Linux · Troubleshooting

Recover a system that won't boot normally

A Linux system that won't boot almost always still gives you a path back in — GRUB's boot menu, recovery mode, or a live USB — the key is knowing which layer is failing (bootloader, kernel, or graphical session) so you fix the right thing instead of reinstalling.

How to open Troubleshooting Boot Issues

  1. Boot normally (not into recovery mode) after applying a fix, to confirm it resolved the issue rather than just working around it.
  2. Check `journalctl -b` for new errors even on a successful boot — some issues degrade rather than fully block booting.
  3. Take a snapshot or note the working kernel/driver version once confirmed stable, so there's a known-good rollback point next time.

What you should know

  • At the GRUB menu (hold Shift or Esc during boot if it's hidden), select "Advanced options" for a recovery mode boot or an older kernel version — useful if the latest kernel or driver update broke booting.
  • If it boots to a black screen instead of the login screen, switch to a TTY with Ctrl+Alt+F3, log in, and check `journalctl -b -1` (previous boot's logs) or `journalctl -xb` (current boot) for the specific failure.
  • A live USB of your distro (or any distro) lets you mount the broken system's partition (`sudo mount /dev/sdXN /mnt`) and inspect or fix config files, back up data, or chroot in to repair GRUB.
  • Reinstall/repair GRUB from a live USB after chrooting: `sudo grub-install /dev/sdX && sudo update-grub` (Debian/Ubuntu) or `sudo grub2-mkconfig -o /boot/grub2/grub.cfg` (Fedora).

Don't panic-reinstall — booting a live USB and reading the actual error in `journalctl` takes a few minutes and usually reveals a fixable one-line cause (bad fstab entry, broken driver, misconfigured GRUB) rather than a reason to start over.

Why it matters

A failed boot is the most stressful Linux troubleshooting scenario for most people because it looks catastrophic, but the underlying causes (bad fstab entry, broken driver, misconfigured GRUB, full disk) are almost always specific and fixable once you can see the actual error instead of just a black screen.

Best practices

  • Keep at least one older kernel installed (most distros do this by default) as a fallback if a new kernel breaks boot.
  • Have a live USB of your distro prepared in advance rather than needing to create one during an actual emergency without another working computer.
  • Take a system snapshot (see Backup & Snapshots) before major updates so a broken boot is a one-command rollback instead of a manual repair.

Common problems and fixes

"Failed to mount" or emergency mode (initramfs prompt) at boot
Almost always a bad `/etc/fstab` entry — from initramfs, remount root read-write (`mount -o remount,rw /`), edit fstab, and reboot.
GRUB menu missing entirely, boots straight past it
Hold Shift (BIOS) or Esc (UEFI) right after power-on to force the menu open, or reinstall GRUB from a live USB if it's not appearing at all.

Tips and tricks

  • `journalctl -b -1 -p err` filters the previous boot's log down to just error-level messages, cutting out a lot of noise when hunting for the actual cause.

Frequently asked questions

Will reinstalling fix a boot problem faster than troubleshooting it?

Rarely — most boot failures are a single misconfigured file or driver, fixable in a few minutes from a live USB, while a reinstall costs much more time and risks personal data if backups aren't current.