PC shuts down for no apparent reason, then reboots

English Install/Boot/Login
1

opensuse Tumbleweed 20261013
KDE Plasma 6.75
Kernal 7.2.9-1-default (64-bit)
Graphics platform X11
Harddrives partitioned EXT4
Asus Z390-A m/board [new], CPU Intel i7-9700K, 32 gb Ram, integrated graphics, PSU 650 W [new]

For a few weeks now I am experiencing my PC shutting down.
This morning I was typing an email and all of a sudden my screen went black and then it went into reboot mode.
There does not seem to be any regularity to it. It shuts down whether I am using it or it is idle.
I have checked my connections of the PC - all is good. External and internal.
I have replaced my APC surge protector.
I have had my 110 circuit tested. No problem there. Also I do have another computer connected to this circuit and it stays up and running while this PC shuts down.
The PC in question always restarts with no problem. I keep my OS up to date, doing zypper dup regularly. The restarting, I think, is the way I have it setup in the bios to react when a power interruption occurs. It always restarts after.

Is there some log or journal that could tell if this is a software or hardware problem. Or at least get me on the track to find this out?

2

journalctl -b -1 will print out a log for the previous boot. Check the end of the log to see what is happening before the shutdown/reboot. If you want to share it here, use a paste service like https://paste.opensuse.org/ and be careful about possibly personally identifying information in the logs.

It sounds more likely to be a hardware issue than software, though.

3

Add the in reverse option so starts at the final journal entry… journalctl -b -1 -r

But as indicated by @ikanotheokara likely a hardware issue, over heating cpu, faulty RAM etc.

1 Like
4

Maybe it’s just me, but -r always does my head in a bit because my brain only wants to read top-to-bottom as older-to-newer. It’s easier for me to just tap End to go to the end of the log then scroll up until I find what I need.

@LaQuirrELL faulty RAM would be my first guess because it’s easy to troubleshoot. Assuming your 32 gigs are two 16 gig sticks, try running with just one at a time to see if the random reboots stop.

You say the motherboard and PSU are new. How new? Did the rebooting problem begin after installing them? I’d first check CPU temps to see if it is an overheat. If that is ruled out I’d suspect the motherboard or PSU, but they aren’t as easy to diagnose.

5

Thank you @ikanotheokara and @malcolmlewis

The m/board and the PSU are approx 6 months old.
The CPU and RAM are 8 years old.
In checking the CPU temps would I go into the UEFI to see or is there a way from command line?
I know there are settings to adjust fan speeds and performance types in the UEFI.

6

I had a similar issue on my system. Power supply was bad. It finally quit one day. Luckily i have a Microcenter nearby.

1 Like
7

You can check sensor data directly in Plasma with a plasmoid/widget.

If you would prefer to use the terminal, you can use sensors to print out all available sensor data, but the plasmoid is easier-to-read and updates in real-time.

You can also add sensor info to System Monitor, but you will have to do a little DIY: Make a new page with +Add new page and then click Edit Page in the top right to configure your own dashboard with the sensors you want to see.

8

This is the output I get when I do “sensors” from a terminal:

dabud@da:~> sensors
acpitz-acpi-0
Adapter: ACPI interface
temp1:        +27.8°C  

nvme-pci-0400
Adapter: PCI adapter
Composite:    +36.9°C  (low  =  -5.2°C, high = +79.8°C)
                       (crit = +84.8°C)

coretemp-isa-0000
Adapter: ISA adapter
Package id 0:  +29.0°C  (high = +86.0°C, crit = +100.0°C)
Core 0:        +27.0°C  (high = +86.0°C, crit = +100.0°C)
Core 1:        +27.0°C  (high = +86.0°C, crit = +100.0°C)
Core 2:        +27.0°C  (high = +86.0°C, crit = +100.0°C)
Core 3:        +27.0°C  (high = +86.0°C, crit = +100.0°C)
Core 4:        +27.0°C  (high = +86.0°C, crit = +100.0°C)
Core 5:        +26.0°C  (high = +86.0°C, crit = +100.0°C)
Core 6:        +28.0°C  (high = +86.0°C, crit = +100.0°C)
Core 7:        +25.0°C  (high = +86.0°C, crit = +100.0°C)

nvme-pci-0200
Adapter: PCI adapter
Composite:    +26.9°C  (low  =  -5.2°C, high = +83.8°C)
                       (crit = +87.8°C)

dabud@da:~>

The temps seem OK to me. Is there any way to capture the temps just before it shuts down? I am using KDE but am not sure if I am in Plasma? How do I tell?
I have removed one of the ram and am running on 16GB.

9

Thermal problem? (-> check sensors)
RAM defect? (-> memtest86)

…argh, I overlooked the other answers here. Mods, please delete this.

10

thank you @shundhammer

11

KDE == Plasma … Plasma == KDE :slight_smile:

Maybe you’re thinking, “is my KDE system running on X11 or Wayland”?

So, let’s re-read a portion of your opening post, quoted here … this should provide both answers :+1: :

I am a bit unclear about “Kernal” … it that how “kernel” is spelled in another language? :slight_smile:

So, one question, which might be difficult to answer:

“What changed” when things were working fine … and then the issue happened??

12

Those temps look pretty cool and comfy, so I don’t think it’s a thermal issue.

If you still suspect thermals, though, you could log it very lazily with something like watch -n 5 "{ date; sensors coretemp-isa-0000; } | tee -a cputemp.log", which will save the following into ~/cputemp.log every 5 seconds (change the value after -n to make it update more or less frequently):

Sat Oct 17 07:57:36 JST 2026
coretemp-isa-0000
Adapter: ISA adapter
Package id 0:  +29.0°C  (high = +86.0°C, crit = +100.0°C)
Core 0:        +27.0°C  (high = +86.0°C, crit = +100.0°C)
Core 1:        +27.0°C  (high = +86.0°C, crit = +100.0°C)
Core 2:        +27.0°C  (high = +86.0°C, crit = +100.0°C)
Core 3:        +27.0°C  (high = +86.0°C, crit = +100.0°C)
Core 4:        +27.0°C  (high = +86.0°C, crit = +100.0°C)
Core 5:        +26.0°C  (high = +86.0°C, crit = +100.0°C)
Core 6:        +28.0°C  (high = +86.0°C, crit = +100.0°C)
Core 7:        +25.0°C  (high = +86.0°C, crit = +100.0°C)

Leave it running in a terminal while you use your computer. If you want to end it early, you can press Q to end the watch process.

Be warned: This file will be growing by 13 lines every n seconds. (9,360 lines every hour at 5 seconds) It will get comically long and unwieldy.

If you are worried about it getting too long, you could use something like: watch -n 5 "{ date; sensors coretemp-isa-0000; } | (tee -a ~/cputemp.log; tail -n 1560 ~/cputemp.log > ~/cputemp.log.tmp && mv ~/cputemp.log.tmp ~/cputemp.log)" which would limit the log to 1560 lines, saving only the most recent 10 minutes worth of readings (assuming a reading every 5 seconds).

You could also use sensors -j with jq to make an easier to read log with JSON, but the watch and tee method is good enough for basic troubleshooting.

Still, my money is on it being RAM, PSU or motherboard, in that order.

13

To the OP if it’s ram, psu, cpu or mobo.
The mobo has a light indicator to show the faulty one.
Just read your mobo manual you will find what I’m suggesting.

1 Like
14

@myswtest
Thank you for clearing up that PLASMA thing. Yes I was confused about Wayland and X11.

Regarding “Kernal” - that must be a typo, never noticed it before. “Kernel” is also the way I know it. I guess someone should point that out to KDE?

PC wise nothing has changed in a long while. Replaced M/board, PSU and power switch six months ago. [Whole other story, but m/board was not a problem]. Since then, PC has been working great. A few weeks ago it began this shutdown thing, which happens without warning unrelated to whatever I am doing.

15

@ikanotheokara
Thank you for the code for getting a temp log. I like the idea that I could save only, say, 1500 lines.

Currently, I have removed one ram and as yet, PC remains running, although typically it may run for 3 to 5 days before a shutdown.

I have been in contact with Kingston regarding the ram [ram is HyperX predator /32] warranty and they want me to run memtest86. I have downloaded it to a USB, but want to finish your suggestions 1st.
I can’t remember, but doesn’t opensuse have memtest86 available when you startup and choose advanced?

16

@conram
Thank you.
The Q-LED indicator on the M/board is what you are refering to? As the PC is running there are no lights on. The next time I have to reboot I will watch the lights to see if there is any indication of a problem.

17

@myswtest further to “kernal” I went back to my info page and it is typed as “kernel”. Apparently it was my typo. :grimacing: i think i am breaking the rule of: “Never admit to anything!” :grin:

@conram the QED indicator lights on startup flash - red then yellow, pause 1 second, white then green, green stays on for 2 seconds and goes out. No lights are on or flashing. It doesn’t appear there is a problem, but I need to find my manual to check.

18

It’s normal for the lights to turn on briefly at boot while the motherboard is checking the status of CPU (red), RAM (yellow), GPU (white) and boot device (green). If they turn off after that, it means the motherboard didn’t detect any issues.

Mind, that doesn’t mean there are no issues, it just means the motherboard didn’t find anything wrong. The motherboard performs only a very basic check. memtest86+ as suggested by @shundhammer and Kingston is much more thorough and definitely worth a try. Run it overnight/while you’re at work, because it takes a long time to complete.

19

Any update?

20

I have run the memtestplus and memtest86 on both ram sticks. I tested the ram sticks separately. Both tests produced 0 errors. As a side note: The core temps reached as high as 46C during the tests. The PC has not had a shut down since I started this post. 10.15.2026
I would cross my fingers, but since it was happening before the tests I am sure it will happen again. So, maybe CPU. I will set up the terminal CPU temp recordings as you have suggested earlier.
NOTE: I thought I had hit the ‘post’ button yesterday. Apparently I had not. My apologies.