Constant Random Crashing. [SOLVED]

redluci4 · 2025-07-06 19:16:40

Edit 1:
looks like this might be a ram issue for now, marked as solved till further diagnosis.

Hello,

This post is to whom ever this may be of concern.

My system became unstable after some updates (I'm NOT saying the update IS causing the crashes, I suspect there might be some hardware issues), and from then on I've wiped and installed arch multiple times, without any improvement, so I just wanted to take a last effort before buying new hardware.

This is my current boot log :

~~jpst.it/4r8jV~~
http://0x0.st/8Gu6.txt

This is my -1 boot log :

~~jpst.it/4r8ki~~
http://0x0.st/8Guk.txt

This is my -2 boot log :

http://0x0.st/8Gu5.txt

This is my -3 boot log (This is a 1GB text file, so I compressed it with `tar -czf text.gz text.log` to 22MB):

http://0x0.st/8GS-.gz

And all the previous boot logs don't exist, idkw.

I'm currently using an Arch install, but due to instability I temporarily use Manjaro live USB for work (but FYI : I sometimes get similar browser crash in that too, but not relevant for this post).

Almost everything is crashing when I'm using my system, Browser, Games, Terminal, kded, baloo, anything from plasma kde, sometimes post install hook while using pacman, sometimes mkinitcpio will crash at the end, sometimes my gpg keys give error - till I run update-ca-trust multiple times, etc. and all this happens on my fresh installation.

I don't know how else to get information about the crashes when it's happening, I just assume my system logs them for future use, but I'm also looking into monitoring the com1 port from another system using a console to usb adapter to log almost everything, not relying on a compromised system for debugging, but for now this is all I have (if you need to know anything else just tell me the command, and I'll report back with updates of the stdout and stderr).

Any help will be appreciated.

Thank you.

Last edited by redluci4 (2025-07-08 23:28:50)

John-Something-Something · 2025-07-06 19:25:04

Hi, i am a newbie, so do not do anything until more experienced guys confirm or disprove any mistake of mine. But your logs are reporting constant problems while loading files and accessing directories. Your SSD may be dying. My one started to be unstable and killed my sistem multiple times during usage, until it just stopped working completely.
The following fatal errors are very important for this conclusion. They are showing that your ssd is not answering some input/output(IO) calls, and corrupted files where found, including the journal itself, other errors while loading firmware also help to conclude the same

Jul 06 19:51:41 ArchAMD kernel: nvme nvme0: missing or invalid SUBNQN field.
Jul 06 19:51:41 ArchAMD kernel: nvme 0000:01:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0011 address=0xfedfc000 flags=0x0050]
Jul 06 19:51:41 ArchAMD kernel: nvme 0000:01:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0011 address=0xfedfc080 flags=0x0050]
Jul 06 19:51:41 ArchAMD systemd-journald[437]: File /var/log/journal/5d94448a5f50465e8005e91359997023/system.journal corrupted or uncleanly shut down, renaming and replacing.

Sorry for my bad english.

Last edited by John-Something-Something (2025-07-06 19:31:48)

redluci4 · 2025-07-06 19:46:58

Thanks bro, I'll do some SSD test with smartmontools on my nvme.

seth · 2025-07-06 21:46:17

(if you need to know anything else just tell me the command, and I'll report back with updates of the stdout and stderr)

Please post your complete system journal for the boot:

sudo journalctl -b | curl -F 'file=@-' 0x0.st

- this will upload the complete system journal of the current boot to 0x0.st and hand you a link.
Please do not use this shitty justpaste.it service that offers me line-wrapped html or line-wrapped pdf and turns the journal into an unreadable mess.

redluci4 · 2025-07-06 22:08:10

seth wrote:

Please post your complete system journal for the boot:
sudo journalctl -b | curl -F 'file=@-' 0x0.st
- this will upload the complete system journal of the current boot to 0x0.st and hand you a link.

Well god-damn, you are a legend, I've been searching for tools like this, since I couldn't get a raw page by putting raw in the URL, I thought of making gist in GitHub for this purpose, but this is far better, even better I can self-host this server.

~~Well, I'll be sure to update~~ My links are updated now with 0x0.st.

Last edited by redluci4 (2025-07-06 22:41:03)

seth · 2025-07-07 07:17:29

So there's possibly something wrong with this disk

Jul 07 02:25:56 ArchAMD kernel: sd 6:0:0:0: [sdc] 1953525168 512-byte logical blocks: (1.00 TB/932 GiB)
Jul 07 02:25:56 ArchAMD kernel: sd 6:0:0:0: [sdc] Write Protect is off
Jul 07 02:25:56 ArchAMD kernel: sd 6:0:0:0: [sdc] Mode Sense: 37 00 00 08
Jul 07 02:25:56 ArchAMD kernel: sd 6:0:0:0: [sdc] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
Jul 07 02:25:56 ArchAMD kernel: sd 6:0:0:0: [sdc] Preferred minimum I/O size 512 bytes
Jul 07 02:25:56 ArchAMD kernel: sd 6:0:0:0: [sdc] Optimal transfer size 33553920 bytes
Jul 07 02:25:56 ArchAMD kernel:  sdc: sdc1
Jul 07 02:25:56 ArchAMD kernel: sd 6:0:0:0: [sdc] Attached SCSI disk
Jul 07 02:25:57 ArchAMD kernel: EXT4-fs (sdc1): mounted filesystem e86edadf-0a3e-4ef6-8f29-47302d00b9fc r/w with ordered data mode. Quota mode: none.
Jul 07 02:26:13 ArchAMD udisksd[1228]: Error probing device: Error sending ATA command IDENTIFY DEVICE to '/dev/sdc': Unexpected sense data returned:

which seems to be /home/luci4/Grano1T
But the last the tow journals are also full of bad page/memory errors. The 3rd has the sense error, but no vma/bad page errors and only one crash in brave
The last journal is just errors about a misconfigured swap - its size is the reason why there're not more journals.

=> memtest86+, "over night" (16+ h) or unitl you hit errors.
Also try to downgrade the kernel and avoid mounting the troublesome disk for the time being.

Off topic PSA: brave faces heavy criticism in recent years, you might want to inquire the situation and consider your options.

redluci4 · 2025-07-07 19:38:56

Shit bro, I know how this looks, but I know what I say when mkinitcpio and pacman seg faulted (it's probably shared libraries than the starting binary), but I wish I didn't lose my old logs.

You know what, I'll put a hold on this until I get the reasonable crash logged, until then I'll just update my normal log in the initial post.

This is so sad, right here a couple of months age I was flexing on everyone running every OS on my PC with just one GPU passing through and splitting with a minimum of 2 VM OS and 1 Hypervisor thinking myself to be a power user, now look at me, I'm still a noob at heart LOL.

Anyway it's been fun debugging and reinstalling my Arch, I'll go figure out how to use my com0 port on my motherboard to bypass my display (The funny thing is, I'm using an old ass Dell laptop to debug my PC with the exact same software - Arch, plasma, Brave, etc. but without none of the bugs, makes me think what happened to the good old days).

So thanks for the help, I'll be back if I get useful info or somehow fix this problem.

Thanks guys.

redluci4 · 2025-07-07 19:53:40

seth wrote:

Off topic PSA: brave faces heavy criticism in recent years, you might want to inquire the situation and consider your options.

Brave and every other browser out there, Firefox, Chromium, LibreWolf, WaterFox, Zen, Tor, Min, etc. are crashing just the same or worse, I'm just using brave because it lasts the longest before crashing, and I've also started using curl to download files in my terminal for that reason too.

But thanks regardless, I'll definitely look into other means of browsing the web.

seth · 2025-07-07 19:55:04

seth wrote:

But the last the tow journals are also full of bad page/memory errors. The 3rd has the sense error, but no vma/bad page errors and only one crash in brave
=> memtest86+, "over night" (16+ h) or until you hit errors.

What you're shown so far has all the hallmarks of corrupt RAM.

but seth also wrote:

Also try to downgrade the kernel and avoid mounting the troublesome disk for the time being.

Edit: brave isn't facing those criticisms because of stability issues, it's more about shady business practices esp. considering their privacy protection claims.

Last edited by seth (2025-07-07 19:56:35)

redluci4 · 2025-07-08 22:39:43

seth wrote:

What you're shown so far has all the hallmarks of corrupt RAM.

Can I ask you about what are the hallmark that gave you signs that my RAM is corrupting?

Because you might be right, I unplugged the ram sticks then did a quick clean and slot swap then did the memtest86+ for about 23 hours and no error so far, now after rebooting the system, I haven't had any browser crash except for a single gpu memory buffer problem:
memtest

Thank you very much btw, I'm going to look for new ram, or diagnose the ones I already have.

seth · 2025-07-08 22:54:26

seth wrote:

But the last the two journals are also full of bad page/memory errors.

It's either the RAM or a very exotic kernel bug nobody else is facing.
You btw. generally might have to run the RAM at lower frequencies or higher timings, they're not only not the same batch, they're completely different vendors.
Running them full speed is likely prone to cause errors.

redluci4 · 2025-07-08 23:25:50

seth wrote:

You btw. generally might have to run the RAM at lower frequencies or higher timings, they're not only not the same batch, they're completely different vendors.

I know it's an uncommon ram setup and how it looks like from a pros POV, I also get what I have to do, it's just that I'm unable to accept it could be a problem after it worked for 5+ years, and done all the loony-bins I've asked it to run and now of all time decides to give up.

But thanks for all the info, tips, solutions, and everything else too.

I'll mark this as solved for now, until I get something to diagnose my pc parts.

seth · 2025-07-09 05:55:59

People have RAM and then later on get more - that's not uncommon at all.
The point is to understand that that this frequently leads to RAM errors when attempting to run both DIMMs at their maximum performance.
Also

and slot swap

because it looks like maybe only *now* you're using both channels?

redluci4 · 2025-07-10 01:58:27

seth wrote:

People have RAM and then later on get more - that's not uncommon at all.
The point is to understand that that this frequently leads to RAM errors when attempting to run both DIMMs at their maximum performance.

That is true, I'll have to look into rams batches with better timing to run them at high freq.

Also
and slot swap
because it looks like maybe only *now* you're using both channels?

Although I might still be a noob, I'm not that Newby of a noob, I simply swapped the second and fourth slots, my idea was that if my ram had a fault, then I'll change the ram order to make it shows up early by having the faulty ram closer to the CPU (not physically, but logically like Slot-2, then Slot-4, then Slot 1 and 3 right.)

And then I ran the test now, then it didn't show any errors. So what every I did (Cleaning my motherboard or swapping the ram) just bought me some time.(Well - I know if I did do the tests before AND after, then I'd have had the data to compare the deltas and come closer to the fault, but now I don't want to poke around because what's working is just working).

But I do appreciate the concern, rest assured I'm gonna go do some ram hunting.

Arch Linux

#1 2025-07-06 19:16:40

Constant Random Crashing. [SOLVED]

#2 2025-07-06 19:25:04

Re: Constant Random Crashing. [SOLVED]

#3 2025-07-06 19:46:58

Re: Constant Random Crashing. [SOLVED]

#4 2025-07-06 21:46:17

Re: Constant Random Crashing. [SOLVED]

#5 2025-07-06 22:08:10

Re: Constant Random Crashing. [SOLVED]

#6 2025-07-07 07:17:29

Re: Constant Random Crashing. [SOLVED]

#7 2025-07-07 19:38:56

Re: Constant Random Crashing. [SOLVED]

#8 2025-07-07 19:53:40

Re: Constant Random Crashing. [SOLVED]

#9 2025-07-07 19:55:04

Re: Constant Random Crashing. [SOLVED]

#10 2025-07-08 22:39:43

Re: Constant Random Crashing. [SOLVED]

#11 2025-07-08 22:54:26

Re: Constant Random Crashing. [SOLVED]

#12 2025-07-08 23:25:50

Re: Constant Random Crashing. [SOLVED]

#13 2025-07-09 05:55:59

Re: Constant Random Crashing. [SOLVED]

#14 2025-07-10 01:58:27

Re: Constant Random Crashing. [SOLVED]

Board footer