Repairing a Linux install (broken bootloader, kernel, fstab) that sits on a LUKS-encrypted LVM volume requires unlocking and activating both layers before you can mount it and chroot in.


Find the Partition

lsblk -f lists block devices with their filesystem type and label, letting you pick out the LUKS-encrypted partition (shown as crypto_LUKS) among the disk’s other partitions.

Unlock the LUKS Container

cryptsetup luksOpen <device> <name> prompts for the passphrase and exposes the decrypted volume at /dev/mapper/<name>.

Activate the LVM Volume Group

The decrypted volume is usually an LVM physical volume, not a filesystem directly. vgscan makes the kernel aware of the volume group inside it, and vgchange -ay activates all its logical volumes, making them appear under /dev/mapper/. Run lsblk -f again to confirm the logical volume’s device name.

Mount for Chroot

Mount the root logical volume, then the separate /boot partition (if any) inside it, then bind-mount the pseudo-filesystems the chrooted environment needs to function (/dev, /proc, /sys, etc.) before running chroot.

Tear Down

Unmount everything, then deactivate the volume group — this locks the LUKS container back up implicitly, since its device-mapper entry depends on the VG being inactive first (a plain luksClose would otherwise fail with “device is busy”).


Cheatsheet

Find the Partition

sudo lsblk -f

Unlock the LUKS Container

sudo cryptsetup luksOpen <device> <name>

Activate the LVM Volume Group

sudo vgscan
sudo vgchange -ay
sudo lsblk -f  # confirm the new /dev/mapper/<vg>-<lv> name

Mount for Chroot

sudo mkdir /mnt/<name>
sudo mount /dev/mapper/<vg>-<lv> /mnt/<name>
sudo mount <boot-partition> /mnt/<name>/boot
for i in /dev /dev/pts /proc /sys /run; do sudo mount -B $i /mnt/<name>/$i; done
sudo chroot /mnt/<name>

Tear Down

sudo sync
sudo umount -R /mnt/<name>
sudo vgchange -an <vg>