Debian : Failed to find logical volume at boot
Thursday, 17 January 2019
Écrit par
Grégory Soutadé

After further investigations, I found a correct fix to this. In facts, its my configuration that is wrong. I have LUKS on LVM schema like this :

system_group (LVM)
    --> system_crypt (LUKS)
swap_group (LVM)
    --> swap_crypt (LUKS)
home_group (LVM)
    --> home_crypt (LUKS)

To be right handled, you need to declare :

In /etc/crypttab :

home--group-home_crypt UUID=349ca075-2922-4c9c-a52b-8dce587767ea /root/home.key luks
swap--group-swap_crypt UUID=4490ce3c-8700-4e90-81df-250cd3573b7c /root/swap.key luks
system--group-system_crypt UUID=95e39100-25c2-41be-829a-bd84fcb21d0a none luks

use blkid command to get right UUID values

In /etc/fstab :

/dev/mapper/system--group-system_crypt /           ext4    errors=remount-ro 0       1
UUID=6866a661-0424-472c-853e-6daa20d15d74 /boot    ext4    defaults          0       2
/dev/mapper/home--group-home_crypt /home           ext4    defaults          0       2
/dev/mapper/swap--group-swap_crypt none            swap    sw                0       0

In /etc/initramfs-tools/conf.d/resume :


Then, do sudo update_initramfs -u and restart.

It's the second time it happens ! After an update my Debian refuse to boot. This time, I wasn't asked for anything !!

I have an LUKS on LVM configuration. I don't use UUID in my cryptroot and Debian scripts only activate devices with it ! I need to manually add "vgchange" command to mount all devices. Unfortunately, the patched script was overwrote by update.

If it happens, follow this procedure :

  • At boot, wait for the rescue shell (~5 minutes)
  • Enter "vgchange -ay"
  • Enter "exit"

Now, you may be able to boot into your system. Then :

  • Edit /usr/share/initramfs-tools/scripts/local-top/cryptroot
  • Add "vgchange -ay" in wait_for_source() function
  • Update initramfs with "sudo update-initrmafs -u"

Safely reboot !

Auteur :

e-mail* :

Le commentaire :

* Seulement pour être notifié d'une réponse à cet article