You are not logged in.
After doing a full system update, booting to arch, systemd won't find the root partition, but the UUID is correct in the fstab and in the efibootmgr. I've tried to arch-chroot from a live usb and doing mkinitcpio -P but nothing, no errors, but still I can't boot to it
Last edited by BlackCherry (2024-12-24 03:32:43)
Offline
I've tried to arch-chroot from a live usb and doing mkinitcpio -P but nothing, no errors, but still I can't boot to it
Do you have a separate /boot/ partition? Is there another /boot/ on the root partition that contains the new kernel & initramfs?
If so run `mount -a` in the chroot to mount everything listed in /etc/fstab and then reinstall the kernel package(s).
Jin, Jîyan, Azadî
Offline
Yes I have a separate boot partition and I have it mounted in my live also (/mnt/boot in my case because the root partition is in /mnt/), just to make the mkinitcpio effective. what do you mean by 'reinstall the kernel package(s)'?
Offline
You will need the kernel image on the /boot/ partition to match the contents of /usr/lib/modules/ so it can load the filesystem drivers, that won't happen unless you reinstall the kernel with /boot/ mounted correctly.
Run fsck on the ESP first though, there must have been a reason it was un-mounted for the upgrade. Perhaps the FAT fileystem got corrupted somehow.
Jin, Jîyan, Azadî
Offline
~#fsck /dev/nvme0n1p2
fsck from util-linux 2.40.2
fsck.fat 4.2 (2021-01-31)
There are differences between boot sector and its backup.
This is mostly harmless. Differences: (offset:original/backup)
65:01/00
1) Copy original to backup
2) Copy backup to original
3) No action
[123?q]?whats this? it's in the ESP
Offline
Looks like it was un-mounted incorrectly. Use the -a option to repair the filesystem automatically.
Jin, Jîyan, Azadî
Offline
~# fsck -a /dev/nvme0n1p2
fsck from util-linux 2.40.2
fsck.fat 4.2 (2021-01-31)
There are differences between boot sector and its backup.
This is mostly harmless. Differences: (offset:original/backup)
65:01/00
Not automatically fixing this.
Dirty bit is set. Fs was not properly unmounted and some data may be corrupt.
Automatically removing dirty bit.
*** Filesystem was changed ***
Writing changes.
/dev/nvme0n1p2: 11 files, 121209/261372 clustersThis will be enough? and what about the root partition?
Offline
That should be fine but I would consider re-formatting the partition and switching the mountpoint to /efi using autofs to avoid exposure.
You can check the root partition if you want but most other filesystems are more resilient than FAT.
Jin, Jîyan, Azadî
Offline
still can't boot
waiting 10 seconds for device /dev/disk/by-uuid/"root partition uuid"
ERROR: device 'UUID="root partitio uuid"' not found. Skipping fsck.Offline
So run the `file` command on the kernel image on the /boot/ partition to confirm the version and then check or share your bootloader configuration.
Also double-check /etc/fstab against the output of `blkid`.
Jin, Jîyan, Azadî
Offline
Check "lsblk -f" whether the UUID is actually correct.
Can you boot the fallback initramfs? Maybe the nvme modules are missing in the initramfs?
Alternatively: Can you install and boot the LTS kernel?
Offline
Can you boot the fallback initramfs?
How can I do that? also I triple checked but UUIDs are correct
Offline
My efibootmgr:
efibootmgr --create --disk /dev/nvme0n1 --part 2 --label "Arch Linux" --loader /vmlinuz-linux --unicode 'root=UUID=157dd383-382e-4399-abfd-79aca335be92 rw initrd=\amd-ucode.img initrd=\initramfs-linux.img' --verboseMy fstab:
# /dev/nvme0n1p3
UUID=157dd383-382e-4399-abfd-79aca335be92 / ext4 rw,relatime 0 1
# /dev/nvme0n1p2
UUID=6E0C-8086 /boot vfat rw,relatime,fmask=0022,dmask=0022,codepage=437,iocharset=ascii,shortname=mixed,utf8,errors=remount-ro 0 2Offline
fstab and efibootmgr matching is meaningless, you need to check "lsblk -f" (from the install iso, before chrooting)
You can select the failsafe/fallback initramfs at your bootloader (it's the same kernel, you're just booting a different initramfs)
Offline
okay we've made progress, initramfs-fallback does work but the main initramfs don't
Offline
I mean, with the initramfs-fallback my system boots perfectly
Offline
Maybe the nvme modules are missing in the initramfs?
sudo lsinitcpio /boot/initramfs-linux.img | grep -i nvmeEither the autodetection fails or you've some bogus stuff in your mkinitcpio.conf?
Offline
# lsinitcpio /boot/initramfs-linux.img | grep -i nvme
usr/lib/modules/6.12.4-arch1-1/kernel/drivers/nvme/
usr/lib/modules/6.12.4-arch1-1/kernel/drivers/nvme/common/
usr/lib/modules/6.12.4-arch1-1/kernel/drivers/nvme/common/nvme-auth.ko.zst
usr/lib/modules/6.12.4-arch1-1/kernel/drivers/nvme/host/
usr/lib/modules/6.12.4-arch1-1/kernel/drivers/nvme/host/nvme-core.ko.zst
usr/lib/modules/6.12.4-arch1-1/kernel/drivers/nvme/host/nvme.ko.zstAnd in the fallback
# lsinitcpio /boot/initramfs-linux-fallback.img | grep -i nvme
usr/lib/modules/6.12.6-arch1-1/kernel/drivers/nvme/
usr/lib/modules/6.12.6-arch1-1/kernel/drivers/nvme/common/
usr/lib/modules/6.12.6-arch1-1/kernel/drivers/nvme/common/nvme-auth.ko.zst
usr/lib/modules/6.12.6-arch1-1/kernel/drivers/nvme/common/nvme-keyring.ko.zst
usr/lib/modules/6.12.6-arch1-1/kernel/drivers/nvme/host/
usr/lib/modules/6.12.6-arch1-1/kernel/drivers/nvme/host/nvme-core.ko.zst
usr/lib/modules/6.12.6-arch1-1/kernel/drivers/nvme/host/nvme-fabrics.ko.zst
usr/lib/modules/6.12.6-arch1-1/kernel/drivers/nvme/host/nvme-fc.ko.zst
usr/lib/modules/6.12.6-arch1-1/kernel/drivers/nvme/host/nvme-rdma.ko.zst
usr/lib/modules/6.12.6-arch1-1/kernel/drivers/nvme/host/nvme-tcp.ko.zst
usr/lib/modules/6.12.6-arch1-1/kernel/drivers/nvme/host/nvme.ko.zst
usr/lib/modules/6.12.6-arch1-1/kernel/drivers/nvme/target/
usr/lib/modules/6.12.6-arch1-1/kernel/drivers/nvme/target/nvme-fcloop.ko.zst
usr/lib/modules/6.12.6-arch1-1/kernel/drivers/nvme/target/nvme-loop.ko.zst
usr/lib/modules/6.12.6-arch1-1/kernel/drivers/nvme/target/nvmet-fc.ko.zst
usr/lib/modules/6.12.6-arch1-1/kernel/drivers/nvme/target/nvmet-rdma.ko.zst
usr/lib/modules/6.12.6-arch1-1/kernel/drivers/nvme/target/nvmet-tcp.ko.zst
usr/lib/modules/6.12.6-arch1-1/kernel/drivers/nvme/target/nvmet.ko.zstI ... don't know what to say x,D
Offline
Why do the fallback modules have a different version?
Did you run the `file` command on the kernel image on the ESP? Have you used `lsblk -f` or `blkid` to confirm the UUIDs?
Jin, Jîyan, Azadî
Offline
What's in your /etc/mkinitcpio.d/linux.preset?
Offline
What's in your /etc/mkinitcpio.d/linux.preset?
# cat /etc/mkinitcpio.d/linux.preset
# mkinitcpio preset file for the 'linux' package
#ALL_config="/etc/mkinitcpio.conf"
ALL_kver="/boot/vmlinuz-linux"
PRESETS=('default' 'fallback')
#default_config="/etc/mkinitcpio.conf"
#default_image="/boot/initramfs-linux.img"
default_uki="/boot/EFI/Linux/arch-linux.efi"
default_options="--splash /usr/share/systemd/bootctl/splash-arch.bmp"
#fallback_config="/etc/mkinitcpio.conf"
#fallback_image="/boot/initramfs-linux-fallback.img"
fallback_uki="/boot/EFI/Linux/arch-linux-fallback.efi"
fallback_options="-S autodetect"Offline
Why is that set up for UKIs when you're not using UKIs?
Offline
yep, that's the funny thing I've discover, I've never have an initramfs because I use a UKI but in november I have a bad update that broke my efiboot entry and searching in the internet I saw that effectively I'm missing the efiboot entry and also my initramfs, but in that moment I forgot my UKI, so basically sorry to all of you that helped me with this it was my fault no remembering that the UKI its a replacement for the initramfs, although I dont know why the initramfs-fallback is up to date, and thank you all so much for the support
Offline
Please always remember to mark resolved threads by editing your initial posts subject - so others will know that there's no task left, but maybe a solution to find.
Thanks.
Offline
I would like to know why my initramfs-fallback is the only updating while having UKI, because with UKI I should not have any initramfs
Offline