nvidia-settings suddenly started smashing CPU at boot
Bug
Summary
At boot, Xorg.bin is using 100% CPU and nvidia-settings is using 50%.
Coolercontrol's UI is unresponsive, the graph does not update
Issuing sudo killall nvidia-settings
(sometimes I need to do this a few times over) fixes everything.
It just magically started doing this today. I'm sure of this because I've another bug with wireplumber/pipewire which has me looking at top
all the time, so I'd have noticed this. I went to sleep, it was fine, while I slept the CPU must have been busy doing something (probably some background maintenance task) because the log is full of the usual errors from nvidia-settings when the fans change (Authorization required, but no authorization protocol specified
- it always does this because it's Wayland and nvidia-settings complains but nothing ever fails) I rebooted tonight and when it came up I noticed this right away (because I've been looking at top
all the time like I said) No packages were updated inbetween any of this, it just started out of nowhere.
Environment
- CoolerControl version: 1.17, also tried 1.16
- Installation method: Repository
- Distribution name and version: OpenSUSE Tumbleweed
- Desktop: KDE (during testing this still occurs on Wayland and X11 sessions, also IceWM, also just from TTY when SDDM is waiting for a login)
- Cooling device model(s): nct6798 on the motherboard
Steps to reproduce
- Boot
- That's it.
Relevant logs and/or screenshots
Nothing seems out of the ordinary in the logs except I don't recall this error 2
Shutdown error: No such file or directory (os error 2)` being there before. All the other stuff is normal. The -40 temp is an external K-type input which is not connected, the unlabelled device is my wifi and it's always like that, I always get the liqctld error.
Sep 06 03:14:45 Pallas systemd[1]: Started Cooler Control Daemon.
Sep 06 03:14:45 Pallas coolercontrold[12027]: Could not establish communication with coolercontrol-liqctld socket connection, retry #1.
error sending request for url (http://127.0.0.1:11986/handshake): error trying to connect: tcp connect error: Connection refused (os error 111)
Sep 06 03:24:44 Pallas coolercontrold (deleted)[12027]: Temperature value: -40 at "/sys/class/hwmon/hwmon2"/temp4_input is outside of usable range. Most likely the sensor is not reporting real readings
Sep 06 03:24:44 Pallas coolercontrold (deleted)[12027]: Temp label doesn't exist for "/sys/class/hwmon/hwmon4"/temp1_label
Sep 06 03:29:26 Pallas systemd[1]: Stopping Cooler Control Daemon...
Sep 06 03:29:27 Pallas coolercontrold (deleted)[12027]: Shutdown error: No such file or directory (os error 2)
Sep 06 03:29:27 Pallas systemd[1]: coolercontrold.service: Deactivated successfully.
Sep 06 03:29:27 Pallas systemd[1]: Stopped Cooler Control Daemon.