Cinnamon broken after 2025-06-15 update (fallback mode, doesn’t start anymore)

I’m on EOS Cinnamon, freshly updated. After the last reboot and login, I see a “Cinnamon is is in fallback mode”. Restarting Cinnamon leads to the same, only when I answer “No” to the restart, some desktop icons appear, and the Welcome app. No panel, no title bars on windows. I could get here by using some info button in the Welcom app only, which started Firefox, although its menus and tabs don’t work when clicking with the mouse.

I tried to get logs using the Welcome app. Since I can’t change windows, I’ll come back with the link, just a mo.

Logs, as far as possible: https://0x0.st/86AJ.txt

journalctl -b 0: https://0x0.st/86Ay.txt

During the day, I did install & try a different theme via the Cinnamon theme setting, and removed it again. (Never mess with the default desktop & theming, aarrgh!). The system ran fine the whole day, but when I opened the lid again this evening, it was stuck, showing a time ~1 hour ago, and no login. Since Ctrl+F2 terminal didn’t let me log in, I had to hard poweroff. After that, problem on the next boot.

Cannot make a screenshot, but changing windows with mouse or Alt+Tab is impossible, and a small window fragment keeps jumping around on the screen.

Thanks for your help!

hmmm this is what i get when i try to get the logs. Looks like I’ll have to try later.

Basically journal says:

process xxx (AppRun) of user 1000 dumped core

then later

process 1166 (cinnamon) of user 1000 terminated abnormally with signal 6/ABRT, processing...

followed by

process 1166 (cinnamon) of user 1000 dumped core

I’m wondering if this was caused by the theme I tried, or by one the updates I did today.

Maybe I could reinstall only cinnamon from the terminal without it messing up all my settings?

was this themed installed to the system or for the user? I mean you could try to remove the cinnamon folders in .config and see if it lets you boot into it then.

Already tried to

cd ~/.config
mv cinnamon cinnamon.sav
mv cinnaom-session cinnamon-session.sav
reboot

No difference.

I installed that theme via the Cinnamon Theme app, by selecting from the available themes, downloading it, and activating it.
No good, so I changed back to standard theme, and deleted it again, also through the theme setting.

Currently can’t get there, since no panel, no Ctr+Alt+T terminal, and that doesn’t work from the Ctrl+F2 terminal of course.

1 Like

did you check .local? to see if there is anything in there? I don’t have Cinnamon installed at the moment so can’t check myself.

There is a ~/.local/share/cinnamon with 4 subfolders, but all empty.

then maybe it goes back to the update :thinking:
can you post the pacman log for the last update maybe someone will have an idea that is actually on cinnamon.

Today’s pacman.log: https://0x0.st/86mZ.log

To follow-up: It must have been one of today’s updates that broke Cinnamon! And Timeshift saved my ass!

Fortunately, my daily Timeshift happened just before I tried to install the theme (2025-06-15 15:00). I restored to that, and all works.

To verify, I re-applied the updates using eos-update --aur (189 packages), and boom! Cinnamon broken!

Restored again, and now don’t dare upgrading anymore…

I include the pacman.log of the update I made after restoring the Timeshift snapshot. (This upgrade apparently broke Cinnamon.)

https://0x0.st/86mj.txt

Hopefully someone of the gurus here can reproduce and/or see what the problem is!

I did an update this morning with Core stuff so I had to reboot. No problem.
There was some other stuff mid-day for the second update.
Just did an update now and it’s a bunch of core/glibc stuff.

I’m afraid to reboot! Which on bit you I wonder?

The process AppRun is from an AppImage. Do you have one running at startup?

Not that I knew of. I do have appimagelauncher but it is set to not automatically do anything. My ~/Applications folder is empty.

Nevertheless, I know that some AUR packages unpack from or use (?) AppImages, so here is my pacman -Qm. Note that all this worked before the update, for several weeks, without problems.

$ pacman -Qm
anydesk-bin 7.0.0-1
appimagelauncher 2.2.0-9
backintime 1.5.5-1
backintime-cli 1.5.5-1
cinnamon-sounds 1.8.9-1
circle-flags 2.7.0-1
cp210x-overclock-dkms 0.1-1
espanso-x11-bin 2.2.3-2
geany-plugin-preview 0.1.2-1
gnome-online-accounts-gtk 3.50.6-2
goldendict-ng 25.5.0-1
hypnotix 4.9-1
ibus-autostart 1.3-1
idjc 0.9.9-1
jellyfin-media-player 1.12.0-3
libeb 4.4.3-12
libpamac-aur 11.7.3-2
libshout-idjc 2.4.6.r1-1
lightdm-settings 2.0.7-1
marktext-bin 0.17.1-2
open-adventure 1.16-3
pamac-aur 11.7.3-1
piper-tts-bin 2023.11.14-1
piper-voices-common 1.0.0-5
piper-voices-de-de 1.0.0-1
piper-voices-en-gb 1.0.0-2
piper-voices-en-us 1.0.0-3
pulsar-bin 1.128.0-1
python-cinemagoer 2023.05.01-1
quickemu 4.9.7-1
simplescreenrecorder 0.4.4-3
thorium-reader-bin 3.1.0-2
timeshift-autosnap 0.9-1
webcamize 2.0.1-1
xwinwrap-git r20.539fc47-1

A quick find for AppImages did also not find any:

$ sudo find / -xdev -iname "*.AppImage"
$

I’m afraid to update… 188 packages waiting…

If I’d know that… Guess some of the gurus here need to step in and help.

I am not sure how you configured your timeshift service. It doesn’t back up $HOME contents unless specifically enabled. If that’s the case, you can skip changes in your $HOME.

It appears that cinnamon crashed (core dumped, sig 6). Stick with the version that is working and wait for further update from pacman. If you really want, you could try to debug the core files. In systemd, they’are located in,

/var/lib/systemd/coredump

You could also list them from cmd line,

$coredumpctl list

Problem is of course I’d like to continue updating and installing software on this machine. So it would be beneficial to find out what exactly caused the crash. I could then maybe uninstall the program causing it.

Otherwise, how would I know when it’s safe to update again?

Timeshift I’ve set to auto-snapshot once a day and before updates (using the Pacman hook), and never to save anything from the home folders. I use Back in Time for that.

coredumpctl shows:

$ coredumpctl list --since "3 days ago"
TIME                         PID  UID  GID SIG     COREFILE EXE                                      SIZE
Fri 2025-06-13 21:02:28 CEST 942 1000 1000 SIGTRAP present  /tmp/.mount_espansA2ygLB/usr/bin/espanso   1M
Sat 2025-06-14 12:40:36 CEST 937 1000 1000 SIGTRAP present  /tmp/.mount_espansU1o4ws/usr/bin/espanso   1M
Sat 2025-06-14 12:57:50 CEST 930 1000 1000 SIGTRAP present  /tmp/.mount_espanspP1eEX/usr/bin/espanso   1M
Mon 2025-06-16 00:13:47 CEST 967 1000 1000 SIGTRAP present  /tmp/.mount_espansCBzQfi/usr/bin/espanso   1M

So just some Espanso messages. It sometimes shows “updated keyboard configuration”, maybe that’s when it crashes. Usually happens at max. once a day so I didn’t care much, since it continued working.

You have “ulimit -c” set properly?

What is ulimit? Tbh, never heard of it or touched it.

It controls various system/usr limits in the system. Try,

$ulimit -a

$ ulimit -a
real-time non-blocking time  (microseconds, -R) unlimited
core file size              (blocks, -c) unlimited
data seg size               (kbytes, -d) unlimited
scheduling priority                 (-e) 0
file size                   (blocks, -f) unlimited
pending signals                     (-i) 63187
max locked memory           (kbytes, -l) 8192
max memory size             (kbytes, -m) unlimited
open files                          (-n) 1024
pipe size                (512 bytes, -p) 8
POSIX message queues         (bytes, -q) 819200
real-time priority                  (-r) 0
stack size                  (kbytes, -s) 8192
cpu time                   (seconds, -t) unlimited
max user processes                  (-u) 63187
virtual memory              (kbytes, -v) unlimited
file locks                          (-x) unlimited

EDIT: If you were looking for the crash dump, that might not be there anymore, since I had to restore to a Timeshift snapshot before the update and crash happened, to get this system working again. But I’ve saved the pacman.log (see above).