Command "tdecmshell mouse" gives "Could not find module 'kde-mouse'".
And kcontrol crashes when trying to configure the mouse.
Reinstallation of the whole TDE-bunch (preliminary stable) brought no
relief. But in normal use the rodent does what it is supposed to do.
Who?
Regards.
Anyone else seeing issues with kaddressbook and distribution lists?
Three of our users, to my knowledge; others might not have noticed yet,
have had distribution lists corrupted in their std.vcf files for no
apparent reason.
For example here's a diff showing the breakage:
*** backup.8/home/steven/.trinity/share/apps/tdeabc/std.vcf Wed Jan 23 16:26:48 2019
--- backup.9/home/steven/.trinity/share/apps/tdeabc/std.vcf Tue Jan 22 16:27:12 2019
***************
*** 121,131 ****
REV:2010-10-13T10:07:43Z
UID:2v3n80lysi
VERSION:3.0
! X-KADDRESSBOOK-DistributionList:;KF0nfbXjcI;W65A8xToAL;TkroRixGpz;sfPshZ13G
! R;LOuCWSQ97o;nv7cFe6Qn3;EwrV6PmgS7;zQitSMdyc5;uhE8AALaot;lda8YjPh7O;iFvalmA
! jur;7blsWNQhCI;jQSplbzEBx;Xr9sO4FR6K;FetVPQzl88;tvfhPIFsnm;y5ouAVz2Xg;dGYAO
! IYZYH;6YVRsssvB3;ECp97esiDY;G8Dg2NNwtf;NEErjVskoA;UFXZpbD4y6;h2LTb23i8o;6Ej
! 2CTR7fR
END:VCARD
BEGIN:VCARD
--- 121,131 ----
REV:2010-10-13T10:07:43Z
UID:2v3n80lysi
VERSION:3.0
! X-KADDRESSBOOK-DistributionList:;KF0nfbXjcI\,;W65A8xToAL\,;TkroRixGpz\,;sfP
! shZ13GR\,;LOuCWSQ97o\,;nv7cFe6Qn3\,;EwrV6PmgS7\,;zQitSMdyc5\,;uhE8AALaot\,;
! lda8YjPh7O\,;iFvalmAjur\,;7blsWNQhCI\,;jQSplbzEBx\,;Xr9sO4FR6K\,;FetVPQzl88
! \,;tvfhPIFsnm\,;y5ouAVz2Xg\,;dGYAOIYZYH\,;6YVRsssvB3\,;ECp97esiDY\,;G8Dg2NN
! wtf\,;NEErjVskoA\,;UFXZpbD4y6\,;h2LTb23i8o\,;6Ej2CTR7fR\,
END:VCARD
BEGIN:VCARD
***************
The '\,' between each entry in the list went AWOL sometime between these
two backup copies which makes kaddressbook come up with the UID as the
target email address, causing a massive fail.
For fun and amusement, I hand edited the \, between the entries back
into the list and it all works again.
Qt: 3.5.0
TDE: R14.0.6 [DEVELOPMENT]
KAddressBook: 3.5.13
Debian stretch.
--
Regards,
Russell
--------------------------------------------------------------------
| Russell Brown | MAIL: russell(a)lls.com PHONE: 01780 471800 |
| Lady Lodge Systems | WWW Work: http://www.lls.com |
| Peterborough, England | WWW Play: http://www.ruffle.me.uk |
--------------------------------------------------------------------
Alas I only get to play on the weekends, and this full write up isn’t going to
be done for at least a month, so figured I’d post the pre-bits to all y’all,
as most of you are programmer enough to read dev notes and understand them
well enough...
As Kate said, “Zim, I like it so far.” It does output almost clean HTML code,
but it’s definitely not a complete solution for writing a local web page and
being able to ‘rip the body guts out’ and paste it into a CMS Body box for
publishing. That said, it’s the best I’ve ever seen, so I’m using it.
## Random Notes
Zim is really cool. The set of Plugins available is extensive. Do go though
the whole list with the help file open in a separate window. Make sure to
configure the ones that can be config’ed!
Templates: HTML export templates exist that can be customized
Crashes a lot, auto saves every 10 seconds by default.
Copy/Paste of links doesn’t always work.
Link creator is lacking and does stuff you didn’t ask it to.
Spellchecker right click menu is in Spanish(?). Spell checker isn’t great,
but works well enough.
Has an insanely cool Table of Contents plugin, but it does not HTML export the
ToC! Q@$%#$%!!!
# # #
Work flow I’ve found that seems to work.
- Write your HTML page in Zim
- Export HTML to see final product in browser(s) [1]
- Once finished with HTML use Notetab Pro, or your equivalent post processor,
to run though your ToC generation*, cleanup, and fix-it scripts.
- Use Quanta Plus, Kate, whatever, to get CSS to your liking.
* If anyone asks I’ll post what I end up with that creates a ToC from H
attributes.
Best,
Michael
[1] Script to help with Exporting to HTML.
michael@local [~/bin]# cat zimexport.sh
#!/bin/bash
#
# Zim export to HTML the default 'Notes' Notebook.
# Change paths as needed to customize.
#
# Add to Zim Custom tools for one click icon in the Zim toolbar:
# Zim -> Tools -> Custom tools
# Click 'green plus' icon
# Name: Export Notes
# Description: Export to HTML the default Notes
# Command: ~/bin/zimexport.sh
# Icon:
# /usr/share/icons/Humanity/actions/48/document-export.svg
# /usr/share/icons/oxygen/48x48/actions/document-export.png
# Show in toolbar: Checked.
#
# Notes:
# 'Command:' needs the full path.
# HTML templates are in /usr/share/zim/templates/html/
# Icons are in: /usr/share/icons
#
# One off script, possibly has bugs, use at your own risk.
#
# You will need {nothing}:
# yum install {nothing}
# apt-get install {nothing}
# (whatever you use to install it)
#
# Script=zimexport.sh
# Author=Michael
# Email=staff_scripts at the commercial domain inet-design
# Website=http://www.inet-design.com/
# License=GPL
# Script repository=http://inet-design.com/tags/scripts
# # # #
zimNotebook="~/Notebooks/Notes"
HTMLOutput="~/Notebooks/html"
if [ ! -d "$HTMLOutput" ] ; then
mkdir -p "$HTMLOutput"
fi
zim --export --overwrite --output="$HTMLOutput" --format=html --template="Default.html" "$zimNotebook"
Sorry; here's another R14.0.6 issue we're seeing (Yes, I can/will report
this as a bug but thought I'd throw it out here first just to see if
anyone's got any ideas).
Logging in (via XDMCP) multiple times as the same user/uid is causing
kdesktop to do weird refresh stuff then get stuck in a loop (100% CPU
usage). I have seen this once-in-a-blue-moon under 14.0.5/Debian Jessie
but 14.0.6/Debian Stretch seems to make it fail every time.
Yes, I know it's bad practise to login multiple times as the same user
but people will do these things despite being told not to :-(
So, before I get into the nitty gritty about creating a failing case, is
there anything obvious I should set/unset to stop this? Is it a known
problem?
How to recreate.
Hard: have a bunch of X-Terminals.
Easy: Use vnc :-) (see below for setup).
User logs in (XDMCP/tdm) on first xterm/vnc session.
Same user logs in again from second xterm/vnc session.
The cursor on each desktop flickers with a busy? sort of pointer and
the desktop refreshes sparodically. NB. Logging out of one session
seems to set a desktop lock on the second.
kdesktop for each session will probably have a high CPU use.
Same user logs in a third time and you end up with a kdesktop running
100% CPU and the desktop(s) get unresponsive (can happen with just two
sessions but three makes it fail quickly).
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
10285 utest 20 0 258196 23212 19128 S 100.0 0.0 45:48.54 kdesktop
21584 lls 20 0 44608 4636 2804 R 26.3 0.0 0:00.07 top
1 lls 20 0 205688 7984 5312 S 0.0 0.0 3:50.32 systemd
Looking at the 100% kdesktop process, it has a thread (10287) running in
a constant poll timeout loop (strace streams the lines below):
poll([{fd=13, events=POLLIN}, {fd=15, events=POLLIN}], 2, 0) = 0 (Timeout)
poll([{fd=13, events=POLLIN}, {fd=15, events=POLLIN}], 2, 0) = 0 (Timeout)
poll([{fd=13, events=POLLIN}, {fd=15, events=POLLIN}], 2, 0) = 0 (Timeout)
poll([{fd=13, events=POLLIN}, {fd=15, events=POLLIN}], 2, 0) = 0 (Timeout)
Looking at /proc/10287/fd I see:
lr-x------ 1 utest tract 64 Feb 6 11:17 0 -> pipe:[11722107]
l-wx------ 1 utest tract 64 Feb 6 11:17 1 -> /home/utest/.xsession-errors
lr-x------ 1 utest tract 64 Feb 6 11:17 10 -> pipe:[11687858]
l-wx------ 1 utest tract 64 Feb 6 11:17 11 -> pipe:[11687858]
lrwx------ 1 utest tract 64 Feb 6 11:17 12 -> socket:[11667305]
lr-x------ 1 utest tract 64 Feb 6 11:17 13 -> pipe:[11688700]
l-wx------ 1 utest tract 64 Feb 6 11:17 14 -> pipe:[11688700]
lrwx------ 1 utest tract 64 Feb 6 11:17 15 -> anon_inode:[eventfd]
lr-x------ 1 utest tract 64 Feb 6 11:17 16 -> /var/tmp/tdecache-utest/tdesycoca
l-wx------ 1 utest tract 64 Feb 6 11:17 2 -> /home/utest/.xsession-errors
lrwx------ 1 utest tract 64 Feb 6 11:17 3 -> socket:[11746354]
lrwx------ 1 utest tract 64 Feb 6 11:17 4 -> socket:[11687855]
lrwx------ 1 utest tract 64 Feb 6 11:17 5 -> socket:[11687856]
lr-x------ 1 utest tract 64 Feb 6 11:17 6 -> pipe:[11687857]
l-wx------ 1 utest tract 64 Feb 6 11:17 7 -> pipe:[11687857]
lrwx------ 1 utest tract 64 Feb 6 11:17 8 -> anon_inode:[eventfd]
lrwx------ 1 utest tract 64 Feb 6 11:17 9 -> anon_inode:[eventfd]
The pipe on fd 13 for the thread seems to be connected to it's parent:
# (find /proc -type l | xargs ls -l | grep 11688700) 2>/dev/null
lr-x------ 1 utest tract 64 Feb 6 11:57 /proc/10285/fd/12 -> pipe:[11688700]
l-wx------ 1 utest tract 64 Feb 6 11:57 /proc/10285/fd/13 -> pipe:[11688700]
lr-x------ 1 utest tract 64 Feb 6 11:58 /proc/10285/task/10285/fd/12 -> pipe:[11688700]
l-wx------ 1 utest tract 64 Feb 6 11:58 /proc/10285/task/10285/fd/13 -> pipe:[11688700]
lr-x------ 1 utest tract 64 Feb 6 11:58 /proc/10285/task/10287/fd/12 -> pipe:[11688700]
l-wx------ 1 utest tract 64 Feb 6 11:58 /proc/10285/task/10287/fd/13 -> pipe:[11688700]
#
-------
VNC setup.
I'm using Debian Stretch. The easiest way to setup a VNC test case is
to add this to /etc/services on the TDE server system (192.168.1.1 in my
case).
# VNC Servers
vnc0 5900/tcp
vnc1 5901/tcp
vnc2 5902/tcp
#
Then create /etc/xinetd.d/vnc0 like so:
service vnc0
{
user = nobody
disable = no
protocol = tcp
socket_type = stream
flags = KEEPALIVE
wait = no
group = tract
instances = 1
log_type = FILE /var/log/vnc
log_on_failure += HOST USERID ATTEMPT
server = /usr/bin/Xvnc
server_args = -inetd -NeverShared -query localhost -once -depth 16 -geometry 1024x768 -SecurityTypes none
}
Copy /etc/xinetd.d/vnc0 to /etc/xinetd.d/vnc1 and change "service vnc0"
to "service vnc1". Repeat for vnc2. Restart xinetd.
From another system on the network fire up a vnc viewer for the first session:
vncviewer 192.168.1.1:5900
Login as test user.
and for the second:
vncviewer 192.168.1.1:5901
Login as same user in on that one.
You'll now see the desktops in each vncviewer window refreshing themselves
and a kdesktop process for each one using significant CPU:
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
26694 utest 20 0 275004 33772 27016 R 49.7 0.0 1:35.19 kdesktop
27099 utest 20 0 275004 33544 26784 S 47.0 0.0 1:38.19 kdesktop
Fire up a third:
vncviewer 192.168.1.1:5902
Login as same user on that.
You'll now get a kdesktop consuming 100% CPU:
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
10805 utest 20 0 278088 37156 27332 S 100.3 0.0 4:15.84 kdesktop
26694 utest 20 0 275004 33772 27016 S 43.3 0.0 3:37.91 kdesktop
27099 utest 20 0 275004 33544 26784 S 42.0 0.0 3:36.62 kdesktop
On this third session, kdesktop becomes unresponsive (not surprising as
it's stuck in a loop!).
NB. I've also tried this using lightdm instead of tdm and exactly the
same thing happens.
--
Regards,
Russell
--------------------------------------------------------------------
| Russell Brown | MAIL: russell(a)lls.com PHONE: 01780 471800 |
| Lady Lodge Systems | WWW Work: http://www.lls.com |
| Peterborough, England | WWW Play: http://www.ruffle.me.uk |
--------------------------------------------------------------------
Hello,
Since I upgraded Trinity,
all the TDE menus are in french,
excepted the TDE menu to turn off my computer,
end current session, reboot, hibernate..., who is in english.
I installed again the package "tde-i18n-fr-trinity",
but no change.
Is it a bug... ?
Thanks,
André
On Sunday 03 February 2019 19.21:34 phiebie(a)drei.at wrote:
> Debian Buster, KDE 3.5.10. aptitude
> As more and more software-updates clash with their dependencies against
> what KDE needs or provides, I wanted to install trinity, preliminary
> stable, alongside KDE. KDE should stay operational till I had
> transferred my settings and programs aso to trinity.
> No go!
> Tdebase was fetched and seemingly installed, but when I restarted the
> computer and wanted to start KDE, there was only a black screen with
> the mouse-cursor and whatever I did with keyboard or mouse, nothing
> happened. Cold reboot to clear memory, same thing.
> Okay, let's then have a look at trinity. A blue screen with the logo
> appeared and after a few seconds something like "no ..... available
> check your installation"and I only could close that window via a warm
> reboot. KDE gave me the black screen again and also TDE said
> "check installation".
> Installed TDE again, same results as before. Half an afternoon had
> passed with no desktop still available.
> Glad, that I had backupped my system before the experiment and could
> restore my working KDE.
> So there's also a clash between TDE and KDE, incredible!
> Where should I look for the culprit for this disaster?
> Kind regards.
OK, so...
a) You say Buster. Where did you install TDE from? What version? As far as I
know, for Buster you need the premilinary builds.
b) You say KDE 3.5.10. I seem to remember that install instructions said to
purge any KDE 3.5 elements befor install. So yes, KDE 3.5 and TDE clash. It's
known. I guess that's because thes share files with the same name, but
different contents.
If you want Trinity, you should better install on a system with no KDE, and
running a testing TDE on a testing Debian, I'm not really surprised you
encounter some problems.
These are caused by your environment, not by Trinity itself.
Thierry
https://www.maketecheasier.com/take-screenshot-of-login-screen-linux/ is the closest I have found
to instructions to use ImageMagick to get one, but tdm seems to be different from lightdm and gdm
such that I can't figure out what to change. I tried logging in and grepping XAUTHORITY from set,
but it apparently cannot be found there.
I want a fullscreen screenshot that has the integrated default theme's menu expanded to show the
selection of session types. Can anyone here help?
--
Evolution as taught in public schools is religion, not science.
Team OS/2 ** Reg. Linux User #211409 ** a11y rocks!
Felix Miata *** http://fm.no-ip.com/