Chroot Rescue Guide
Chroot = enter your broken system from the outside and operate on it directly. The Recovery Environment automates common cases; here’s the manual craft.
1. Boot the live ISO, identify partitions
lsblk -f
# typical UEFI layout: p1 = EFI (vfat), p2 = root
2. Mount everything in place
sudo mount /dev/nvme0n1p2 /mnt
sudo mount /dev/nvme0n1p1 /mnt/boot/efi # UEFI only
for d in dev proc sys run; do sudo mount --rbind /$d /mnt/$d; done
sudo cp -L /etc/resolv.conf /mnt/etc/ # DNS for network inside
Btrfs subvolume roots: mount with -o subvol=@ first (Btrfs Management).
3. Enter
sudo arch-chroot /mnt # ships on the ISO (arch-install-scripts)
You now have your system’s shell, packages, and configs.
4. Fix things
pacman -Syu # finish interrupted upgrades
pacman -S linux mkinitcpio # reinstall kernel + regenerate images
grub-mkconfig -o /boot/grub/grub.cfg
passwd youruser # password resets
nano /etc/fstab # UUID corrections
journalctl -b -1 -p err # inspect the failed boot
Gotchas
-
Encrypted root:
cryptsetup open /dev/nvme0n1p2 cryprootbefore mounting /dev/mapper/cryproot. -
Wrong-mode GRUB (installing BIOS GRUB on UEFI box) — match targets:
--target=x86_64-efi. -
If pacman complains about locks post-crash:
rm /var/lib/pacman/db.lckinside chroot.
Related: structured boot triage · the automated path.