Failed to boot after reboot

FTouzain

Just one thing: when I do sudo clr-boot-manager list-kernels. No kernel seems to be selected. I made a sudo clr-boot-manager update, but it does not seem to change things (no star in front of at less one kernel). Is it normal?

FTouzain

Must I do:
sudo clr-boot-manager set-kernel com.solus-project.current.6.12.17-314 ?

Here are the proposed kernels:

  com.solus-project.current.6.12.17-314
  com.solus-project.current.6.12.12-313
  com.solus-project.lts.6.6.80-265
  com.solus-project.lts.5.15.77-189

The command gives:

Generating grub configuration file ...
Warning: os-prober will be executed to detect other bootable partitions.
Its output will be used to detect bootable binaries on them and create new boot entries.
done

But no kernel is selected.

Is it normal @Harvey please ?

FTouzain

My current live kernel seems to be still Linux fragile 6.12.12-313.current.
I have not rebooted or shutdown my computer because I am not sure it will reboot clearly as not kernel is selected according to clr-boot-manager.

I only switched to standby

So:

can I reboot safely?
if not, what must I do to select last kernel as the current one before rebooting please?

Thank you in advance.

alfisya

It should have been fine. You do not need to check clr-boot-manager everytime a kernel gets updated, eopkg will automatically do it for you (via usysconf trigger). If it ends up not fine, we have a rescue guide in the help center. Cheers!

FTouzain

Many thanks for this information 🙂

FTouzain

It failed to boot in grub, telling:
‘Error : file « /boot/null » not available ´

In addition, when I go in grub command, it does know sudo command, and I don’t see usual directories in /dev but:

urandom
lp1
lp2
lp3
lp0
null
What must I do to rescue my boot please, usual advices of the help section does not seem applyable.

The only line of the grub menu is
Solus 4.7 Endurance (6.2.17-314.current).
I have 4 com… kernel files in the /boot directory

alfisya

Try this guode https://help.getsol.us/docs/user/troubleshooting/boot-rescue

FTouzain

Not applyable as no sudo nor fdisk command.

In the following booting command, what must I write to replace (null) please ?:
linux /boot/(null) root=UUID=… (null)

Please ?

stocc

FTouzain To follow that guide you'll need to boot from a live USB. You can't do it from the Grub prompt. Unfortunately, I'm not familiar enough with Grub to know whether you can get the system to boot from where you are at the Grub prompt. Perhaps others are best placed to advise on that.

FTouzain

Thank you.

I followed the guide, but when doing:

sudo clr-boot-manager update

last step to Regenerate UEFI boot entries before rebooting the system

I get:

[FATAL] cbm (../src/bootman/bootman.c:L548): Cannot determine boot device

All steps before seems ok, and /etc/fstab corresponds to the output of ls -lah /dev/disk/by-uuid

At least, now, sudo clr-boot-manager list-kernels lists all my four kernels (but still without star in front of one of them).
Any idea of how to tell to solus what is my boot device?

I apologize not to have seen that, to follow the guide, I needed to have booted on a live solus usb key

stocc

How was Grub installed initially? Was it by Solus, or some other operating system? Is it a dual-boot machine? It may help if you could provide some system information to help others determine what the issue might be. I think 'inxi -bz' would be a good start.

FTouzain

stocc

Yes, grub was installed with Solus. There is no other system on it.

The booting partition is a nvme (first partition swap, second ext4 for system, no encryption).
I have 5 other disks for data: 1 nvme, 1 sata3 ssd and 3 harddrives.

I looked at the /etc/kernel directory because of the following issue that shows an error looking like mine: https://github.com/getsolus/packages/issues/3742
and I have only this into, the file /etc/kernel/cmdline.d/10_resume.conf in which it is written:

resume=UUID=uuid_of_my_swap_partition

The command inxi -bz gives me:

System:
  Kernel: 6.12.9-312.current arch: x86_64 bits: 64
  Desktop: Budgie v: 10.9.2 Distro: Solus 4.7 endurance
Machine:
  Type: Desktop Mobo: ASRock model: Z370 Professional Gaming i7
    serial: <filter> UEFI: American Megatrends v: P1.10 date: 09/08/2017
CPU:
  Info: 6-core Intel Core i7-8700K [MT MCP] speed (MHz): avg: 800
    min/max: 800/4700
Graphics:
  Device-1: NVIDIA GP104 [GeForce GTX 1070] driver: N/A
  Device-2: Logitech QuickCam Vision Pro driver: snd-usb-audio,uvcvideo
    type: USB
  Display: server: X.org v: 1.21.1.16 with: Xwayland v: 24.1.6 driver: X:
    loaded: nvidia gpu: N/A
  API: OpenGL v: 4.5 vendor: mesa v: 24.3.4 note: incomplete (EGL sourced)
    renderer: llvmpipe (LLVM 19.1.6 256 bits)
  Info: Tools: api: eglinfo,glxinfo gpu: nvidia-settings,nvidia-smi
    x11: xdpyinfo, xprop, xrandr
Network:
  Device-1: Intel Ethernet I219-V driver: e1000e
  Device-2: Intel Dual Band Wireless-AC 3168NGW [Stone Peak] driver: iwlwifi
  Device-3: Intel I211 Gigabit Network driver: igb
  Device-4: Aquantia AQtion AQC107 NBase-T/IEEE 802.3an Ethernet [Atlantic
    10G] driver: atlantic
Drives:
  Local Storage: total: 13.66 TiB used: 365.05 GiB (2.6%)
Info:
  Memory: total: 64 GiB available: 62.72 GiB used: 1.98 GiB (3.1%)
  Processes: 370 Uptime: 1h 19m Shell: Bash inxi: 3.3.37

Maybe originally, I had either ubuntu or elementaryOS before I discovered Solus, but the computer had always only one linux system

FTouzain

Maybe important, the kernel listed above in the output does not correspond to one of the 4 kernels listed by the command sudo clr-boot-manager list-kernels (but I have never removed the 6.12.9-312 manually)

stocc

Perhaps as suggested by @BuzzPCSOS in https://discuss.getsol.us/d/11447-hello-folks, you may want to download and use Rescatux to try to repair your system. It can be downloaded here: https://sourceforge.net/projects/rescatux/. The website (https://www.supergrubdisk.org/rescatux/) has some tutorials which you should look through.

FTouzain

Sorry, I had not seen your answer @stocc .

~~I realized that my fstab seemed to have been changed by the last upgrade~~. I had created a long time ago a swap file on one of my device (in addition of the current swap partition). In the new fstab, upgrade seems to have changed the UUID of one of my disk (my second nvme) by the UUID of my (unused) old swap partition (first portion of the first nvme) and my swap partition UUID for the data disk UUID

Edited 2025 03 12: no fstab modification, swap is swap and booting partition is booting partition, just an error from me. The error is still the same.

Therefore, I followed again all the process of boot rescue with my corrected /etc/fstab file

When I run last command sudo clr-boot-manager update, I get the following message:

[FATAL] cbm (../src/bootman/bootman.c:L548): Cannot determine boot device

I was wondering if it was because my partitions were not mounted. Therefore I mounted them, and retry with the same result..
I have the feeling that everything is ok but clr-boot-manager still tells me that it cannot determine booting device.
And when I do fdisk -l, my supposed boot partition is recognized as Linux and as a booting partition (there is a star in the dedicated column). It doesn't seem that I have EFI except on the USB live key.

Maybe rescatux is a good solution, but I had really bad experiences with partitions and repairs already (it was on a mac, but anyway).

I think my original installation was ubuntu with a swap partition. When I installed Solus a long time ago, I did it on the ubuntu Linux partition.

stocc

I'm no expert, but if your system is UEFI, I thought Solus used systemd-boot rather than Grub?
I personally had an issue recently when I installed another distro which uses Grub, replacing my Solus partition. After deciding to change back to Solus, I re-installed to that same partition and rebooted. I was greeted with the Grub rescue prompt... I then rebooted and accessed the PCs one-time boot menu (F12 at boot up on my Dell) to boot Solus directly. I then mounted the boot partition using clr-boot-manager and deleted the other distro related entries from /boot/EFI and rebooted. Voilà, everything worked as expected.
Your issue seems different, but I hope my experience may give you, or someone else, some ideas...

Sorry that I haven’t been able to be of more help.

FTouzain

stocc

Thank you for your feedback. I have always see grub mentionned with the boot menu, so I believe this is grub2. In addition, fdisk does not detect UEFI partition other that the one of the live USB key. Maybe I will try rescatux if I see no complementary answers or no other way to handle my grub problem (grub menu show only one kernel and is not able to boot on the related drive while 4 kernels are shown by clr-boot-manager).
So, maybe I won't have choice to try your proposition. Thanks again.

BuzzPCSOS

FTouzain While you are working with boot discs it would be a good time to check the SMART data for your hard drive. There are tools to do this on Solus Gnome, Budgie, XFCE and Plasma.
Plasma - use KDE Partition Manager.
Gnome and Budgie - use Disks.
Not sure about XFCE.
Bad drive health could be the cause of the problem. Unfortunately if this is the case then the best fix is to replace your hard drive. But definitely worth a look as it could save many coins going into the swear jar.

FTouzain

Smart control of both nvme is ok "PASSED" (if it means really something for nvme drive), thanks.

So, I must find another usb key for rescatux if no other solution.