System Crashing - not new but more information - just looking for ideas

[LuigiCalamar]

I've mentioned before that my system will crash from time to time. I have to unplug my PC to get it back. I've shared some logs and such in the past but I have some more information - just looking for thoughts/ideas if this is making any sense:

At the bottom of this post I'll include my PC specs.

So the main things I'm running on this machine are:

• Firefox (often multiple tabs)
• QBitorrent
• Plex Media Server
• Beeper (communication package)
• Windscribe (VPN)
• I also use Radaar and Sonaar to auto find shows (they don't seem to use much )

That's 95% of my processes.

Usually when it crashes I'm running all the above.

Lately as I've been watching this - it seems that FireFox is by far the largest resource user.... I do have a number of add-ons.
--- of these tabs - I use ProtonMail which seems to be especially resource consuming.

I do 90% of my PC activities in Firefox... so if I'm on my computer I'm likely in Firefox.

Now - as soon as I'm done using my PC I shutdown Firefox... shutdown Beeper (which is also a big user of resources). I typically leave Qbitorrent up - but will often just quit it also (I need to put it on a schedule to only run at night...)

Quitting Firefox (and other apps) seems to greatly reduce my system crashes.

Note: I've also increased my swap from 500gb to 4tb... didn't do much.

If it makes sense that Firefox etal are overloading my system - any ideas how to manage (other than quitting) - or how to better log what is crashing things... (logs haven't done much in the past but perhaps looking for specifics can help?)



Operating System: Kubuntu 26.04 LTS
KDE Plasma Version: 6.6.6
KDE Frameworks Version: 6.24.0
Qt Version: 6.10.2
Kernel Version: 7.0.0-31-generic (64-bit)
Graphics Platform: Wayland
Processors: 16 × 13th Gen Intel® Core™ i7-13620H
Memory: 32 GiB of RAM (31.1 GiB usable)
Graphics Processor: Intel® Graphics
Manufacturer: GMKtec
Product Name: NucBox K7 PLUS
System Version: V1.0

[die.boer]

Firefox may be contributing to the problem, but with 32 GB RAM I wouldn't assume that simply having a lot of Firefox tabs open should hard-lock the entire machine. The fact that you have to physically power-cycle it makes me more suspicious of a kernel/GPU/driver/hardware issue, with Firefox potentially acting as the trigger.

I'd first try Firefox with hardware acceleration disabled, then Firefox with extensions disabled, and separately test without Windscribe. If one of those makes the crashes disappear, we've got a much better direction to investigate.

I'd also check the previous boot's kernel log immediately after a crash:

journalctl -b -1 -k -p 0..4

and particularly look for i915, drm, gpu, oom, lockup, watchdog, MCE, or hardware error.
or post the output of command here

Increasing swap isn't likely to help if the actual problem is a GPU/driver/kernel lockup.

I've also increased my swap from 500gb to 4tb

is that a typo ? Please also post output of

swapon --show

free -h

[LuigiCalamar]

yes - a mistype... 4 gb

[LuigiCalamar]

ep 21 07:54:12 plextop kernel: #1 #3 #5 #7 #9 #11
Sep 21 07:54:12 plextop kernel: hpet_acpi_add: no address or irqs in _CRS
Sep 21 07:54:12 plextop kernel: i8042: PNP: PS/2 appears to have AUX port disabled, if this is incorrect please boot with i8042.nopnp
Sep 21 07:54:12 plextop kernel: ENERGY_PERF_BIAS: Set to 'normal', was 'performance'
Sep 21 07:54:12 plextop systemd-journald[392]: File /var/log/journal/dc4210b5870e4fdc8f33ad960f0f9435/system.journal corrupted or uncleanly shut down, renamin>
Sep 21 07:54:12 plextop kernel: ACPI BIOS Error (bug): Could not resolve symbol [^^RP28.PXSX.WIST], AE_NOT_FOUND (20251212/psargs-332)
Sep 21 07:54:12 plextop kernel:
Sep 21 07:54:12 plextop kernel: No Local Variables are initialized for Method [IFUN]
Sep 21 07:54:12 plextop kernel:
Sep 21 07:54:12 plextop kernel: Initialized Arguments for Method [IFUN]: (4 arguments defined for method invocation)
Sep 21 07:54:12 plextop kernel: Arg0: 000000000c316fd0 <Obj> Buffer(16) 2C 17 66 72 0B 22 29 4B
Sep 21 07:54:12 plextop kernel: Arg1: 00000000d36b6fa3 <Obj> Integer 0000000000000000
Sep 21 07:54:12 plextop kernel: Arg2: 00000000d70a0b8f <Obj> Integer 0000000000000002
Sep 21 07:54:12 plextop kernel: Arg3: 000000009097c573 <Obj> Buffer(4) 01 00 00 00
Sep 21 07:54:12 plextop kernel:
Sep 21 07:54:12 plextop kernel: ACPI Error: Aborting method \_SB.PC00.CNVW.IFUN due to previous error (AE_NOT_FOUND) (20251212/psparse-529)
Sep 21 07:54:12 plextop kernel: ACPI Error: Aborting method \_SB.PC00.CNVW._DSM due to previous error (AE_NOT_FOUND) (20251212/psparse-529)
Sep 21 07:54:12 plextop kernel: spi-nor spi1.0: supply vcc not found, using dummy regulator
Sep 21 07:54:14 plextop kernel: nvme nvme0: using unchecked data buffer
Sep 21 07:54:14 plextop kernel: block nvme1n1: No UUID available providing old NGUID
Sep 21 07:54:15 plextop systemd-journald[392]: File /var/log/journal/dc4210b5870e4fdc8f33ad960f0f9435/user-1000.journal corrupted or uncleanly shut down, rena>
Sep 21 07:54:18 plextop kernel: warning: `kdeconnectd' uses wireless extensions which will stop working for Wi-Fi 7 hardware; use nl80211
Sep 21 08:00:24 plextop kernel: kauditd_printk_skb: 221 callbacks suppressed
Sep 21 08:38:28 plextop kernel: workqueue: output_poll_execute hogged CPU for >10000us 4 times, consider switching to WQ_UNBOUND
Sep 21 08:38:28 plextop kernel: workqueue: i915_hpd_poll_init_work [i915] hogged CPU for >10000us 4 times, consider switching to WQ_UNBOUND
Sep 21 08:38:51 plextop kernel: workqueue: output_poll_execute hogged CPU for >10000us 5 times, consider switching to WQ_UNBOUND
Sep 21 08:38:51 plextop kernel: workqueue: i915_hpd_poll_init_work [i915] hogged CPU for >10000us 5 times, consider switching to WQ_UNBOUND
Sep 21 08:39:36 plextop kernel: workqueue: output_poll_execute hogged CPU for >10000us 7 times, consider switching to WQ_UNBOUND
Sep 21 08:39:37 plextop kernel: workqueue: i915_hpd_poll_init_work [i915] hogged CPU for >10000us 7 times, consider switching to WQ_UNBOUND
Sep 21 08:41:07 plextop kernel: workqueue: output_poll_execute hogged CPU for >10000us 11 times, consider switching to WQ_UNBOUND
Sep 21 08:41:07 plextop kernel: workqueue: i915_hpd_poll_init_work [i915] hogged CPU for >10000us 11 times, consider switching to WQ_UNBOUND
Sep 21 08:44:06 plextop kernel: workqueue: output_poll_execute hogged CPU for >10000us 19 times, consider switching to WQ_UNBOUND

Sep 21 08:44:06 plextop kernel: workqueue: i915_hpd_poll_init_work [i915] hogged CPU for >10000us 19 times, consider switching to WQ_UNBOUND
Sep 21 08:50:12 plextop kernel: workqueue: output_poll_execute hogged CPU for >10000us 35 times, consider switching to WQ_UNBOUND
Sep 21 08:50:12 plextop kernel: workqueue: i915_hpd_poll_init_work [i915] hogged CPU for >10000us 35 times, consider switching to WQ_UNBOUND
Sep 21 10:00:53 plextop kernel: workqueue: i915_hpd_poll_init_work [i915] hogged CPU for >10000us 67 times, consider switching to WQ_UNBOUND
Sep 21 10:01:15 plextop kernel: workqueue: output_poll_execute hogged CPU for >10000us 67 times, consider switching to WQ_UNBOUND
Sep 21 11:20:25 plextop kernel: workqueue: i915_hpd_poll_init_work [i915] hogged CPU for >10000us 131 times, consider switching to WQ_UNBOUND
Sep 21 11:20:47 plextop kernel: workqueue: output_poll_execute hogged CPU for >10000us 131 times, consider switching to WQ_UNBOUND
Sep 21 12:39:47 plextop kernel: workqueue: i915_hpd_poll_init_work [i915] hogged CPU for >10000us 259 times, consider switching to WQ_UNBOUND
Sep 21 12:40:32 plextop kernel: workqueue: output_poll_execute hogged CPU for >10000us 259 times, consider switching to WQ_UNBOUND
Sep 21 14:16:28 plextop kernel: workqueue: i915_hpd_poll_init_work [i915] hogged CPU for >10000us 515 times, consider switching to WQ_UNBOUND
Sep 21 14:17:13 plextop kernel: workqueue: output_poll_execute hogged CPU for >10000us 515 times, consider switching to WQ_UNBOUND
Sep 21 17:50:49 plextop kernel: workqueue: i915_hpd_poll_init_work [i915] hogged CPU for >10000us 1027 times, consider switching to WQ_UNBOUND
Sep 21 17:51:57 plextop kernel: workqueue: output_poll_execute hogged CPU for >10000us 1027 times, consider switching to WQ_UNBOUND
Sep 22 00:35:16 plextop kernel: workqueue: i915_hpd_poll_init_work [i915] hogged CPU for >10000us 2051 times, consider switching to WQ_UNBOUND
Sep 22 00:37:31 plextop kernel: workqueue: output_poll_execute hogged CPU for >10000us 2051 times, consider switching to WQ_UNBOUND
lines 32-47/47 (END)


lines 1-16

[LuigiCalamar]

That is the output from > journalctl -b -1 -k -p 0..4 I do see the i915 error.

Doing some research - I tried this suggestion:
"I found that to fix this I had to enable the nvidia DRM kernal module to do this I added a file in the /etc/modprobe.d/ folder called NVIDIA_DRM.conf (can be any name so long as it ends in .conf) and in the contents of that file shoud be options nvidia_drm modeset=1

You can do this with the command

echo "options nvidia_drm modeset=1" | sudo tee /etc/modprobe.d/NVIDIA_DRM.conf
"
Haven't seen if that helps at all yet.

[die.boer]

Those logs actually give us a much more interesting lead than Firefox simply using too much RAM.

The thing that jumps out is the repeated:

i915_hpd_poll_init_work [i915] hogged CPU for >10000us

along with:

output_poll_execute hogged CPU for >10000us

Those messages start around 08:38 and then increase from 4 occurrences to 5, 7, 11, 19, 35, 67, 131, 259, 515, 1027 and eventually 2051 occurrences.

i915 is Intel's graphics driver, and HPD is Hot Plug Detect, so this looks like something in the Intel graphics/display stack is repeatedly getting bogged down.

That doesn't necessarily mean Firefox is innocent. Firefox makes heavy use of GPU acceleration, so it could be triggering an existing i915/display problem rather than actually exhausting your RAM.

I'd try one simple test first: temporarily disable Firefox hardware acceleration under Settings → General → Performance, then see whether the crashes continue.

I'd also be interested in what displays are connected to this machine. Are you using HDMI/DisplayPort, a TV, multiple monitors, an HDMI switch/receiver, etc.? The i915 HPD messages make the display setup relevant.

Could you also post the output of:

inxi -Gxx

I wouldn't concentrate too much on the ACPI messages yet. They're worth noting, but the repeated i915 workqueue messages are considerably more interesting given the complete freezes you're experiencing.

Also, the "journal corrupted or uncleanly shut down" messages are probably just a consequence of having to forcibly power the machine off after a crash, rather than the cause.

[LuigiCalamar]

I turned off firefox hardware acceleration.... I did turn that off a couple days ago but I'd since done a firefox reset - and it was turned back on... should have caught that... so going forward I'll leave it off.

Here is the output from the inxi -Gxx :

Graphics:
Device-1: Intel Raptor Lake-P [UHD Graphics] driver: i915 v: kernel arch: Xe
ports:active: HDMI-A-1 empty: DP-1, DP-2, DP-3, HDMI-A-2, HDMI-A-3,
HDMI-A-4 bus-ID: 00:02.0 chip-ID: 8086:a7a8
Display: wayland server: X.org v: 1.21.1.22 with: Xwayland v: 24.1.10
compositor: kwin_wayland driver:gpu: i915 display-ID: 0
Monitor-1: HDMI-A-1 model: LG (GoldStar) FULL HD res: 1920x1080 hz: 60
dpi: 102 diag: 551mm (21.7")
API: EGL v: 1.5 platforms:device: 0 drv: iris device: 1 drv: swrast gbm:
drv: iris surfaceless:drv: iris wayland:drv: iris x11:drv: iris
API: OpenGL v: 4.6 compat-v: 4.5 vendor: intel mesa v: 26.0.8-1ubuntu0.3
glx-v: 1.4 direct-render: yes renderer: Mesa Intel Graphics (RPL-P)
device-ID: 8086:a7a8 display-ID: :0.0
API: Vulkan v: 1.4.341 surfaces: N/A device: 0 type: integrated-gpu
driver: mesa intel device-ID: 8086:a7a8 device: 1 type: cpu
driver: mesa llvmpipe device-ID: 10005:0000
Info:Tools:api: clinfo, eglinfo, glxinfo, vulkaninfo
de: kscreen-console,kscreen-doctor wl: wayland-info x11: xdriinfo,
xdpyinfo, xprop, xrandr

[LuigiCalamar]

Note: I've also suspected that transcoding in Plex can lead to shutdowns... so what I've been trying is to turn off all apps outside Plexmediaserver if I'm not on thecomputer... especially Firefox... thought that was helping but then - that hardware acceleration got turned back on.

Note: I've also turned off all plexmediaserver transcoding options that use hardeware acceleration...

I know I'm asking a lot on this small mini-PC... but I would expect slowdowns - not crashes...

[cookiemuncher]

Most problems I have ever had with PC's rebooting was due to faulty RAM sticks. Do a complete test (without interruption) to make sure. I used memtest86+

[die.boer]

I turned off firefox hardware acceleration.... I did turn that off a couple days ago but I'd since done a firefox reset - and it was turned back on... should have caught that... so going forward I'll leave it off.

Here is the output from the inxi -Gxx :

Graphics:
Device-1: Intel Raptor Lake-P [UHD Graphics] driver: i915 v: kernel arch: Xe
ports:active: HDMI-A-1 empty: DP-1, DP-2, DP-3, HDMI-A-2, HDMI-A-3,
HDMI-A-4 bus-ID: 00:02.0 chip-ID: 8086:a7a8
Display: wayland server: X.org v: 1.21.1.22 with: Xwayland v: 24.1.10
compositor: kwin_wayland driver:gpu: i915 display-ID: 0
Monitor-1: HDMI-A-1 model: LG (GoldStar) FULL HD res: 1920x1080 hz: 60
dpi: 102 diag: 551mm (21.7")
API: EGL v: 1.5 platforms:device: 0 drv: iris device: 1 drv: swrast gbm:
drv: iris surfaceless:drv: iris wayland:drv: iris x11:drv: iris
API: OpenGL v: 4.6 compat-v: 4.5 vendor: intel mesa v: 26.0.8-1ubuntu0.3
glx-v: 1.4 direct-render: yes renderer: Mesa Intel Graphics (RPL-P)
device-ID: 8086:a7a8 display-ID: :0.0
API: Vulkan v: 1.4.341 surfaces: N/A device: 0 type: integrated-gpu
driver: mesa intel device-ID: 8086:a7a8 device: 1 type: cpu
driver: mesa llvmpipe device-ID: 10005:0000
Info:Tools:api: clinfo, eglinfo, glxinfo, vulkaninfo
de: kscreen-console,kscreen-doctor wl: wayland-info x11: xdriinfo,
xdpyinfo, xprop, xrandr

ok everthing looks ok there , what i would suggest is :
-Disable Firefox hardware acceleration and see if crashes stop.
-Try Plasma X11 for a while, leaving everything else unchanged.

we want to see if Firefox GPU acceleration or Wayland makes the problem disappear.

btw...is the LG display directly connected via HDMI , connected through an HDMI switch/receiver/adapter , occasionally going into standby or being turned on/off while the PC remains running?


if you don't see a x11 option at login screen , you can install it with

sudo apt install plasma-session-x11

, then log out / in ,choosing x11.

[LuigiCalamar]

the lg display is directly connected by hdmi cable will update after trying out x11