On Sat Jun 13 08:41, Brice Goglin wrote:
> On Sat, Aug 11, 2007 at 01:24:40AM +0100, Richard Thrippleton wrote:
> > Package: xserver-xorg-video-i810
> > Version: 2:2.1.0-2
> > Severity: important
> >
> > I've consistently found that after a suspend to RAM,
I can reproduce this issue. Furthermore, it only happens for ssh clients that
have port forwarding (LocalForward) enabled.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Package: xterm
Version: 232-1
Severity: important
xterm, run with whatever arguments, exits with a segfault before bringing up a
window. I've graded this as important rather than anything higher because it
may work on other X servers, especially based on the gdb output.
The relevant part of the
Package: pwsafe
Version: 0.2.0-3
Severity: normal
The manpage states that -E will "Force echo'ing of entry to stdout. This is
selected by default if $DISPLAY is not set."
However, in this version of pwsafe, "pwsafe -up " will respond with
"Going to copy login and password to X selection" when $DIS
On Sat Aug 11 10:42, Brice Goglin wrote:
> Richard Thrippleton wrote:
> > Attached. This was run on a fresh X server before a suspend. Do you want a
> > second output from after a suspend?
> >
>
> No need to send the whole output again. Just save /var/log/Xorg.0.log
On Sat Aug 11 02:38, Julien Cristau wrote:
> Please run '/usr/share/bug/xserver-xorg-core/script 3>&1' and send the
> output to this bug.
Attached. This was run on a fresh X server before a suspend. Do you want a
second output from after a suspend?
Richard
Contents of /var/lib/x11/X.roster:
xserve
Package: xserver-xorg-video-i810
Version: 2:2.1.0-2
Severity: important
I've consistently found that after a suspend to RAM, many GL applications fail
to render correctly, or sometimes completely lock up the system. GL behaves
fine up until after the first suspend. A workaround is to restart the X
Package: gxine
Version: 0.5.8-3
Severity: important
On startup, and for playing any kind of media (DVD or mpg/divx file), I get a
dialogue box entitled "Error from the xine engine" containing "The xine engine
failed to start" and "No demuxer found - stream format not recognised", and
playback does
On Wed Jul 04 21:33, Vincent Fourmond wrote:
>
> Hello,
>
> I've recently taken over maintenance of pmount, and I'm going through
> bug reports. I am not convinced that the problem you have is really
> pmount's fault : when I have a stale smb mount and that I ls (--color)
> on the parent dire
I've had the 'persist' option fail for me similarly, but without the strange
'tcsetattr' bug. This is with PPPoE (for an ADSL connection), and has occurred
in the past when the PPP connection has been terminated at the ISP end (during
faults, or sometimes load balancing).
Jul 4 01:10:01 bluesk
Package: sylpheed-claws-gtk2
Version: 2.6.0-1.1
Severity: grave
Justification: renders package unusable
After initially encountering the error, I removed all .sylpheed* files from my
home directory, and started sylpheed-claws-gtk2. After configuring it with
name, address, servers etc., an "Error"
Package: postfix
Version: 2.4.0-2
Severity: wishlist
The sendmail interface under exim supports restrictions on who can set an
arbitrary sender with "-f", enforcing that some local users can only send out
mail with their authentic address. It would be useful if postfix had such a
feature to preven
>
> MboxStyle=qmail means that ~/Maildir is prefered to /var/mail/$USER when
> `imget -s local'. imget always uses MH style for message destination.
>
> This isn't a bug, but wishlist.
I thought that initially, but then found evidence to the contrary - look in
LocalMbox.pm , searching will show
Package: im
Version: 1:148-6
Severity: normal
In both /etc/im/SiteConfig and in ~/.im/Config, "MBoxStyle=qmail" is set. When
using a command of the form imget -s nntp:[EMAIL PROTECTED] -d =news.group
, the contents are downloaded and a mailbox at the correct path is created,
filled with the expect
Package: pmount
Version: 0.9.13-1+b1
Severity: important
Attempting to pmount a device hangs for a very long time while I happened to
have some stale network mounts open on other mountpoints. strace (as root, so
the suid didn't get clobbered) indicated hanging at;
lstat64("/mnt/bulk",
(where /
As the person who found and has thoroughly tested this bug, I can confirm
firsthand that this isn't just a case of apache being vulnerable with "-F"! I
specifically mentioned using the init script in the original report over a
month ago, not "-F". That is, the circumstances required to exploit this
Yes, Apache closes fds 0,1 and 2 (well, reopens them to /dev/null) , but
retains the controlling tty - this can be accessed via opening /dev/tty . The
controlling tty is a process property that is separate from the fds it has
open, and allows certain ioctl powers on any fd matching that. A 'ps' lis
To fix this problem, revert the 033_-F_NO_SETSID patch. It only introduced
minor functionality, and if anyone really cares about having it, they can try
submitting a new patch. One that doesn't introduce a root exploit into the
package, ideally.
Richard
--
To UNSUBSCRIBE, email to [EMAIL PROTEC
I noticed that suexec doesn't abdicate the controlling terminal either, and I'd
not be surprised to find out that non-suexec CGI invocation didn't do this
either. The result is that I've just been able to escalate from local user -->
root by writing a hostile CGI script - this is no longer just a c
If I have understood this correctly, preserving the controlling tty like this
allows an escalation from www-data to root. If, for example, I run
"/etc/init.d/apache start" from a root shell which I don't close soon after,
a resulting apache process running as www-data will share a controlling tty
w
Package: ivman
Version: 0.6.12-5
Severity: normal
As far as I can tell, ivman can only match and exec on conditions by name, not
by value. This is a problem on my toshiba system, where hal emits conditions
for hotkeys like this;
computer condition ButtonPressed = brightness-down
computer conditio
Package: icewm
Version: 1.2.28-4
Severity: normal
When running an application inside icewm that hangs (i.e. stops updating,
presumably some underlying crash) while one of its menu or drop-down list
widgets is in use and in focus, icewm itself refuses to acknowledge any user
input. No toolbar butto
I have more examples, it's definitely not just a one-off. Again, did an apt-get
update, then attempted installing (this time with an apt-get upgrade)
immediately afterwards.
Failed to fetch
http://ftp.uk.debian.org/debian/pool/main/g/gtk+2.0/libgtk2.0-common_2.8.20-3_all.deb
MD5Sum mismatch
F
Package: apt
Version: 0.6.45
Severity: normal
For a long time now, apt has been failing to install packages due to an md5sum
mistmatch. There is no particular pattern to the packages that fail, it seems
random. An example;
Failed to fetch
http://ftp.uk.debian.org/debian/pool/main/p/pingus/pingu
On Tue Oct 10 14:59, Peter Eisentraut wrote:
> Richard Thrippleton wrote:
> > As of a recent package upgrade, ntpdate no longer includes the script
> > in init.d to update the time.
>
> That is intentional. Read NEWS.Debian.
OK. The implication is that ntp should be the l
Package: ntpdate
Version: 1:4.2.2+dfsg.2-3
Severity: important
As of a recent package upgrade, ntpdate no longer includes the script in init.d
to update the time. I've had to modify scripts in order to keep my use ntpdate
working.
-- System Information:
Debian Release: testing/unstable
APT pre
xserver-xorg-video-i810 at version 1:1.5.1.0-2
libgl1-mesa-dri at version 6.4.2-1
Exactly the same message about I830DRIRec not matching size. I'm assuming the
library is a newer version than when the bug was reported, as the original
reported didn't say, and it's been over 3 months since it was a
Package: pmount
Version: 0.9.13-1
Severity: normal
I'm having problems mounting an iso9660 formatted DVD-R. pmount can cope with
mounting it, but pmount-hal cannot;
[EMAIL PROTECTED]:~$ pmount-hal
'/org/freedesktop/Hal/devices/volume_label_SGA_2_2'
mount: block device /dev/hdc is write-protect
Package: gtkam
Version: 0.1.12-2.2
Severity: wishlist
Currently, gtkam will not delete directories off of the camera until all of the
files/directories within it have been deleted manually. Recursive delete would
ease the tedium!
-- System Information:
Debian Release: testing/unstable
APT pref
Package: gtkam
Version: 0.1.12-2.2
Severity: normal
With a single camera plugged in over USB, gtkam auto-detects it on startup and
allows normal use. When I select 'Rescan' from the 'Camera' menu, the following
error dialogue box comes up;
"Could not get file list for folder '/'
An error occur
I have this bug on my system, saw the bug that this was blocking on, and
noticed that I didn't even have xfs installed - is the X Font Server a
dependency for playing Flash? Or is the problem deeper than that?
Richard
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe"
Package: xjump
Version: 2.7.5-2
Severity: wishlist
Playing xjump for even a short period of time is extremely harsh on the eyes,
seemingly due to the very fine brick pattern in the background of the main
gameplay. Vision remains blurry for a while after play, and I've had somebody
with better eye
Package: cdrecord
Version: 4:2.01+01a03-5
Followup-For: Bug #377145
The patch seems to apply cleanly on the latest source in Debian, and doesn't
appear to break functionality (I burnt a CD without problems). It does however
make using cdrecord less annoying.
Can somebody confirm that the "do not
Bug #374345 seems fairly similar, which I've added detail to. Appropriate to
reassign/merge ?
Richard
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
I can confirm this (I'm not too happy having pointless setuid applications
around either - instead I just have rw access to the cdrom device). Here's the
sample output;
[EMAIL PROTECTED]:~$ cdrecord /mnt/bulk/isos/ubuntu-5.10-live-i386.iso
cdrecord: No write mode specified.
cdrecord: Asuming
Package: beep-media-player
Version: 0.9.7.1+cvs20050803-1.1
Severity: normal
With WM decorations turned on (that is, it spawns normal windows rather than
bypassing the window manager), when resizing the playlist window using WM
controls, the window border resizes but not the playlist content wit
My locale was en_GB.UTF-8 ... switching to POSIX fixed the bug (though the
fonts looked subtly different). I'm assuming it's some problem with the global
gtkrc for en_GB or utf-8. I may investigate further if I can find the time.
Richard
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subj
Package: xdm
Version: 1:1.0.1-6
Severity: normal
This could either be a bug in /etc/X11/xdm/Xresources , or xdm itself. The
resources named xlogin*Font and similar purport to set the fonts used in the
login widget, but have no such effect. Changing the resource name to login*Font
works.
I think xd
I can confirm that there's more than one person suffering from this - I've
encountered this in xmms and other gtk1.2 apps. See bug #365678 .
Richard
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Package: xmms
Version: 1.2.10+cvs20060429-1
Severity: important
While the skinned GUI is fine, any gtk-style menus and dialogue boxes are
entirely devoid of text. In the preferences dialogue, the tabs at the top are
shrunken, indicating that the text just isn't there rather than a font or
colours
Package: gxine
Version: 0.5.4-0.1
Severity: normal
First bug - I can set video contrast/brightness from the Video/Settings menu,
and it persists for the session, but upon quitting and reloading, the defaults
are restored. The .gxine/config file is being written (according to the
modified time), b
On Mon, Mar 20, 2006 at 01:20:50AM +0100, Marco d'Itri wrote:
> reassign 357858 linux-2.6
> thanks
>
> Unless it's #350235/#355441 again (and it should not be, since that
> happens even before you can try using the device) then it's some similar
> kernel-related problem.
> Anyway, udev just crea
Package: udev
Version: 0.087-2
Severity: normal
Using a compact flash card in the card slot in my laptop, which is presented as
pcmcia-ide, while udev creates the device node correctly, it apparently removes
and recreates the device on any attempt to mount it. This makes mounting
impossible witho
Package: postfix
Version: 2.2.8-9
Severity: normal
When run as _any_ user, 'postqueue' allows them to see the contents of the queue
with -p , and also to flush it with -f. The former is what I'd consider a user
privacy violation, the latter seems to be putting administrative functions in
the hand
Package: sox
Version: 12.17.9-1
Severity: normal
When attempting to play or record to an alsa device, sox fails with an ioctl
error. In my case;
[EMAIL PROTECTED]:~$ sox msg.WAV -t alsa /dev/snd/pcmC0D0p
sox: ioctl operation failed 22
[EMAIL PROTECTED]:~$ sox -t alsa /dev/snd/pcmC0D0p reco
Package: kphone
Followup-For: Bug #288598
While I can use inband signalling to get most DTMFs recognised by Asterisk,
whenever I attempt to issue a '#' with kphone, asterisk recognises it as '*'
according to the verbose logs.
-- System Information:
Debian Release: testing/unstable
APT prefers t
Package: kphone
Severity: normal
Whenever music or any other sound is playing from another application, an
incoming call will hang the entire application, as will making an outgoing
call. Hanging in this cases refers to the windows failing to redraw. I'm using
alsa for all sound on the system, and
Package: dnsmasq
Version: 2.22-2
Severity: wishlist
When you specify only one interface (or several) in dnsmasq.conf, it still
binds to all interfaces to listen for requests. It would be preferable if it
only listened on the specified interfaces, leaving other interfaces free to
have different dn
Package: msmtp
Version: 1.4.1-1
Severity: normal
Using the version in woody, one was able to use the option --from="" to start
an SMTP transaction with an empty return path (for example, when passing on a
bounce from mailer-daemon). This no longer works in the current version and
behaves as if th
Turns out this problem was caused simply by running out of memory space -
strace indicated a flood of calls returning ENOMEM just before a segfault.
Making more memory available fixed the problem for good.
I suggest that this bug be downgraded to something just indicating that firefox
(and probably
Package: mozilla-firefox
Version: 1.0.4-2
Severity: important
Upon attempting to visit any URL, by either hitting enter, or clicking on a
bookmark/home icon, no response ... a second repeat always attempts to go to
the URL but always stays on 'Looking up ' before timing out as
usual with non-exist
51 matches
Mail list logo
