If your server fails to boot, it can be caused by filesystem corruption, misconfigured boot settings, a failed update, or a full disk. This guide walks you through diagnosing and repairing boot issues using the web console and rescue mode.
Symptoms of Boot Failure
- Server shows as "Running" in the dashboard but is unreachable via SSH.
- The web console shows error messages during boot (kernel panic, filesystem errors, GRUB errors).
- The server is stuck at a boot prompt or login screen but does not respond.
- Dashboard shows the server repeatedly restarting.
Step 1 — Check the Web Console
The web console is your primary diagnostic tool when SSH is unavailable:
See Using the Web Console for console details.
Common Boot Error Messages
| Error | Likely Cause |
|---|---|
Kernel panic - not syncing | Corrupt kernel or missing initramfs |
GRUB error: unknown filesystem | Corrupt or missing GRUB configuration |
fsck failed or filesystem errors | Disk corruption |
A start job is running for... (hangs) | A service is failing to start |
No space left on device | Disk is full, preventing boot |
Failed to start (various services) | Misconfigured service blocking boot |
Step 2 — Boot Into Rescue Mode
If normal boot fails, boot into rescue mode:
Rescue mode boots a minimal Linux environment from a separate image, giving you access to your server's disk without loading its operating system.
Step 3 — Mount the Server's Filesystem
In rescue mode, mount your server's disk:
# List available disks
lsblkMount the main partition (adjust device name as needed)
mount /dev/vda1 /mntIf using LVM
vgscan
vgchange -ay
mount /dev/mapper/vg0-root /mntOnce mounted, you can access and repair files on your server's disk.
Common Fixes
Fix Filesystem Corruption
# Unmount first (if mounted)
umount /mntRun filesystem check
fsck -y /dev/vda1Remount after repair
mount /dev/vda1 /mntFix a Full Disk
# Check disk usage
df -h /mntFind large files
du -sh /mnt/var/log/*
du -sh /mnt/tmp/*Clean up logs
truncate -s 0 /mnt/var/log/syslog
rm -f /mnt/var/log/*.gzClear package cache
rm -rf /mnt/var/cache/apt/archives/*.debFix a Failed Service Blocking Boot
# Chroot into the mounted system
mount --bind /dev /mnt/dev
mount --bind /proc /mnt/proc
mount --bind /sys /mnt/sys
chroot /mntDisable the problematic service
systemctl disable problematic-serviceExit chroot
exitFix SSH Configuration
If a misconfigured SSH prevents access:
# Edit the SSH config on the mounted disk nano /mnt/etc/ssh/sshd_configFix the issue (e.g., restore PermitRootLogin, fix port number)
Save and exit
Restore GRUB
# Chroot into the mounted system
chroot /mntReinstall GRUB
grub-install /dev/vda
update-grubExit chroot
exitStep 4 — Reboot Normally
After making repairs:
umount /mntPrevention
- Keep disk usage below 85% — A full disk is a common cause of boot failure.
- Test updates — Apply major updates during maintenance windows.
- Create snapshots — Take a snapshot before making significant changes. See Backup Strategy.
- Monitor your server — Set up alerts for disk usage and service health. See Server Metrics.
When to Contact Support
Contact support if:
- Rescue mode is not available or does not boot.
- The disk appears to have hardware-level issues.
- You are unable to repair the filesystem.
- The problem persists after all repair attempts.
What to Do Next
- Using the Web Console — Console access guide.
- Monitoring Server Performance — Prevent issues proactively.
- Backup Strategy — Protect against data loss.
- Slow Server Performance — Post-repair optimization.