Steam Deck won’t boot after unexpected shutdown and failed SteamOS reinstall

[Big Boss]

Hi everyone,

I was playing a game on my Steam Deck when it suddenly shut down out of nowhere. I decided to reinstall SteamOS to fix any potential issues, but during the installation process, the USB drive disconnected. Since then, my Steam Deck hasn’t been able to recognize the SSD.

Now, every time I try to reinstall SteamOS using different USB drives, I keep running into errors, and it seems like the system can’t properly detect or mount the SSD. I’m not sure if the SSD is dead or if it’s just not formatted correctly.

I’ve attached a picture of the error screen I’m getting. Has anyone experienced something similar? Any advice on how to fix this or check if the SSD is still functional?

this is the error in my screen


[ 0.228041] 2 M3 A5 06 9F
[ 0.310757] Expanded Resource Reserved due to conflict with PCI Bus 0000:00
:: Running early hook [udev]
Starting systemd-udevd version 253.5-2-arch
:: Triggering uevents...
[ 0.667353] i915 0000:00:02.0: [drm] *ERROR* No MOCS PTEs? W/A workaround disabled, if this is incorrect please boot with 10042.nopm
[ 0.702523] udevd[169]: unknown parameter 'speed_dev' ignored
:: Running hook [systemd]
:: Running hook [teatmnts]
:: Loading compiled-in X.509 certificates
Scanning for PCI devices
[ 1.176808] BTRFS: device label "steam" devid 1 transid 21 /dev/nvme0n1p3 scanned by systemd-udevd (176)
[ 1.176828] BTRFS critical (device nvme0n1p3): unable to find logical device
Mounting /dev/disk/by-partlabel/steam-root as root
Waiting for root device /dev/disk/by-partlabel/steam-root...
[ 2.108179] BTRFS error: device /dev/sda3 already registered with a higher generation, found 21 expect 25
Setting root target to /dev/disk/by-partlabel/steam-rootfs
:: Running hook [resume]
ERROR: resume: no device specified for hibernation
Mounting /dev/disk/by-partlabel/steam-rootfs on real root
[ 2.163203] BTRFS error: device /dev/sda3 already registered with a higher generation, found 21 expect 25
mount: mounting /new_root failed: File exists
ERROR: failed to mount /dev/disk/by-partlabel/steam-rootfs on real root
You are now being dropped into an emergency shell.
sh: can't access tty; job control turned off
[rootfs ]#
[rootfs ]#
[rootfs ]#


Thanks in advance for any help!

[thetargos]

Originally posted by Big Boss:

Hi everyone,

Now, every time I try to reinstall SteamOS using different USB drives, I keep running into errors, and it seems like the

[    0.228041]    2  M3  A5  06  9F
[    0.310757] Expanded Resource Reserved due to conflict with PCI Bus 0000:00
:: Running early hook [udev]
Starting systemd-udevd version 253.5-2-arch
:: Triggering uevents...
[    0.667353] i915 0000:00:02.0: [drm] *ERROR* No MOCS PTEs? W/A workaround disabled, if this is incorrect please boot with 10042.nopm
[    0.702523] udevd[169]: unknown parameter 'speed_dev' ignored
:: Running hook [systemd]
:: Running hook [teatmnts]
:: Loading compiled-in X.509 certificates
Scanning for PCI devices
[    1.176808] BTRFS: device label "steam" devid 1 transid 21 /dev/nvme0n1p3 scanned by systemd-udevd (176)
[    1.176828] BTRFS critical (device nvme0n1p3): unable to find logical device
Mounting /dev/disk/by-partlabel/steam-root as root
Waiting for root device /dev/disk/by-partlabel/steam-root...
[    2.108179] BTRFS error: device /dev/sda3 already registered with a higher generation, found 21 expect 25
Setting root target to /dev/disk/by-partlabel/steam-rootfs
:: Running hook [resume]
ERROR: resume: no device specified for hibernation
Mounting /dev/disk/by-partlabel/steam-rootfs on real root
[    2.163203] BTRFS error: device /dev/sda3 already registered with a higher generation, found 21 expect 25
mount: mounting /new_root failed: File exists
ERROR: failed to mount /dev/disk/by-partlabel/steam-rootfs on real root
You are now being dropped into an emergency shell.
sh: can't access tty; job control turned off
[rootfs ]#
[rootfs ]#
[rootfs ]#

Based solely on these messages it would seem as if the SSD is indeed faulty. HOWEVER, having experienced somewhat inconsistent results myself when I upgraded my SD's storage (from 64GiB to 1 TiB), it resulted to be corrupted recovery image. I had to redo the process four times. To add insult to.injury, Valve does not provide a SHA1SUM file for the correct HASH of the image, plus also you may run into issues when creating the Bootable media, plus the boot process of the recovery image lacks a self test of the image.

My suggestion would be to boot into the recovery KDE session (if you get as far), open partition manager, delete all partitions on the main drive and start the install process.

[Big Boss]

Thanks for the response! That makes sense. I’ll try installing Windows on a microSD to see if it detects the SSD and allows me to format it. Right now, I can’t access the KDE recovery session, so unless there’s a way to boot into KDE from external storage without relying on the SSD, I’ll have to go with the Windows method first.

I also tried installing Bazzite, but it wouldn’t let me install it on either the microSD or the SSD. The SSD appears almost completely full, even though I can’t access it properly.

I’ll update once I test it. Let me know if you have any other suggestions!


EDIT: WINDOWS NOT WORK

[Cqllel]

I had this issue with my laptop. It began randomly black screening outa nowhere one day. It randomly shut off, sometimes immediate black screens to the boot screen (ASUS logo for me)

One day it shut off and immediately opened up to BIOS where it said "No Storage Device Detected", and that confirmed it

I ended up buying a new SSD for it. So if you're having black screening issues, it's likely your SSD is corrupted or faulty

[thetargos]

If you are able to get the recovery image to the Live session (you should be able to it in every other distronlive media, actually), you could try and attempt to diagnose the main drive, but since it also fails with /dev/sda (assuming here that it is the USB media), is why I thought of the image being corrupt, but it could be the unfortunate event that you have both, a corrupted recovery image and a failing drive. Though given the age of the SD, I would not think of the latter unless you have unequivocal evidence by using diagnostic tools