Hi! As the title says, I get random freezes. I thought it was only for the browsers (currently using Edge and Zen Browser), but then it also started happening with other apps (calendar, terminal, and so on). At first I also thought that it was only when watching videos, but it also happens just reading forums, websites, using the system, and so on.
Before, with F40 I never had this issue and everything worked smoothly, but now I get this freezes very often, for a few seconds, making the whole system unresponsive. When that happens, I usually get a few freezes back to back, maybe 6 or 7 of them at the same time one after other.
This is too annoying and make the system unusable. I have to use Windows 11 instead to have a normal browsing session.
I have uploaded a screenshot with my specs. Also, the 3 lines that I always see in the journal when the freezes happen. It seems to be related to Vulkan and Intel drivers. I have an integrated Intel GPU (Intel Corporation WhiskeyLake-U GT2 [UHD Graphics 620]), I do not have Nvidia or any other dedicates graphic card.
EDIT: Now after creating this topic I got like 6 freezes. And now editing it I am getting a few more, making it almost impossible to edit the message since the freezes keeps coming up and cannot write or save the edit.
Thanks Mike! I had already tried with those: gl and ngl, and it does not fix the issue. I still get the freezes.
I thought that it could work because some users on forums and Reddit said this workaround stopped the freezes for them. But unfortunately, in my case, it does not.
Or maybe I am not doing it correctly. This is what my /etc/enviroment looks like:
Would you like me to create a new user profile and try to reproduce the issue? Right now I only have one user.
Sometimes it takes like one or a couple of hours to start happening. Sometimes starts sooner. It is kind of random I would say. Or I have to do something to trigger it which I do not know what that could be.
I will create a new user and try to reproduce it. I will get back to you after seeing how that goes.
I’ve been unable to reproduce the issue in any of the user profiles so far. It’s been a persistent problem for weeks, but now, when I need to report back, it’s not occurring.
It will happen again for sure, so I will paste the journal lines here when that happens again.
echo $GSK_RENDERER
That shows nothing in the Terminal. Probably because as the GSK_RENDERER=gl (or ngl) did not fix the problem for me yesterday, I deleted the line from the enviroment file.
That’s what I wanted to see: if you did keep the env. variable active or not. I was wondering if the vulkan warning still appearing when the graphics renderer is being switched to gl or ngl.
I happened to stumble on this previous post that was edited recently. It is generally not recommended to edit existing posts (except for typos or very recent changes), because there is a high chance they get missed.
The /etc/environment file seems to have been edited correctly. You know for sure if it is ok, if the echo $GSK_RENDERER command returns the correct value after a reboot. Now that I look again closer, I see a typo, i.e. “enviroment” instead of “environment”.
If this typo happened also initially, as you tried the first time and it failed, and not tested the variable, then there are good chances that the env. variable wasn’t set up correctly.
To make sure, you could run the one-line command I linked above in my earlier post. You can do this now, or later, in case the issue reappears.
Damn! Indeed there is a typo. English is not my native language so sometimes it happens…
I will do it as you say this time and will report back how it goes. Thanks so much!
In the meantime, it has happened again a few min ago, so I will paste here the lines from the journal. It happened around 00:20, and also around 00:24:
nov 20 00:19:10 fedora systemd[1]: Started snapper-cleanup.service - Daily Cleanup of Snapper Snapshots.
nov 20 00:19:10 fedora audit[1]: SERVICE_START pid=1 uid=0 auid=4294967295 ses=4294967295 subj=system_u:system_r:init_t:s0 msg='unit=snapper-cleanup comm="systemd" exe=">
nov 20 00:19:10 fedora systemd[1]: Starting snapperd.service - DBus interface for snapper...
nov 20 00:19:10 fedora systemd[1]: Started snapperd.service - DBus interface for snapper.
nov 20 00:19:10 fedora audit[1]: SERVICE_START pid=1 uid=0 auid=4294967295 ses=4294967295 subj=system_u:system_r:init_t:s0 msg='unit=snapperd comm="systemd" exe="/usr/li>
nov 20 00:19:10 fedora systemd-helper[75582]: Running cleanup for 'home'.
nov 20 00:19:10 fedora systemd-helper[75582]: Running number cleanup for 'home'.
nov 20 00:19:10 fedora systemd-helper[75582]: Running timeline cleanup for 'home'.
nov 20 00:19:10 fedora systemd-helper[75582]: Suprimiendo instantánea de home:
nov 20 00:19:10 fedora systemd-helper[75582]: 851
nov 20 00:19:10 fedora systemd-helper[75582]: Suprimiendo instantánea de home:
nov 20 00:19:10 fedora systemd-helper[75582]: 894
nov 20 00:19:10 fedora systemd-helper[75582]: Running empty-pre-post cleanup for 'home'.
nov 20 00:19:10 fedora systemd-helper[75582]: Running 'btrfs qgroup clear-stale /home'.
nov 20 00:19:10 fedora systemd-helper[75582]: Running cleanup for 'root'.
nov 20 00:19:10 fedora systemd-helper[75582]: Running number cleanup for 'root'.
nov 20 00:19:10 fedora systemd-helper[75582]: Running timeline cleanup for 'root'.
nov 20 00:19:10 fedora systemd-helper[75582]: Suprimiendo instantánea de root:
nov 20 00:19:10 fedora systemd-helper[75582]: 1061
nov 20 00:19:10 fedora systemd-helper[75582]: Suprimiendo instantánea de root:
nov 20 00:19:10 fedora systemd-helper[75582]: 1164
nov 20 00:19:10 fedora systemd-helper[75582]: Running empty-pre-post cleanup for 'root'.
nov 20 00:19:10 fedora systemd-helper[75582]: Running 'btrfs qgroup clear-stale /'.
nov 20 00:19:11 fedora systemd[1]: snapper-cleanup.service: Deactivated successfully.
nov 20 00:19:11 fedora audit[1]: SERVICE_STOP pid=1 uid=0 auid=4294967295 ses=4294967295 subj=system_u:system_r:init_t:s0 msg='unit=snapper-cleanup comm="systemd" exe="/>
nov 20 00:20:11 fedora systemd[1]: snapperd.service: Deactivated successfully.
nov 20 00:20:11 fedora audit[1]: SERVICE_STOP pid=1 uid=0 auid=4294967295 ses=4294967295 subj=system_u:system_r:init_t:s0 msg='unit=snapperd comm="systemd" exe="/usr/lib>
nov 20 00:20:42 fedora systemd[2916]: Started dbus-:1.2-org.gnome.Logs@1.service.
nov 20 00:20:43 fedora gnome-logs[75789]: MESA-INTEL: warning: ../src/intel/vulkan/anv_formats.c:763: FINISHME: support YUV colorspace with DRM format modifiers
nov 20 00:20:43 fedora gnome-logs[75789]: MESA-INTEL: warning: ../src/intel/vulkan/anv_formats.c:794: FINISHME: support more multi-planar formats with DRM modifiers
nov 20 00:20:53 fedora systemd[2916]: dbus-:1.2-org.gnome.Logs@1.service: Consumed 1.740s CPU time, 180.2M memory peak.
nov 20 00:20:54 fedora systemd[2916]: Started dbus-:1.2-com.gexperts.Tilix@3.service.
nov 20 00:20:55 fedora systemd[2916]: Started vte-spawn-8d04c2d0-9761-4eb7-a811-286309a7d8b9.scope - VTE child process 75857 launched by tilix process 75850.
nov 20 00:21:08 fedora geoclue[3242]: Failed to query location: Query location SOUP error: Not Found
log file: udo nano /etc/enviroment sudo nano /etc/enviroment
nov 20 00:19:10 fedora systemd-helper[75582]: Suprimiendo instantánea de root:
nov 20 00:19:10 fedora systemd-helper[75582]: 1061
nov 20 00:19:10 fedora systemd-helper[75582]: Suprimiendo instantánea de root:
nov 20 00:19:10 fedora systemd-helper[75582]: 1164
nov 20 00:19:10 fedora systemd-helper[75582]: Running empty-pre-post cleanup for 'root'.
nov 20 00:19:10 fedora systemd-helper[75582]: Running 'btrfs qgroup clear-stale /'.
nov 20 00:19:11 fedora systemd[1]: snapper-cleanup.service: Deactivated successfully.
nov 20 00:19:11 fedora audit[1]: SERVICE_STOP pid=1 uid=0 auid=4294967295 ses=4294967295 subj=system_u:system_r:init_t:s0 msg='unit=snapper-cleanup comm="systemd" exe="/>
nov 20 00:20:11 fedora systemd[1]: snapperd.service: Deactivated successfully.
nov 20 00:20:11 fedora audit[1]: SERVICE_STOP pid=1 uid=0 auid=4294967295 ses=4294967295 subj=system_u:system_r:init_t:s0 msg='unit=snapperd comm="systemd" exe="/usr/lib>
nov 20 00:20:42 fedora systemd[2916]: Started dbus-:1.2-org.gnome.Logs@1.service.
nov 20 00:20:43 fedora gnome-logs[75789]: MESA-INTEL: warning: ../src/intel/vulkan/anv_formats.c:763: FINISHME: support YUV colorspace with DRM format modifiers
nov 20 00:20:43 fedora gnome-logs[75789]: MESA-INTEL: warning: ../src/intel/vulkan/anv_formats.c:794: FINISHME: support more multi-planar formats with DRM modifiers
nov 20 00:20:53 fedora systemd[2916]: dbus-:1.2-org.gnome.Logs@1.service: Consumed 1.740s CPU time, 180.2M memory peak.
nov 20 00:20:54 fedora systemd[2916]: Started dbus-:1.2-com.gexperts.Tilix@3.service.
nov 20 00:20:55 fedora systemd[2916]: Started vte-spawn-8d04c2d0-9761-4eb7-a811-286309a7d8b9.scope - VTE child process 75857 launched by tilix process 75850.
nov 20 00:21:08 fedora geoclue[3242]: Failed to query location: Query location SOUP error: Not Found
nov 20 00:21:31 fedora microsoft-edge.desktop[68477]: [68470:68470:1120/002131.471428:ERROR:CONSOLE(138)] "%cfluid:telemetry:fluid:telemetry:ConnectionManager %csubmitSi>
nov 20 00:21:58 fedora cupsd[1993]: REQUEST localhost - - "POST / HTTP/1.1" 200 185 Renew-Subscription successful-ok
nov 20 00:23:23 fedora microsoft-edge.desktop[68477]: [68470:68470:1120/002323.143874:ERROR:CONSOLE(143)] "fluid:telemetry:OdspDriver:DeltaConnection:ServerDisconnect, c>
nov 20 00:23:23 fedora microsoft-edge.desktop[68477]: [68470:68470:1120/002323.145321:ERROR:CONSOLE(143)] "fluid:telemetry:Container:ConnectionStateChange_Disconnected, >
nov 20 00:23:23 fedora microsoft-edge.desktop[68477]: [68470:68470:1120/002323.147012:ERROR:CONSOLE(143)] "fluid:telemetry:Container:ConnectionStateChange_EstablishingCo>
nov 20 00:23:23 fedora microsoft-edge.desktop[68477]: [68470:68470:1120/002323.151974:ERROR:CONSOLE(143)] "fluid:telemetry:OdspDriver:DeltaConnection:ServerDisconnect, c>
nov 20 00:23:23 fedora microsoft-edge.desktop[68477]: [68470:68470:1120/002323.153390:ERROR:CONSOLE(143)] "fluid:telemetry:Container:ConnectionStateChange_Disconnected, >
nov 20 00:23:23 fedora microsoft-edge.desktop[68477]: [68470:68470:1120/002323.155551:ERROR:CONSOLE(143)] "fluid:telemetry:Container:ConnectionStateChange_EstablishingCo>
nov 20 00:23:23 fedora microsoft-edge.desktop[68477]: [68470:68470:1120/002323.756798:ERROR:CONSOLE(143)] "fluid:telemetry:Container:ConnectionStateChange_CatchingUp, ca>
nov 20 00:23:23 fedora microsoft-edge.desktop[68477]: [68470:68470:1120/002323.814119:ERROR:CONSOLE(143)] "fluid:telemetry:Container:ConnectionStateChange_CatchingUp, ca>
nov 20 00:23:31 fedora systemd[2916]: Started dbus-:1.2-org.gnome.Logs@2.service.
nov 20 00:23:32 fedora gnome-logs[76401]: MESA-INTEL: warning: ../src/intel/vulkan/anv_formats.c:763: FINISHME: support YUV colorspace with DRM format modifiers
nov 20 00:23:32 fedora gnome-logs[76401]: MESA-INTEL: warning: ../src/intel/vulkan/anv_formats.c:794: FINISHME: support more multi-planar formats with DRM modifiers
nov 20 00:23:39 fedora systemd[2916]: dbus-:1.2-org.gnome.Logs@2.service: Consumed 1.200s CPU time, 118M memory peak.
nov 20 00:26:08 fedora geoclue[3242]: Failed to query location: Query location SOUP error: Not Found
nov 20 00:26:47 fedora microsoft-edge.desktop[68477]: [68470:68470:1120/002647.083470:ERROR:interface_endpoint_client.cc(725)] Message 2 rejected by interface blink.mojo>
nov 20 00:26:47 fedora microsoft-edge.desktop[68477]: [68470:68470:1120/002647.342931:ERROR:CONSOLE(31)] "The workspace has been cleared, the document is going to be del>
nov 20 00:27:30 fedora microsoft-edge.desktop[68477]: [68470:68470:1120/002730.250312:ERROR:CONSOLE(31)] "The workspace has been cleared, the document is going to be del>
nov 20 00:27:43 fedora gnome-shell[63275]: Window manager warning: last_focus_time (22868476) is greater than comparison timestamp (22868431). This most likely represen>
nov 20 00:27:43 fedora gnome-shell[63275]: Window manager warning: last_user_time (22868476) is greater than comparison timestamp (22868431). This most likely represent>
nov 20 00:27:43 fedora gnome-shell[63275]: Window manager warning: W96 (Tilix: javier@fedora:~) appears to be one of the offending windows with a timestamp of 22868476.
I also get freezes with KDE Plasma so its not only Gnome. I did a fresh installation of Fedora 41 KDE Spin two days ago and i did not fiddle around with the system at all so there is no way i broke something. I just enabled nonfree/free rpms and installed the nvidia drivers and a few apps like discord and that kind of stuff. Left everything else at default settings except i switched to breeze dark theme.
I got a few freezes for a few seconds when i used the firefox browser. But also in other situations. For example i downloaded a file and everytime i opened the downloads folder the whole desktop stopped responding for 30 seconds and froze. After trying to access the downloads folder a few times after 3-4 freezes the issue magically stopped happening and worked.
I also had similar issues after i did a fresh installation of EndeavourOS. As of now the only distro where i did not encounter a weird glitch or freeze was Linux Mint 22 Cinnamon.
While you might have the same issue as the OP, there are also changes it’s not. In case this thread won’t reveal your issue, you might want to open a new one, given that nvidia drivers can cause their own set of problems.
Hi Mike! After testing the fix fo a couple of days, I must say it seems I get less freezes, but I still get them.
I cannot see the “vulkan” error in the journal anymore. Now, when Edge freezes, I see no lines in the journal that could be linked to the freeze at the time it happens, or even no lines at all (for Edge).
Right now, I just got a few freezes for Betterbird (Thunderbird fork). It always ran smoothly, but now it is just frustrating and I cannot even work on my emails. I do see in the journal some lines for Betterbird at 18:01-18:02:
EDIT: I just got two freezes for Edge at 18:58, but only two is already an improvement compared to all the freezes I would get before. I think it is related the Snapper cleanup service that runs, I think, every hour. But this does not seem to be related to the freezes for Betterbird above.
If you would like to troubleshoot further, you’d need to identify the exact apps that are still causing freezes. That way you might you could identify if it is still a GNOME specific issue (e.g. if GNOME core apps are still freezing), or just an issue with specific apps (e.g. Betterbird is not a GNOME app, might not be as well tested as Thunderbird, it’s a Flatpak etc; when issue is Edge, test if similar issues happen with Firefox, which comes preinstalled etc).
It happens with Edge, Betterbird, Tilix, BTRFS assistant. The usual apps I use. I had been using all of those apps in Fedora 40 for months with no issues. It was only after upgrading to F41 when all these freezes started.
Now it is even worse, not only the freezes are more frequent (again) but now to write this message Edge is very laggy. It is taking me long time to be able to write this post.
I have no make any changes or install any other apps in F41. Same apps and config as in F40.
Honestly, I have no time to try to troubleshoot all this and I will probably not be able to fix it. I might just have to format Fedora and try another distro or just stick with Windows 11. I really liked Fedora. But I cannot use such an unstable system (F41 in my laptop I mean, many other people probably have no issues). I do not know what they have done with F41 but, for me, it works very bad and unstable.
And my laptop is a Lenovo X1 Carbon. Supposedly they have some kind of collaboration with Fedora for support, drivers, and so on.
When this happens are you able to switch to another application or is the entire system basically paused or a second or so?
Can you, for example, switch to a different session (Ctrl+Shift+Fn) where you could have a session already logged in to the cli, running htop to see if there’s a large spike CPU elsewhere, or a burst of paging/ram reclamation/compression/something. I can understand a spiky system struggling with a driver issue or a rendering pipeline problem, but struggling to see keyboard presses rendering in a text box within a browser smells like something fundamentally soiling itself.
I assume there’s nothing obvious in the system log when this happens, or any of the other logs.
Are all of these applications which are giving you grief ALL installed via snap or flatpak?
Ever get the same kind of glitch/pause/stutter/whatever with applications which are installed directly - vim, emacs, mutt, whatever else you happen to use which is not in some kind of container?
Folks … JJ Magnesium mentioned that when he created another user, the issue did not happen. This suggests that it has to do with the contents of ~/.config.
JJ, try logging into the system via the command line and moving the .config directory in your home directory to something like DOT-config and then log in as you normally do. Please note that this resets your basic environment to its default.
I think I have found the culprits. I am still doing some testing to make sure about it and discard what might not be affecting the performance.
As a summary:
I removed the line of GSF_RENDERER=gl and went back to Vulkan because the system felt laggier and more unstable with GL. It seems that even I have often the Vulkan error on the journal, is not really affecting, or at least not much, to cause these freezes.
I deactivated the Snapper cleanup service which run every hour. After doing this, everything improved. But it still did not feel right.
BTRFS quota. I deactivated this option. I think, if I recall well, I activated one or two days before upgrading to F41, or one or two days after upgrading. I think that this is the main culprit. Along with Snapper cleanup service, which made this point even worse.
Not sure yet because the freezes and low performance was kind of random. But so far it seems it is working correctly. Just some small lags every now an then while browsing with Edge. It could be Edge itself (although previously did not happen) or related to this error on the journal, (related to Intel GPU drivers):
i915 0000:00:02.0: [drm] *ERROR* Atomic update failure on pipe A
Although when these lags, that are not a big deal, happens, sometimes I see no lines whatsoever on the journal, and sometimes I find the line above.
If everything keeps working correctly I will then activate the cleanup service again (without the quotas) to see how it goes.
.
