Hello, this is my first post about an issue here. I’m on Manjaro XFCE since 2021 and I have to say that everything went mostly smoothly until now. I’ll try to explain it… Some weeks ago, on my main machine, I updated Manjaro after quite some time (I know, I know…). There were some issues, and I thought that the long time without updates created some mess. So, I reinstalled everything from scratch (twice!) and wiped away all my /home folders to do a clean reinstall, but the issues persist (which, by the way, had never happened before in this machine). They are quite weird, and I haven’t been able to find a soultion by searching in the forum. The version in which I’m seeing right now these issues is 24.0.3 (installed with the .iso from the official webpage)
These are some things that I observe right now :
After reinstalling and configuring the OS to my preferences, after various reboots, then it happens that the desktop misses its icons (not the panel, though). I can’t even right-click in the background to change the desktop settings, theme, etc. Since I use a solid color background, all I see now is a black screen with the panel and the Manjaro whisker menu. I tried resetting the xfce-minimal-settings but nothing changes with any reboot. If I open the desktop settings from the whisker menu, I can’t change anything…
Thunar acts very strangely; I have to double-click several times (sometimes killing some Thunar stuck process in the task manager) to make it run. It also, randomly, doesn’t run it’s root mode with the right-click option. Besides, when I try to run it both from the panel icon or from the whisker menu, it sometimes shows a sign saying that there was an input/output error. After that, it can happen that it runs fine after retrying (or not), Everything is quite unpredictable. Hence, I installed Nemo alongside Thunar, and I use both depending on the occasion.
Notepadqq (and the Mousepad) tend to shutdown soon after I open any simple text file. This is really strange. Sublime-text doesn’t crash, so I’ll stay with it for now.
If I see or remember other strange issues like these, I’ll post them here. Needless to say, these things never happened during these years with Manjaro. I can’t speak about a couple of other machines I have, since I have just reinstalled the OS in them (these things seem to happen, as I said, after a number of reboots). I’m clueless about the cause of all of this; I tweak the machine for realtime audio work and do some other modifications, but I’ve been doing it for years and nothing weird ever happened. I wonder if it could be hardware-related (since the machine in which I have these issues is an I7 from 2013).
I even tried to switch the kernel from 6.9 to 6.6, but the problem persist. These things don’t reach (for now) the dealbreaker status (and I’d like not to move to another distro since Manjaro is the most convenient for me), but they are really weird and it feels rather uncomfortable to deal with all of this. I hope there can be a solution, if it’s not related to my hardware.
While I’m no Linux wizard, I’ve learnt a thing or two during these Manjaro years, and I’m certainly not afraid of the terminal. Any insight about this will be deeply appreciated.
Thank you for editing your post and changing the presentation of your terminal output from Blockquote to Preformatted text; it is much easier to read now. As a general practice, it is also helpful to include the initial command with the output, so that everyone knows what they are looking at. For example
While this may or may not be related, I note that kernel 6.7 reached EOL some 3 months (or more) ago; which means it’s no longer receiving any updates, security or otherwise.
If you do actually use the 6.7 rt it might be beneficial to switch back to 6.6 rt, and remove 6.7 rt, until the next realtime kernel is released.
I note that your mainboard has a BIOS update available;
albeit rather old, it’s more recent than the BIOS you’re currently using;
Perhaps it will make some difference:
P8H77-V LE BIOS 1307
Version 1307
4.19 MB
2014/05/09
Improve memory compatibility.
Did you remove any orphans at some point? Blindly removing them can unintentionally remove some important packages, so if you did, please post which orphans you removed.
Thanks so much for all the replies! Beginning with these ideas, I’ll check out the disk, you’re right. I’ve have had some debacle with a SSD in the past, so it’s a possibility. Regarding swap, I have 6 GBs of it (without hibernation) in a machine with 32 GBs of RAM, so I think that I’m fine with that…
Thanks for the suggestions, I’ll take note. These are my first posts here, and I was rather nervous yesterday with these issues.
Definitely! Anyway, I could even switch back to 6.9, since the issues appear the same in any context. For realtime audio work, I understand that the “normal” kernel is just fine with the threadirqs flag in grub…
Yes, you’re right. I never updated the BIOS. I’ll do it now, since these strange things are happening…
Good question… I can only think of two things :
I removed cmake dependencies while installing a couple of things (I don’t think this can be related, but I mention it anyway).
I removed all things pipewire, to use just Pulseaudio, ALSA and JACK. Maybe I erased something important?
Apart from these things, I didn’t really have too much time to break stuff, since I had just reinstalled the OS. I’ll do another reinstall today (since being Sunday, I have a couple of hours). I’ll try not to remove anything, to see if it does the same during these days. And I’ll check the disk and load the BIOS update, to see if there’s any improvement…
I’ll report back after all of this. Thanks again for the ideas!
I hear you, definitely! A couple of questions, regarding 6.6 : If I switch from 6.9 to 6.6, do I have to mark it as non-upgradable? That is, until I see that another LTS kernel is released? Or LTS kernels stay automatically without upgrades until another one of the same type appears?
And, besides, just out of curiosity : is the RT version of 6.6 supposed to be as stable as the “normal” one?
Regarding a possible disk failure : I did an extended test with GSmartControl and it showed no errors on the disk. I don’t know if there is another deeper way to check it…
They are different packages.
And you can have as many kernels installed as you like - with the ability to choose them at boot. linux66 will continue to receive updates for as long as it is supported.
Which, according to kernel.org, will be until the end of 2026.
Sorry for the delay in the answer… Yes, that could be the case here. What do you recommend about it?
I have to click several times before it starts. Sometimes, the thunar process is seen on the task manager, but after a long while, nothing happens anyway. When I kill once or twice that process, then it starts. Sometimes, it even gives a warning sign saying that there is an input/output error. All of this doesn’t seem to happen before the desktop is supposedly disabled.
I know… I have only just used it to place some vst/vst3 plugins in a folder. But I see your point.
Maybe… The same applies with it as in the Thunar case. It seemingly starts to misbehave when the desktop appears as if it was disabled.
Should be something in Desktop Settings. (Settings Manager > Desktop)
I believe you can check/uncheck items in the Icons tab, but there should also be an option for “Icon Type” which if set to “None” would disable the interactive Desktop.
Please review settings in this area.
Though I just re-read that in the original situation you could not affect any changes to these settings.
Is this still the case?
Have you tried launching thunar from the terminal to see if it spits out anything useful?
Similarly you may go checking in the journal for any clues.
In the mean time I can leave some general ‘make sure you are up to date’ commands;
sudo pacman-mirrors --continent
sudo pacman -Syu
Oh, and so we can see what you did remove we can check the pacman log:
Well, none of the settings there seem to change anything. No icons appears, no matter how many options you check/uncheck… Regarding the original situation, (which also happens right now), I can´t right-click in the background of the desktop to access the settings.
Well, the first time I try to run it, it shows this :
(thunar:1294): thunarx-python-WARNING **: 08:36:41.831: g_module_open libpython failed: /usr/lib/lib.so.1.0: cannot open shared object file: No such file or directory
ImportError: could not import gobject (error was: ImportError('/usr/lib/python3.12/site-packages/gi/_gi.cpython-312-x86_64-linux-gnu.so: undefined symbol: PyUnicode_FromFormat'))
(thunar:1294): thunarx-python-WARNING **: 08:36:41.853: pygobject initialization failed
(thunar:1294): thunarx-python-WARNING **: 08:36:41.853: thunarx_python_init_python failed
Traceback (most recent call last):
File "/usr/share/thunarx-python/extensions/insync-thunar-plugin.py", line 15, in <module>
import socket
File "/usr/lib/python3.12/socket.py", line 52, in <module>
import _socket
ImportError: /usr/lib/python3.12/lib-dynload/_socket.cpython-312-x86_64-linux-gnu.so: undefined symbol: PyUnicode_FromFormat
Segmentation fault (core dumped)
And on a third try :
(thunar:1822): thunarx-python-WARNING **: 08:38:57.210: g_module_open libpython failed: /usr/lib/lib.so.1.0: cannot open shared object file: No such file or directory
ImportError: could not import gobject (error was: ImportError('/usr/lib/python3.12/site-packages/gi/_gi.cpython-312-x86_64-linux-gnu.so: undefined symbol: PyUnicode_FromFormat'))
(thunar:1822): thunarx-python-WARNING **: 08:38:57.233: pygobject initialization failed
(thunar:1822): thunarx-python-WARNING **: 08:38:57.233: thunarx_python_init_python failed
Traceback (most recent call last):
File "/usr/share/thunarx-python/extensions/insync-thunar-plugin.py", line 15, in <module>
import socket
File "/usr/lib/python3.12/socket.py", line 52, in <module>
import _socket
ImportError: /usr/lib/python3.12/lib-dynload/_socket.cpython-312-x86_64-linux-gnu.so: undefined symbol: PyUnicode_FromFormat
On this third try, the process is stuck and I have to kill in the Task Manager.
Then, on the fourth try it runs, while giving these lines in the terminal :
(thunar:1872): thunarx-python-WARNING **: 08:40:58.308: g_module_open libpython failed: /usr/lib/lib.so.1.0: cannot open shared object file: No such file or directory
ImportError: could not import gobject (error was: ImportError('/usr/lib/python3.12/site-packages/gi/_gi.cpython-312-x86_64-linux-gnu.so: undefined symbol: PyUnicode_FromFormat'))
(thunar:1872): thunarx-python-WARNING **: 08:40:58.332: pygobject initialization failed
(thunar:1872): thunarx-python-WARNING **: 08:40:58.332: thunarx_python_init_python failed
Traceback (most recent call last):
File "/usr/share/thunarx-python/extensions/insync-thunar-plugin.py", line 15, in <module>
import socket
File "/usr/lib/python3.12/socket.py", line 52, in <module>
import _socket
ImportError: /usr/lib/python3.12/lib-dynload/_socket.cpython-312-x86_64-linux-gnu.so: undefined symbol: PyUnicode_FromFormat
(thunar:1872): thunar-WARNING **: 08:40:58.558: Name 'org.freedesktop.FileManager1' lost on the message dbus.
I`ll show the output of the journalctl command. In the beginning it says “Linux 6.9.5-1”, but I’m on 6.6.40-1 right now (at some point, it just changed to 6.9 again without my intervention, and I had to revert to 6.6 at boot. I thought that once you chose a kernel, it just kept loading it on every boot…
I finally discovered what was causing the desktop debacle in my system : an aur package related to Insync, the application I use to sync my Google Drive folder across computers. The package is insync-thunar, and it simply broke the desktop due to some python stuff mess. I removed it with yay -Rsn and voilá! The XFCE desktop returned as usual…