users November 2025

users@lists.fedoraproject.org

53 participants
35 discussions

X11 graphics problem
by Robert McBroom 18 Nov '25

18 Nov '25

While xscreensaver is trying to run the following messges keep coming to the terminal--- [GFX1-]: glxtest: ManageChildProcess failed [GFX1-]: glxtest: libEGL initialize failed [GFX1-]: glxtest: X error, error_code=2, request_code=151, minor_code=3 [GFX1-]: RenderCompositorSWGL failed mapping default framebuffer, no dt [GFX1-]: glxtest: ManageChildProcess failed Ideas as to what is missing from the installation would be much appreciated.

2 1

How to disable a globally-enabled systemd user service for the gdm user
by Ben 17 Nov '25

17 Nov '25

Hello, Running on Fedora 43 with latest updates. I created a systemd user service that is globally enabled (ie, sudo systemctl --global enable servicename.service) and configured to launch when any user logs in to the system (hopefully, only for graphical logins): --- [Unit] Description=Registration Reminder After=graphical-session.target PartOf=graphical-session.target After=polkit.service [Service] Type=forking ExecStart=/usr/share/Project/wrapperscript Restart=no [Install] WantedBy=default.target --- (I've modified some of the details since they are internal.) The wrapperscript launches a python gtk dialog with pkexec that requires a user to do some stuff and configures their system. After a reboot, the service always fails to start the first time it is started. Sometimes, the attempt to initialize the pygtk env cores with "Can't create a GtkStyleContext without a display connection." After debugging this, I realized that systemd was trying to start this service for the gdm user. (I guess the display environment that systemd starts for gdm doesn't stick around for too long, which is why sometimes it cores.) I don't want this service to run for the gdm user, and Google suggested this command: sudo systemctl --user -M gdm@ disable servicename.service Checked the systemd manpage, and it seems legit. However, whenever I try and run this command, it always fails: sudo systemctl --user -M gdm@ disable servicename.service Failed to connect to system scope bus via local transport: No such file or directory Failed to disable unit: Transport endpoint is not connected There is no /run/user/42/bus path for gdm like I have for my regular user. Anyone know how to disable a globally enabled systemd user service for gdm? Thanks, Ben

4 4

Installing Signal Desktop on Fedora 43
by Max Pyziur 17 Nov '25

17 Nov '25

Greetings, The flip to F43 has caused my installation of Signal Desktop to stop functioning. I came across these Debian packages on the Signal webpage: https://signal.org/download/linux/ Is there a way of installing these on F43? Many thanks, Max Pyziur pyz(a)brama.com

4 5

what channel is the exit code stored in?
by ToddAndMargo 16 Nov '25

16 Nov '25

Hi All, There is STDOUT, STDERR, and the exit code ($? in bash). What channel is it stored in? STDEXIT? Just curious about the name. -T

4 6

Chrome Will Not Start
by Richard Kimberly Heck 14 Nov '25

14 Nov '25

I have several machines running Fedora 42. One one of them, all of a sudden, Chrome will not start. > google-chrome-stable Created TensorFlow Lite XNNPACK delegate for CPU. [13280:13315:1113/115904.140753:ERROR:google_apis/gcm/engine/registration_request.cc:292] Registration response error message: DEPRECATED_ENDPOINT And then it is frozen, and has to be killed. I updated everything. Same problem. Uninstalled and re-installed. No luck. I created a new profile and even a new system account and tried there: Same problem. Any ideas? Riki

4 3

Fedora 43 vs BlueTooth, etc
by Jonathan Ryshpan 14 Nov '25

14 Nov '25

I have two small Linux Systems, one of which is used to interface with a Davis weather station, the other for general use. Shortly after the release of Fedora 43 both stopped running my current release, namely Fedora 42. They would start the boot process , which would then fail (Post hoc ergo propter hoc?) I then installed Fedora 43 on both machines; however when this was done two of my four mouses stopped working, the symptom being that they appear+ to pair and connect (as reported by the systems), but that moving the mouse doesn't move the cursor. Here's the breakdown: Working mouses: Logitech M325 wireless mouse (Proprietary communication with its dongle) HP Wired 3 button mouse (very old) Nonworking mouses Dell wireless mouse (Gift of a friend; no more info available) MS Bluetooth Ergonomic Mouse Also lately, the Bluetooth system doesn't start; here is the latest status message indicating the problem: $ systemctl status bluetooth. bluetooth.service bluetooth.target jonrysh@localhost-live:~$ systemctl status bluetooth.service ○ bluetooth.service - Bluetooth service Loaded: loaded (/usr/lib/systemd/system/bluetooth.service; enabled; preset: enabled) Drop-In: /usr/lib/systemd/system/service.d └─10-timeout-abort.conf Active: inactive (dead) Docs: man:bluetoothd(8) Nov 13 09:49:00 localhost-live systemd[1]: bluetooth.service - Bluetooth service was skipped because of an unmet condition check (ConditionPathIsDirectory=/sys/class/blue> Nov 13 09:50:45 localhost-live systemd[1]: bluetooth.service - Bluetooth service was skipped because of an unmet condition check (ConditionPathIsDirectory=/sys/class/blue> My current plan is to buy a new Bluetooth dongle and see if it works; but it would be nice to know what is going on. There a number of threads on the Web about similar issues; the most useful looks like: https://www.reddit.com/r/archlinux/comments/zweejv/how_to_fix_error_bluetoo… How to fix error? Bluetooth service was skipped because of a failed condition check (ConditionPathIsDirectory=/sys/class/bluetooth). Here's the info on my system: Operating System: Fedora Linux 43 KDE Plasma Version: 6.5.2 KDE Frameworks Version: 6.19.0 Qt Version: 6.10.0 Kernel Version: 6.17.7-300.fc43.x86_64 (64-bit) Graphics Platform: Wayland Processors: 8 × Intel® Core™ i7-4790K CPU @ 4.00GHz Memory: 16 GiB of RAM (15.5 GiB usable) Graphics Processor: Intel® HD Graphics 4600 Manufacturer: ASUS Product Name: All Series -- Jonathan Ryshpan <jonrysh(a)pacbell.net> Berkeley Linux Team

4 3

OT, Having problems with Yahoo mail and Thunderbird
by Robert McBroom 14 Nov '25

14 Nov '25

Yahoo Mail won't accept the password any more. Get the message Edge: To many requests. Can log in fine in Firefox. Tried generating a app password for Thunderbird as suggested on the web but no success. Is there another email app that people are successfully accessing yahoo mail?

8 15

F42 refuses to start
by Terry Hurlbut 14 Nov '25

14 Nov '25

Yesterday I turned on my computer to boot F42 after an overnight shutdown. It refused to load. The display got through all three points of the ellipsis, then showed a blinking underscore for five seconds. Then the underscore showed steady, and the booting got no further. Ten times I shut it down with the Power button and waited for it to reboot, with the same result. My device has a pair of SSDs, with the OS and most applications on one, and all my data on the other. So if I could boot the machine from anything, and then do any kind of installation of F43, I don't think I would lose my data. So would someone tell me, please, how to restore functionality to my desktop, with or without upgrading to F43, whether with a clean install or by using DNF5 (sp ) ? By the way, I have used KDE for years, and would prefer to stay in that environment and not switch to GNOME. Temlakos <temlakos(a)gmail.com>

8 31

SELinux: or how I learned to start worrying and hate the bomb
by Sam Varshavchik 14 Nov '25

14 Nov '25

According to Google Skynet, Fedora Core 3 was the first release that had SELinux enabled by default. That was more than 20 years ago. I think that 20 years is more than long enough for this kind of technology to mature, and work the kinks out, and then it's smooth sailing from that point on. Yet, for the umpteenth time I have to create yet another bug for an AVC failure, and watch it being dupe-hammered, and everyone else's reports also having the same fate, for the next year or so. Why? I have to conclude that there's something fundamentally broken with SELinux. SELinux AVC denials should be rare. Like once in a blue moon. The fact that they still come out of the firehose, non-stop, and some poor soul has to chase down as many as possible, letting all others expire and autoclose – that is not right. This shouldn't happen. The current state of affairs would not be unreasonable say, within the first five years of SELinux's existence. But not 20 years later. Come on. Either fix the fscking thing, or get rid of it. When I have to install a cron job to run restorecon every five minutes, while the corresponding bug ages, that's a big honking clue that something is wrong in the state of Denmark. I might be wrong, but AFAIK Fedora/RHEL is the only Linux distribution that still screws around with SELinux. Google Skynet also mentioned that something called "Amazon Linux" also has SELinux enabled. Who? That doesn't count. So, resuming: if anyone ever wonders why only Fedora/RHEL bothers with SELinux, hopefully this clears it up. I believe that SELinux is fundamentally broken, and I don't think it's fixable in its current state. Fortunately, it is easy to turn it off. But is that what SELinux's advocates will say is the solution? Really? And if not, what /is/ the solution? Can someone come up with the answer to the following answer: what can, and should be done, to fix the constant pain point of AVC denials, permanently?

8 14

Re: F41-43 Mate setting window size
by Steven I Usdansky 11 Nov '25

11 Nov '25

Control Center -> Look and Feel -> Windows -> Placement uncheck Enable windows tiling (may need to log out and back in) -- Steven I Usdansky, PhD Traveling Geologist

2 1