OPINIONATED OMARCHY
TROUBLESHOOTING CORPUS
790
RECORDS
790
AUDITED
0
UNCHECKED
2388
SOURCES
LIVE
boot-kernel critical common DISPUTE

Fix "Kernel panic - not syncing: VFS: Unable to mount root fs on unknown-block(0,0)"

archomarchyendeavouroscachyosmanjarogrubsystemd-bootliminedesktoplaptop
CORRECTED An auditor found the fix wrong or incomplete and rewrote it.
AUDIT NOTE

Cause analysis and the chroot sequence are correct, and the EndeavourOS anecdote (config regen, not grub-install, was the cure) is a genuinely useful detail. Two problems: sudo bootctl update is presented as the systemd-boot equivalent of regenerating boot config, which it is not. It only replaces the systemd-boot EFI binary on the ESP and leaves stale entries untouched. And pacman -S linux in a chroot whose database may be mid-upgrade is how people compound a partial upgrade. pacman -Syu first is the safe order.

The cause above was not rewritten and may still contain the error described. The fix below is the corrected version.
RISK

You are operating on the bootloader from a chroot. Mount the correct ESP at /mnt/boot before running grub-mkconfig or bootctl, or you will write boot files into a directory on the root filesystem where the firmware will never find them.

Symptom

The screen fills with a kernel trace immediately after the bootloader hands over:


Kernel panic - not syncing: VFS: Unable to mount root fs on unknown-block(0,0)
CPU: 0 PID: 1 Comm: swapper/0 Not tainted 6.12.x-arch1-1

Sometimes with a QR code (systemd's kernel panic screen). Happens after a botched update, after repartitioning for dual boot, or after a motherboard/BIOS change.

Cause

unknown-block(0,0) means the kernel got **no initramfs at all**, or an initramfs that ended without ever mounting a root filesystem. Either the bootloader entry's initrd line points at a file that does not exist (deleted, or wiped by a full /boot), or the initramfs was never regenerated after the kernel changed, or the bootloader config still references an old kernel/root layout after repartitioning.

Fix

Boot a live USB, chroot in (see the chroot record, and get the Btrfs subvolume and ESP mount point right first), then:


lsblk -f
sudo cryptsetup open /dev/nvme0n1p2 cryptroot   # only if LUKS
sudo mount -o subvol=@ /dev/mapper/cryptroot /mnt
sudo mount /dev/nvme0n1p1 /mnt/boot             # confirm against /mnt/etc/fstab
sudo arch-chroot /mnt

# inside the chroot
cat /etc/fstab                 # verify /boot vs /boot/efi vs /efi before anything else
pacman -Syu                    # finish any half-done upgrade FIRST
pacman -S linux                # reinstalls vmlinuz + triggers mkinitcpio
mkinitcpio -P
ls -l /boot                    # vmlinuz-linux AND initramfs-linux.img must both exist

Then regenerate the loader's own config:


grub-mkconfig -o /boot/grub/grub.cfg    # GRUB
limine-mkinitcpio                        # Omarchy / Limine

For **systemd-boot**, bootctl update only updates the EFI binary. It does not write entries. Check and fix the entry itself:


bootctl list
cat /boot/loader/entries/*.conf    # the `initrd` line must name a file that exists

Then:


exit
sudo umount -R /mnt
reboot

Verify

ls -l /boot/vmlinuz-linux /boot/initramfs-linux.img both exist with current timestamps, and the boot entry's initrd path matches a real file. The machine boots to a login prompt.

Sources

kernel-panic-vfs-unable-to-mount-root-fs