Frequent crashes in my arch linux

har936 · 2026-04-20 11:40:04

When i was using my arch (gnome), suddenly firefox crashed, then all the icons disappeared, it looked broken, then it will go to the loading linux linux and loading initial ramdisk will come as if it is booting, then it will come Failed to start Generate shutdown-ramfs, then i had to long press my power button in my lap to kill. first actually, i know it is wrong, but i took journalctl logs and pasted in chatgpt to see what is the error it told. it mainly pointed on it being gnome crash error. Again i was in gnome, suddenly it crashed and hit kernel panic with a big qr code with log report. I thought i was doomed.So, i also have hyprland too, so i switched to hyprland, but i kept my laptop for 2 hours in sleep and it still crashed when i opened. when i asked chatgpt parallely, it mainly focused on it being ssd failing. the error was that there was read timeout. I even once hit kernel panic and a big QR code appeared. When i checked my ssd with my service guy, he checked and told the health is 98%. he mentioned the crashes might be reason for the 2%. but i faced crashes in only in arch. In windows it hung for a brief time, didnt crash. Can you guys help to identify is there issue with the linux.

My kernel panic:
Kernel panic

this is output for my journalctl level 3 errors in a boot it crashed:
https://pastebin.com/1bGFrJAp

ps: this is my first post, idk how to post stuff, if the format of posting the kernel panic log and the crash log in a url is wrong, please tell. I hope this format doesn't trigger anyone and is open to help me .

Last edited by har936 (2026-04-20 12:07:25)

cryptearth · 2026-04-20 13:14:16

Apr 15 19:39:31 archlinux kernel: I/O error, dev sda, sector 4096 op 0x0:(READ) flags 0x880700 phys_seg 1 prio class 2
Apr 15 19:39:31 archlinux kernel: device offline error, dev sda, sector 4096 op 0x0:(READ) flags 0x800000 phys_seg 1 prio class 2

your drive is dead - you have to replace it

seth · 2026-04-20 13:19:01

Apr 15 19:39:31 archlinux kernel: I/O error, dev sda, sector 4096 op 0x0:(READ) flags 0x880700 phys_seg 1 prio class 2
Apr 15 19:39:31 archlinux kernel: device offline error, dev sda, sector 4096 op 0x0:(READ) flags 0x800000 phys_seg 1 prio class 2
Apr 15 19:39:31 archlinux kernel: Buffer I/O error on dev sda, logical block 512, async page read

[ 1578.098749] systemd-journald[484]: Failed to vacuum /var/log/journal/0b9e0a9bf6a746e7b72fca340ef6e518, ignoring: Input/output error
[ 1578.098793] EXT4-fs warning (device nvme0n1p8): dx_probe:791: inode #788561: lblock 0: comm systemd-journal: error -5 reading directory block
[ 1578.127375] fbcon: Taking over console
[ 1578.134085] Console: switching to colour frame buffer device 240x67
[ 1578.141248] wlan0: deauthenticating from a8:da:0c:61:74:81 by local choice (Reason: 3=DEAUTH_LEAVING)
[ 1578.141766] coredump: 1872(wpa_supplicant): |/usr/lib/systemd/systemd-coredump pipe failed
[ 1578.152946] coredump: 2678(udisksd): |/usr/lib/systemd/systemd-coredump pipe failed
[ 1579.141710] coredump: 696(iwd): |/usr/lib/systemd/systemd-coredump pipe failed
[ 1579.142983] systemd-journald[484]: Failed to write entry to /var/log/journal/0b9e0a9bf6a746e7b72fca340ef6e518/system.journal (21 items, 515 bytes) despite vacuuming, ignoring: Input/output error (Dropped 120 similar message(s))
[ 1579.314434] systemd[1]: segfault at 7ffce557aff8 ip 00007f5b5029cdc8 sp 00007ffce557b000 error 6 in libsystemd-shared-260.1-1.so[29cdc8,7f5b50075000+32a000] likely on CPU 4 (core 8, socket 0)
[ 1579.315928] Code: 1a 00 ba 38 00 00 00 48 8d 35 0f ab 13 00 48 8d 3d 8b 46 17 00 48 89 e5 ff 15 e4 fc 29 00 0f 1f 40 00 f3 0f 1e fa 55 48 89 e5 <53> 48 83 ec 18 48 85 ff 0f 84 c2 01 00 00 48 8b 42 28 49 89 f1 49
[ 1579.316220] coredump: 1(systemd): |/usr/lib/systemd/systemd-coredump pipe failed
[ 1579.316224] Kernel panic - not syncing: Attempted to kill init! exitcode=0x0000000b

Please post your complete system journal for the boot:

sudo journalctl -b -1 | curl -s -H "Accept: application/json, */*" --upload-file - 'https://paste.c-net.org/'

my ssd with my service guy, he checked and told the health is 98%

And also "smartctl -a /dev/sda" and "nvme smart-log /dev/nvme0"

Is there a parallel windows installation?

Edit: @cryptearth, don't underestimate cable/bus errors

Last edited by seth (2026-04-20 13:20:19)

har936 · 2026-04-20 13:36:27

cryptearth wrote:

Apr 15 19:39:31 archlinux kernel: I/O error, dev sda, sector 4096 op 0x0:(READ) flags 0x880700 phys_seg 1 prio class 2
Apr 15 19:39:31 archlinux kernel: device offline error, dev sda, sector 4096 op 0x0:(READ) flags 0x800000 phys_seg 1 prio class 2

your drive is dead - you have to replace it

But, it is working fine now
Like it is been 5 days since i had a crash....

seth · 2026-04-20 13:42:49

In that case you'll likely have to increase the boot offset (-1 is the previous boot)

har936 · 2026-04-20 14:07:20

seth wrote:

Apr 15 19:39:31 archlinux kernel: I/O error, dev sda, sector 4096 op 0x0:(READ) flags 0x880700 phys_seg 1 prio class 2
Apr 15 19:39:31 archlinux kernel: device offline error, dev sda, sector 4096 op 0x0:(READ) flags 0x800000 phys_seg 1 prio class 2
Apr 15 19:39:31 archlinux kernel: Buffer I/O error on dev sda, logical block 512, async page read

[ 1578.098749] systemd-journald[484]: Failed to vacuum /var/log/journal/0b9e0a9bf6a746e7b72fca340ef6e518, ignoring: Input/output error
[ 1578.098793] EXT4-fs warning (device nvme0n1p8): dx_probe:791: inode #788561: lblock 0: comm systemd-journal: error -5 reading directory block
[ 1578.127375] fbcon: Taking over console
[ 1578.134085] Console: switching to colour frame buffer device 240x67
[ 1578.141248] wlan0: deauthenticating from a8:da:0c:61:74:81 by local choice (Reason: 3=DEAUTH_LEAVING)
[ 1578.141766] coredump: 1872(wpa_supplicant): |/usr/lib/systemd/systemd-coredump pipe failed
[ 1578.152946] coredump: 2678(udisksd): |/usr/lib/systemd/systemd-coredump pipe failed
[ 1579.141710] coredump: 696(iwd): |/usr/lib/systemd/systemd-coredump pipe failed
[ 1579.142983] systemd-journald[484]: Failed to write entry to /var/log/journal/0b9e0a9bf6a746e7b72fca340ef6e518/system.journal (21 items, 515 bytes) despite vacuuming, ignoring: Input/output error (Dropped 120 similar message(s))
[ 1579.314434] systemd[1]: segfault at 7ffce557aff8 ip 00007f5b5029cdc8 sp 00007ffce557b000 error 6 in libsystemd-shared-260.1-1.so[29cdc8,7f5b50075000+32a000] likely on CPU 4 (core 8, socket 0)
[ 1579.315928] Code: 1a 00 ba 38 00 00 00 48 8d 35 0f ab 13 00 48 8d 3d 8b 46 17 00 48 89 e5 ff 15 e4 fc 29 00 0f 1f 40 00 f3 0f 1e fa 55 48 89 e5 <53> 48 83 ec 18 48 85 ff 0f 84 c2 01 00 00 48 8b 42 28 49 89 f1 49
[ 1579.316220] coredump: 1(systemd): |/usr/lib/systemd/systemd-coredump pipe failed
[ 1579.316224] Kernel panic - not syncing: Attempted to kill init! exitcode=0x0000000b

Please post your complete system journal for the boot:

sudo journalctl -b -1 | curl -s -H "Accept: application/json, */*" --upload-file - 'https://paste.c-net.org/'

my ssd with my service guy, he checked and told the health is 98%

And also "smartctl -a /dev/sda" and "nvme smart-log /dev/nvme0"

Is there a parallel windows installation?

Edit: @cryptearth, don't underestimate cable/bus errors

My entire system boot: (thanks for that offset info, i searched up all the boots and found it is 23rd one. this is the full journal of the crashed one)
complete system journal for the boot

This is the output for "sudo smartctl -a /dev/sda":
Output

But i dont have sda, i have only nvme0n1.
So i did: "sudo smartctl -a /dev/nvme0n1". Its output is:
https://paste.c-net.org/GoodmanLucked

output for "sudo nvme smart-log /dev/nvme0n1" is:
https://paste.c-net.org/HarpyRange

edit: also yeah, i have dual booted arch and windows.

The above one is normal crash one i think, because it doesnt have proper end (Journal didnt stop last line)
this below full journal is the kernel panic crash one i think. (i had multiple crashes...)
https://paste.c-net.org/McintyreLifeboat

also i have MSI Thin 15 B13VE laptop, not a desktop..

Last edited by har936 (2026-04-20 14:40:58)

cryptearth · 2026-04-20 14:12:15

seth wrote:

don't underestimate cable/bus errors

agreed - based upon many issues over at github/zfs about failing pools which turned out to be caused by bad cables / bad power ... but would be unusual for nvme ... unless one uses a janky usb adapter ... or flimsy flatflex and raspi like Jeff Geerling

seth · 2026-04-20 15:28:57

But i dont have sda, i have only nvme0n1.

Apr 15 19:39:31 archlinux kernel: Buffer I/O error on dev sda, logical block 512, async page read

You had…

edit: also yeah, i have dual booted arch and windows.

3rd link below. Mandatory.
Disable it (it's NOT the BIOS setting!) and reboot windows and linux twice for voodo reasons.

If that's not it and

1987:5021  Phison Electronics Corporation PS5021-E21 PCIe4 NVMe Controller (DRAM-less)
…
Apr 15 19:25:39 archlinux kernel: pci 10000:e1:00.0: can't override BIOS ASPM; OS doesn't have ASPM control

about the nvme, see https://wiki.archlinux.org/title/Solid_ … leshooting
There isn't gonna be any journal entry about the crash after a kernel panic - does it typically relate to sleeping/hibernating the device?

har936 · 2026-04-21 08:34:12

seth wrote:

But i dont have sda, i have only nvme0n1.
Apr 15 19:39:31 archlinux kernel: Buffer I/O error on dev sda, logical block 512, async page read
You had…
edit: also yeah, i have dual booted arch and windows.
3rd link below. Mandatory.
Disable it (it's NOT the BIOS setting!) and reboot windows and linux twice for voodo reasons.
If that's not it and
1987:5021  Phison Electronics Corporation PS5021-E21 PCIe4 NVMe Controller (DRAM-less)
…
Apr 15 19:25:39 archlinux kernel: pci 10000:e1:00.0: can't override BIOS ASPM; OS doesn't have ASPM control
about the nvme, see https://wiki.archlinux.org/title/Solid_ … leshooting
There isn't gonna be any journal entry about the crash after a kernel panic - does it typically relate to sleeping/hibernating the device?

That time of that crash i think i had my external hdd connected. but i dont think that is the reason, it is perfectly working hdd, and crashes happened even without it. i think it just detected the external hdd.

And i disabled the windows startup as instructed in wiki(including hybernation ig, in the arch wiki, it was recommended both to be turned off).

for the NVME troubleshooting, setting nvme_core.default_ps_max_latency_us=0 , i am worried about my battery health and heating issues. do i have to set this? Also, it has been a week since i had a crash too.. should i disable low power mode for my ssd?

And yeahh, it mainly happened during after i left the laptop for sleep for some time and turned it back on.

seth · 2026-04-21 15:43:27

APST is particularly an issue wrt S3/s2idle - but so is a hibernating windows.
If the situation doesn't stabilize and you get more nvme related issues you should™ absolutely test to disable it as well as the IOMMU ("iommu=soft")
It will consume more battery while idling IO (you can check this w/ powertop - claims are 2-3W) but the system won't overheat (unless it's extremely poorly designed) and the immediate goal is to figure whether this will stabilize the system.

har936 · 2026-04-22 08:14:32

seth wrote:

APST is particularly an issue wrt S3/s2idle - but so is a hibernating windows.
If the situation doesn't stabilize and you get more nvme related issues you should™ absolutely test to disable it as well as the IOMMU ("iommu=soft")
It will consume more battery while idling IO (you can check this w/ powertop - claims are 2-3W) but the system won't overheat (unless it's extremely poorly designed) and the immediate goal is to figure whether this will stabilize the system.

for now i haven’t added the nvme power fix, since things seem stable. i’ll just keep using it and monitor — if the crashes come back, i’ll try disabling apst and see if that fixes it.
thanks for the help :)

har936 · 2026-05-08 11:37:22

So this many days i didnt have any issue. Now all of a sudden, what happened again is, i was using firefox, suddenly firefox closed, all the gnome icons disappeared, then boot screen that "loading linux linux
loading initial ramdisk" came, then [FAILED] Failed to start Generate shutdown-ramfs. i was trying to hold down my powerbutton to shutdown.
Then it hit linux kernel panic.
This is the complete boot: https://paste.c-net.org/TalentsLapel

This is the kernel panic report: https://paste.c-net.org/MailerPlatt
what should i do..

cryptearth · 2026-05-08 11:40:18

replace the drive - it's dead, Jim

har936 · 2026-05-08 14:50:02

cryptearth wrote:

replace the drive - it's dead, Jim

It is not dead.. it worked this many days perfectly. I don't want to conclude that my drive is dead

5hridhyan · 2026-05-08 15:18:11

har936 wrote:

for now i haven’t added the nvme power fix, since things seem stable. i’ll just keep using it and monitor — if the crashes come back, i’ll try disabling apst and see if that fixes it.

May 08 15:56:32 archlinux kernel: Command line: BOOT_IMAGE=/vmlinuz-linux root=UUID=7b8744cf-4b77-4dc8-a91f-97e44794e71d rw loglevel=3 quiet

what should i do.

seth wrote:

the immediate goal is to figure whether this will stabilize the system.

first, try adding

nvme_core.default_ps_max_latency_us=0 iommu=soft

as Kernel Parameters, to rule out NVMe power state issues.

if the problem persists, you can additionally test w/

pcie_aspm=off pcie_port_pm=off

as the wiki said.
And/or nvme.noacpi=1

If it's actually failing, back up your data w/o wasting time...

Last edited by 5hridhyan (2026-05-08 15:55:11)

cryptearth · 2026-05-08 15:57:45

har936 wrote:

cryptearth wrote:
replace the drive - it's dead, Jim
It is not dead.. it worked this many days perfectly. I don't want to conclude that my drive is dead

test your back then or prepare for data loss
the drive is failing - it's only a matter of when, not if

seth · 2026-05-08 16:05:40

5hridhyan wrote:

If it's actually failing, back up your data w/o wasting time...

The drive isn't operating reliably, whatever the cause, you cannot trust it and HAVE to have backups!

it's only a matter of when, not if

"Yesterday" - the remaining question is "why", next to bogus APST it could also be environmental (termperature/humidity, badly seated, tension, …) but

It is not dead.. it worked this many days perfectly.

Things work and then they're broken - the circle of life.
Nobody wants to see their drive broken, there's some emotional element of uncertainty if your storage betrays you. I get that.
But data loss weighs heavier, so whatever you decide to do: if you don't already have a backup strategy, you need one NOW.

har936 · Yesterday 09:41:07

yeah i understand, i took proper backup of my stuff. (The thing i worry about the hardisk failing is all the things i have set up like the arch setup itself, hyprland configuraiton, windows config, settings, etc., )
Anyway failing of hardisk means, i would have to change that, financially also, with the prices, i want to try to save the disk.

Today too, i hit kernel panic. again it crashed.
But when i was using windows for a while, it didnt seem crashing or anything. i only had crashes with arch. (i dont have enough evidence too, like long time usage of windows or multiple boots of windows. if needed i can try do it, but all my stuff are in arch, like what i am working on, have to migrate all to there).
My doubt is, i feel like windows is fine and arch is crashing a lot. can i, or is there anyway to like, point the crashes to arch itself? kind of like to link the issue with arch and not disk. Windows seems pretty fine to me.

Meanwhile, I will try adding
nvme_core.default_ps_max_latency_us=0 iommu=soft

latest crash log if needed:

This is the complete boot: https://paste.c-net.org/DeceitVacate
This is the kernel panic report: https://paste.c-net.org/FamilyTaking

Arch Linux

#1 2026-04-20 11:40:04

Frequent crashes in my arch linux

#2 2026-04-20 13:14:16

Re: Frequent crashes in my arch linux

#3 2026-04-20 13:19:01

Re: Frequent crashes in my arch linux

#4 2026-04-20 13:36:27

Re: Frequent crashes in my arch linux

#5 2026-04-20 13:42:49

Re: Frequent crashes in my arch linux

#6 2026-04-20 14:07:20

Re: Frequent crashes in my arch linux

#7 2026-04-20 14:12:15

Re: Frequent crashes in my arch linux

#8 2026-04-20 15:28:57

Re: Frequent crashes in my arch linux

#9 2026-04-21 08:34:12

Re: Frequent crashes in my arch linux

#10 2026-04-21 15:43:27

Re: Frequent crashes in my arch linux

#11 2026-04-22 08:14:32

Re: Frequent crashes in my arch linux

#12 2026-05-08 11:37:22

Re: Frequent crashes in my arch linux

#13 2026-05-08 11:40:18

Re: Frequent crashes in my arch linux

#14 2026-05-08 14:50:02

Re: Frequent crashes in my arch linux

#15 2026-05-08 15:18:11

Re: Frequent crashes in my arch linux

#16 2026-05-08 15:57:45

Re: Frequent crashes in my arch linux

#17 2026-05-08 16:05:40

Re: Frequent crashes in my arch linux

#18 Yesterday 09:41:07

Re: Frequent crashes in my arch linux

Board footer