Constant Input/Output Error [Solved]

Heretic12 · 2024-11-30 19:34:57

No, I didn't put that nvme back. An also I didn't change any nvme settings and didn't create any RAID arrays...

seth · 2024-11-30 20:11:11

So why's the broken nvme0 showing up there w/ the good old IO errors?
Though interestingly enough, nvme0 is now on 03:00.0. what used to be nvme1 - maybe the indexing changed, but there's certainly something foul about those nvmes.

From the arch iso:

lspci -tvnn

The system installation is probably fine.

Nov 21 17:46:40 HereticPC kernel: DMI: Gigabyte Technology Co., Ltd. Z490 AORUS XTREME WF/Z490 AORUS XTREME WF, BIOS F23 12/20/2023

Did I ask whether there're BIOS updates available?
Can you constrain the CPU clocks from the firmware?

Heretic12 · 2024-11-30 20:52:42

As for nvme numbers - they change all the time, they are not persistent, so no, I didn't put that device back) The BIOS is updated to the latest version already (well, it was in september during the first installation of Arch, but the latest update came out in 22, so I doubt something has change althou I'll check) and yes, I can lower the multiplier.

Yup, the BIOS is of the latest version

It worked, finally managed to boot after lowering the multiplier to 37. Although for some reason I can't type anything with my keyboard

Lspci -tvnn output: http://0x0.st/XRD5.txt

Last edited by Heretic12 (2024-11-30 21:37:30)

seth · 2024-12-01 07:17:19

Although for some reason I can't type anything with my keyboard

Lower the multiplier more.
If this doesn't work, we'll need a journal from that boot.

Ftr, you're not currently overclocking the system?
You also didn't just happen to have forgotton to plug the dedicated 6/8-pin power supply into your GPU?

Heretic12 · 2024-12-01 15:42:47

Ftr, you're not currently overclocking the system?
You also didn't just happen to have forgotton to plug the dedicated 6/8-pin power supply into your GPU?

No, double checket it, XMP is off, multiplier is 20, both 8-pin graphics card cables are properly attached and the card gives led signals that both power cables are connected. All the voltage settings are on auto.
The result is that I boot into getty on TTY1 but when I try to type in the login (or anything) -nothing happens like the keyboard is not even plugged in, although it is, and when I try to unplug it and plug it back - I get system messages on screen about it. Like this:

Arch Linux 6.6.62-1-1ts (tty1)
HereticPC login: [ 417.513160] usb 1-2.1: USB disconnect, device number 4
[ 418.0252111] usb 1-2.2: USB disconnect, device number 6
[ 423.535771] usb 1-2.1: new full-speed USB device number 15 using xhci_hcd
[ 423.627769] usb 1-2.1: New USB device found, idVendor=1532, idproduct=0226, bcdDevice= 2.00
[ 423.6278991] usb 1-2.1: New USB device strings: Mfr=1, Product=2, SerialNumber=3
[ 423.6280021] usb 1-2.1: Product: Razer Huntsman Elite
[ 423.628073] usb 1-2.1: Manufacturer: Razer
[ 423.8624391] usb 1-2.2: new full-speed USB device number 16 using xhci_hcd
[ 423.9540251] usb 1-2.2: New USB device found, idVendor=1532, idProduct=0244, bcdDevice= 2.00
[ 423.954156] usb 1-2.2: New USB device strings: Mfr=1, Product=2, SerialNumber=0
[ 423.9542581] usb 1-2.2: Product: Razer Tartarus Pro
[ 423.954326] usb 1-2.2: Manufacturer: Razer

Last edited by Heretic12 (2024-12-01 15:48:09)

seth · 2024-12-01 21:59:06

Do you have a less fancy keyboard?
Did you try a different usb port?

Heretic12 · 2024-12-02 01:41:25

Doesn't matter. Due to lowering cpu multiplier I've actually managed to reinstall nvidia-dkms and run mkinitcpio and after removing nomodeset I've managed to boot into SDDM. Point is - there is no signal from the mouse either - it's like it does't see any usb devices yet it - as I' shown - definitely does. Hell of a bug... I would conclude tgat it is a hardware issue, close the topic and apologize for taking so much time, but nvmes are OK and there are no reasons to believe that cpu has decided to unexpectedly die...

seth · 2024-12-02 09:05:54

Can you plug and mount usb keys?
Try to remove all detacheable usb devices (internal an external) and only add a $5 office-supply keyboard in case some™ other device is jamming the bus.

Heretic12 · 2024-12-02 16:15:35

No use. I think it's a hard reinstall situation - just a big cluster F - now even usb is not working properly...

UPD
Fixed the problem - remove all new kernel parameters - those that were disabling plymouth, nomodeset and boot delay. So..

Last edited by Heretic12 (2024-12-02 16:48:41)

seth · 2024-12-02 16:40:09

The only thing that could break things this har (esp. if only booting the multi-user.target) is the kernel. Maybe some zealous power management tool.
If this affects the LTS kernel and maybe 6.11.x (you can downgrade the kernel in isolation, this is fine for the moment) and because of your history of issues and the impact of the CPU clock, … have you checked whether you currently can still boot and use grml (slightly older and completely different SW stack and probably not misconfigured) to rule out this being defective hardware?
The keyboard still works in the UEFI?
Maybe the handover fails? Can you configure sth. like "legacy usb"?

Heretic12 · 2024-12-02 17:13:51

OK, managed to boot! But... now Network Manager doesn't see wifi devices)

seth · 2024-12-02 17:15:19

Managed to boot what? The LTS kernl? 6.11? grml?
Do you rely on some OOT module (aur package) for wifi to work?

lspci -k
rfkill

Heretic12 · 2024-12-02 17:24:06

Linux-lts, boot was as usual (commented everything in fstab except / and with cpu multiplier being 20). Wifi works in boot iso btw

Last edited by Heretic12 (2024-12-02 17:26:23)

seth · 2024-12-02 17:32:34

seth wrote:

Do you rely on some OOT module (aur package) for wifi to work?
lspci -k
rfkill

Heretic12 · 2024-12-02 18:00:37

No, only Network Manager. rfkill returns

rfkill: cannot open /dev/rfkill: No such file or directory

seth · 2024-12-02 20:50:59

NM isn't a driver, it's more like the broadcom wl driver etc. Please post the output of

lspci -k
modinfo rfkill
lsmod | grep rfkill

Heretic12 · 2024-12-02 22:44:39

So, I ran this commands from chroot and from booted system (saved results into files and posted them from chroot)
Command lsmod | grep rfkill didn't return anything in the system, but returned a result in chroot (included in the following file):

rfkill 40960 5 iwlmvm,bluetooth,cfg80211

Command lspci -k returned from system: http://0x0.st/X7Na.txt and from chroot: http://0x0.st/X7NB.txt
Command modinfo rfkill both in chroot and booted system returned

modinfo: ERROR: Module rfkill not found.

seth · 2024-12-02 22:52:11

00:14.3 Network controller: Intel Corporation Comet Lake PCH CNVi WiFi
	DeviceName: Onboard - Ethernet
	Subsystem: Intel Corporation Dual Band Wi-Fi 6(802.11ax) AX201 160MHz 2x2 [Harrison Peak]

00:14.3 Network controller: Intel Corporation Comet Lake PCH CNVi WiFi
	DeviceName: Onboard - Ethernet
	Subsystem: Intel Corporation Dual Band Wi-Fi 6(802.11ax) AX201 160MHz 2x2 [Harrison Peak]
	Kernel driver in use: iwlwifi

https://bbs.archlinux.org/viewtopic.php?id=298631

Heretic12 · 2024-12-02 22:55:07

Ah, and the most important - the journal of a previous boot: http://0x0.st/X7NM.txt

UPD
Neither boot_delay=30 or rootdelay=30 fixed the issue with wifi device. But still I have a workaround - save everything to files and post their content from chroot

Last edited by Heretic12 (2024-12-02 23:26:21)

ReDress · 2024-12-02 23:45:44

Heretic12 wrote:

Did 1 full run with memtest86+ with XMP on - passed, no errors

This gets me wondering whether memtest86+ tests whether everything works as expected while at the same time checking for errors.

Or, it runs broken tests to test whether they fail.

seth · 2024-12-03 05:00:03

Ah, and the most important - the journal of a previous boot: http://0x0.st/X7NM.txt

I suspect the LTS kernel is too old for your wifi NIC, what else sticks out in the LTS journal is that you've nit installed nvidia-lts and are running on nouveau.
Does nouveau allow you to boot/run the main kernel?

@ReDress, https://bbs.archlinux.org/viewtopic.php … 7#p2209367 (it's a looong thread … )

ReDress · 2024-12-03 06:20:23

For what it's worth.

The first approach seems like testing. The second sounds like debugging. Both might achieve the desired result.

Guess it's a matter of choice.

Heretic12 · 2024-12-03 16:01:37

I suspect the LTS kernel is too old for your wifi NIC

LTS previuosly worked normal - it stopped working only recently - just with mkinitcpio which returns something like this (took a picture with a phone):

==> Building image from preset: /etc/mkinitcpio.d/linux-lts.preset: 'default
==> Using default configuration file: '/etc/mkinitcpio.conf
  -> -k /boot/vmlinuz-linux-lts -g /boot/initramfs-linux-lts.img
==> ERROR: Invalid option -g -- '/boot/initramfs-linux-lts.img' must be writable
==> Building image from preset: /etc/mkinitcpio.d/linux-lts.preset: 'fallback
==> Using default configuration file: '/etc/mkinitcpio.conf
  -> -k /boot/vml inuz-linux-lts -g /boot/initramfs-linux-lts-fallback.ing -S autodetect
==> ERROR: Invalid option -g -- '/boot/initramfs-linux-lts-fallback.img' must be uritable
==> Building image from preset: /etc/mkinitcpio.d/linux.preset: default
==> Using default configuration file: '/etc/mkinitcpio.conf
  -> -k /boot/vmlinuz-1inux -g /boot/initramfs-linux. img
==> ERROR: Invalid option -g -- '/boot/initramfs-linux.img' must be writtable
==> Building image from preset: /etc/mkinitcpio.d/linux.preset: fallback
==> Using default configuration file: '/etc/mkinitcpio.conf
  -> -k /boot/vmlinuz-linux -g /boot/initramfs-linux-fallback. img -S autodetect
 ==> ERROR: Invalid option -g -- "/boot/initramfs-linux-fallback. img' must be writable

This is most likely due to my own unprofessional attempts to fix things) Main issue still remains though - I/O Error (and it's connection to CPU clocks and possibly videogames)...

UPD
Managed to boot with /home and /boot partition uncommented in fstab
But after adding ntfs drives - the boot process started to fail again

Last edited by Heretic12 (2024-12-03 16:25:11)

seth · 2024-12-03 17:15:51

cat /etc/mkinitcpio.d/linux-lts.preset
stat /boot/initramfs-linux.img
mount | grep boot

Edit: the ntfs partitions are likely flagged as dirty, https://wiki.archlinux.org/title/NTFS#U … rked_dirty but preferably use chkdsk on windows.

Last edited by seth (2024-12-03 17:18:01)

Heretic12 · 2024-12-03 18:27:39

http://0x0.st/X7Be.txt

Arch Linux

#76 2024-11-30 19:34:57

Re: Constant Input/Output Error [Solved]

#77 2024-11-30 20:11:11

Re: Constant Input/Output Error [Solved]

#78 2024-11-30 20:52:42

Re: Constant Input/Output Error [Solved]

#79 2024-12-01 07:17:19

Re: Constant Input/Output Error [Solved]

#80 2024-12-01 15:42:47

Re: Constant Input/Output Error [Solved]

#81 2024-12-01 21:59:06

Re: Constant Input/Output Error [Solved]

#82 2024-12-02 01:41:25

Re: Constant Input/Output Error [Solved]

#83 2024-12-02 09:05:54

Re: Constant Input/Output Error [Solved]

#84 2024-12-02 16:15:35

Re: Constant Input/Output Error [Solved]

#85 2024-12-02 16:40:09

Re: Constant Input/Output Error [Solved]

#86 2024-12-02 17:13:51

Re: Constant Input/Output Error [Solved]

#87 2024-12-02 17:15:19

Re: Constant Input/Output Error [Solved]

#88 2024-12-02 17:24:06

Re: Constant Input/Output Error [Solved]

#89 2024-12-02 17:32:34

Re: Constant Input/Output Error [Solved]

#90 2024-12-02 18:00:37

Re: Constant Input/Output Error [Solved]

#91 2024-12-02 20:50:59

Re: Constant Input/Output Error [Solved]

#92 2024-12-02 22:44:39

Re: Constant Input/Output Error [Solved]

#93 2024-12-02 22:52:11

Re: Constant Input/Output Error [Solved]

#94 2024-12-02 22:55:07

Re: Constant Input/Output Error [Solved]

#95 2024-12-02 23:45:44

Re: Constant Input/Output Error [Solved]

#96 2024-12-03 05:00:03

Re: Constant Input/Output Error [Solved]

#97 2024-12-03 06:20:23

Re: Constant Input/Output Error [Solved]

#98 2024-12-03 16:01:37

Re: Constant Input/Output Error [Solved]

#99 2024-12-03 17:15:51

Re: Constant Input/Output Error [Solved]

#100 2024-12-03 18:27:39

Re: Constant Input/Output Error [Solved]

Board footer