed to
/sbin/blkid -o udev -p ...
Probably the initramfs-hook to copy that new
Hi,
since 2.2.11-01-1+b1 apache fails to read .htaccess files which aren't
world-readable. For the sake of testing I reverted my vhost setup to
AssignUserId www-data www-data and tried it with a .htaccess file in the
DocumentRoot owned by www-data:www-data set to 640. An attempt to retrieve
Hi,
My point is, if people disable it, they should hopefully know what they
are doing.
Apparently they don’t, I received yet another bug today which is
probably caused by that.
I'm sorry to have to open crossfire :-)
This is my situation:
Hi,
after upgrading gconf to 2.24.0-7 the login on the machine itself via gdm or
plain startx works flawlessly. However I can't start my vncserver anymore, it
just hangs after the desktop background is shown. My xstartup looks like this:
$ cat ~/.vnc/xstartup
#!/bin/sh
x-session-manager
The
20K config vs. 1.6M image size? Seems a bit an odd comparison to me. On the
other hand, the uncompressed .config in /boot takes ~80K which is four times as
much (ok, it also lies on disk and not in RAM :-)
My point is, having the config available during runtime might come in handy
when handling
Sorry, something went wrong with my previous post.
I just wanted to ask the kernel team to enable
# grep IKCONFIG config-2.6.24-1-amd64
CONFIG_IKCONFIG=y
CONFIG_IKCONFIG_PROC=y
for all x86 kernel packages like it is the case for amd64.
Thanks,
Daniel
--
To UNSUBSCRIBE, email to [EMAIL
triggered by the
mount targets of the iso-images. As you can see in the patch attached,
for the conditional awk output to work as desired, another pair of hyphen
enclosing the awk match calls is required.
HTH,
Daniel Reichelt
*** tmp/update-initramfs.patch
--- update-initramfs2007-11-28 19:42
the HTTPS envvar and act accordingly.
HTH,
Daniel Reichelt
-- debconf information excluded
--- /usr/lib/cgi-bin/dwww-org 2007-04-01 20:58:22.0 +0200
+++ /usr/lib/cgi-bin/dwww 2007-12-15 00:32:54.0 +0100
@@ -76,9 +76,11 @@
elsif (($type eq ) or ($location eq
101 - 173 of 173 matches
Mail list logo
