Restore Failed Playbook
Who this is for
Users whose restore operation failed mid-way, or who are seeing an error when trying to start a restore.
Pre-Restore Error Playbook
These errors occur before the restore even starts.
| Error | Meaning | Fix |
|---|---|---|
INVALID_STATE | Backup status is not completed | Only restore from completed backups. If the backup is still running, wait. If it failed, create a new backup first. |
SNAPSHOT_RESTORE | Trying to restore a snapshot through CloudAIPilot | Snapshots must be restored via the cloud provider console. See KB-05-05. |
RESTORE_IN_PROGRESS | Another restore is already running | Wait for the current restore to finish, or cancel it — see KB-05-06. |
Mid-Restore Failure Playbook
If the restore started but failed mid-way, the server may be in a partially restored state.
Step 1 — Check What Failed
- Open the Activity Center.
- Find the restore job.
- Expand to see which step failed.
Step 2 — Assess Server State
| Last successful step | Server state |
|---|---|
| Restore not yet started | Server is unchanged |
| Files partially overwritten | Partial files — site may be broken |
| Database partially imported | Database in inconsistent state |
Step 3 — Recover Using Safe Mode Backup
If you had Safe Mode enabled when starting the restore (the default), CloudAIPilot created a pre-restore backup.
- Go to Server detail → Backups tab.
- Find the backup labeled "Pre-restore backup" (created at the time the restore started).
- Restore from that backup to return to the state before the failed restore.
If safe mode was disabled: you will need to restore from your most recent healthy backup.
Common Mid-Restore Failures
| Error message | Meaning | Fix |
|---|---|---|
No space left on device | Disk full during restore | Delete files on the server, then retry restore |
Permission denied | cloudpilot user cannot write to the document root | Fix file ownership: sudo chown -R cloudpilot:cloudpilot /var/www/... |
mysqldump: Got error: Access denied | MySQL credentials in backup don't match current DB | Manually update DB credentials after restore |
SSH connection dropped | Server SSH became unavailable mid-restore | Check SSH connectivity — see KB-12-05 — then retry |
ERROR 1005: Can't create table (errno: 150) | Foreign key constraint failure during DB import | Try with safeMode: false and a fresh DB, or manually fix the dump |
After a Failed Restore: Is the Site Working?
- Visit the site in a browser.
- If it loads normally: the restore recovered enough to be functional. Check for data integrity manually.
- If it shows a 502/500 error or blank page:
- Restart Nginx:
sudo systemctl restart nginx(via server terminal in CloudAIPilot). - Check the site's error log:
/var/log/nginx/error.logor/var/www/./logs/
- If the database is corrupt: restore from the pre-restore backup (see Step 3 above).