Can someone verify some gtk-qt-engine package weirdness?
* Ensure another DE is installed.
* Remove any traces of TDE from /etc/ld.so.conf and run ldconfig.
* Disable related TDE /etc/profile.d scripts.
* Boot into the other DE.
* Launch a terminal window.
* Ensure no traces of TDE are in the environment variables, $PATH, etc.
* Type a GTK GUI command to launch.
* Verify TDE is trying to intercept the command to control the GTK theme.
In my case I was trying to launch meld from the command line. I
repeatedly encountered a TDE crash dialog:
TDECrash: Application 'python-gtk-tqt-application' crashing.
The message is derived from the package qt_qt_wrapper.cpp.
This happened to me in both KDE and Xfce.
Removing the gtk-qt-engine package resolves the problem.
Being buggy or stale code is one thing, but why is TDE and the
gtk-qt-engine package even functional when using a different DE?
DA
What is the preferred way to configure environment variables for root?
The starttde script establishes environment variables for the user's
session, but those settings do not apply to root. Running su - results
in the loss of these environment variables.
The same happens when using the konsolesu.desktop and
konquerorsu.desktop files. Environment variables such as $PATH do not
include $TDEDIR/bin. So although the su - session is in a full TDE
session, no TDE commands can be used unless typing full paths.
When using konquerorsu.desktop, Konqueror launches okay but odd things
happen. The Settings dialogs are empty. The Help button in those empty
dialogs fails to find the Konqueror handbook. The handbook functions
fine from the main Help menu.
When launching Konq this way there seems to be something hard-coded to
$HOME/.trinity. I have $TDEHOME configured to $HOME/.tde rather than
$HOME/.trinity and $TDEROOTHOME configured to /root/.tde rather than
/root/.trinity. These variables are exported globally on login through
/etc/profile.d. Yet launching Konq as root results in these environment
variables being ignored and $HOME/.trinity is created with default
configurations.
I do not mind learning I am doing something wrong but I do not know
where to look.
Thanks.
DA
I am trying to create build environments. With some packages I am
receiving the following fatal error:
/usr/bin/ld: cannot find -lpng
I can build the packages in another almost identical system. With
respect to this error, the libpng packages are the same. I don't know
what I should look for or what might be missing.
Thanks.
DA
Hi All,
When selecting not to use any login splash there is nonetheless some
kind of mini startup dialog. The dialog does not appear when using a
splash or possibly is hidden by the splash. Is there a way to disable
this mini startup dialog? None should mean none.
Thanks for your time!
DA
Hi,
is someone using pinentry-tqt?
Is it somehow in the list of available applications?
I was cleaning up and saw the conversion work that now I removed, because
the work was passed to GnuPG.
As I compile TDE myself and do not actually know what is available in the
package managers, I'm asking here.
BR
--
FCD6 3719 0FFB F1BF 38EA 4727 5348 5F1F DCFE BCB0
Every PC I try to remove avahi from wants to disgorge most of TDE, even though all
of them are running systemd-network with static configuration for eth0 and have no
wireless hardware, just like my KDE3 installations that don't require avahi. What
wouldn't work if avahi wasn't installed with TDE? Why does TDE require it?
--
Evolution as taught in public schools is, like religion,
based on faith, not based on science.
Team OS/2 ** Reg. Linux User #211409 ** a11y rocks!
Felix Miata
The following is an extreme example, very old hardware, one PC, tried with two old
NVidia GPUs.
# lspci -nnk | grep -A2 VGA
01:00.0 VGA compatible controller [0300]: NVIDIA Corporation NV11 [GeForce2 MX/MX 400] [10de:0110] (rev a1)
Subsystem: Palit Microsystems Inc. NV11 [GeForce2 MX/MX 400] [1569:0110]
Kernel driver in use: nouveau
# lspci -nnk | grep -A2 VGA
01:00.0 VGA compatible controller [0300]: NVIDIA Corporation NV34 [GeForce FX 5200] [10de:0322] (rev a1)
Kernel driver in use: nouveau
Kernel modules: nouveau
# dpkg-query -l tdm-trinity
Desired=Unknown/Install/Remove/Purge/Hold
| Status=Not/Inst/Conf-files/Unpacked/halF-conf/Half-inst/trig-aWait/Trig-pend
|/ Err?=(none)/Reinst-required (Status,Err: uppercase=bad)
||/ Name Version Architecture Description
+++-==============-===========================-============-=================================
ii tdm-trinity 4:14.0.13-0debian12.0.0+0~a i386 X display manager for TDE
https://paste.debian.net/hidden/1d9bf7ac/ Xorg.0.log from failed first start
https://paste.debian.net/hidden/f1a49e73/ log from successful subsequent start
# systemd-analyze
Startup finished in 28.020s (kernel) + 2min 53.529s (userspace) = 3min 21.549s
graphical.target reached after 2min 53.482s in userspace.
From Xorg.0.logs, it looks like the server works fine, just takes longer than TDM's
default timeout. NAICT, the 120s default timeout isn't really all that long for some
hardware, but it appears TDM may not always try a second time, even though default
tries is 4.
https://bugzilla.opensuse.org/show_bug.cgi?id=1188954 may be related. It's much more
easily reproducible with TDM than with other DMs, and on openSUSE than elsewhere. Age
of hardware or drivers used don't seem to impact whether or not this happens. On an
given PC, it can't be counted on to happen every time. On the current one, after a
switch back to the newer of the two GPUs, it booted normally, though quite slowly.
I've seen this mostly on openSUSE, but today's is obviously from (32bit) Bookworm.
Have others encountered $SUBJECT? What's necessary to determine whether fault is
in TDM or elsewhere?
--
Evolution as taught in public schools is, like religion,
based on faith, not based on science.
Team OS/2 ** Reg. Linux User #211409 ** a11y rocks!
Felix Miata
