Boot into a grml.org rescue system.

Rescue workplace

How to setup my workplace, if the disk which needs rescue is /dev/vda2:

fsck /dev/vda2
mkdir -p /mnt/rescue
mount /dev/vda2 /mnt/rescue
for i in /dev /dev/pts /sys /proc; do mount -B $i /mnt/rescue$i; done
chroot /mnt/rescue /bin/bash
export LC_ALL=C

Rescue an encrypted disk

cryptsetup luksOpen /dev/sda3 sda3_crypt
vgchange -a y
mount /dev/xy-vg/root-lv /mnt/target

Rescue a UEFI based system

Inspired by this page.

Boot into a grml.org rescue system and check that the computer is booted in EFI mode:

[ -d /sys/firmware/efi ] && echo "EFI boot on HDD" || echo "Legacy boot on HDD"
should return "EFI boot on HDD".

chroot /mnt/rescue /bin/bash
export LC_ALL=C
apt-get install --reinstall grub-efi
grub-install --recheck /dev/sda
update-grub

efibootmgr --verbose | grep debian
file /boot/efi/EFI/debian/grubx64.efi
/boot/efi/EFI/debian/grubx64.efi: PE32+ executable (EFI application) x86-64 (stripped to external PDB), for MS Windows

Rescue a broken initrd image

Boot into a grml.org rescue system, according to above.

apt-get install --reinstall linux-image-3.16.0-4-amd64
update-grub

If it is a UEFI system, check the bootloader:

efibootmgr --verbose | grep debian
file /boot/efi/EFI/debian/grubx64.efi
/boot/efi/EFI/debian/grubx64.efi: PE32+ executable (EFI application) x86-64 (stripped to external PDB), for MS Windows
export LC_ALL=C

grub-install –force “dummy” failed

grub-install: warning: vars_set_variable: write(): failed: No space on device

With a preseeded unattended install the filesystem structure of /boot /boot/efi always looks the same. The filesystem is not really reset. In this example /dev/sda1 is /boot/efi:

dd if=/dev/zero of=/dev/sda1; sync

Links