Write a simple disaster-recovery plan

Backups answer “can I get the data back?” A disaster-recovery (DR) plan answers “how fast can the whole site be running again, and who does what?” You can write a useful one in an afternoon.

Two numbers to decide

• RPO (Recovery Point Objective) — how much data you can afford to lose. If you back up daily, your RPO is up to 24 hours. Need less? Back up more often.
• RTO (Recovery Time Objective) — how long you can be down. This drives how rehearsed and automated your restore needs to be.

Write these down per site — they tell you if your current backup schedule is enough.

What to protect

• Files — site web roots, uploads, config.
• Databases — dumps of every database.
• The restic repository password — store it somewhere separate and durable. Without it, backups can’t be restored.
• DNS + domain access — know where your domain is managed so you can repoint it.
• Credentials — provider logins, in a password manager.

The rebuild runbook

Write the actual steps so anyone can follow them under pressure:

1. Provision a new server on a supported OS and install OrbixPanel.
2. Restore files and databases from the latest snapshot → restore from backup.
3. Reconnect apps to their databases; recreate email and cron.
4. Repoint DNS to the new server; issue SSL.
5. Verify with the post-migration checklist.

This is essentially a server move done under time pressure — which is why rehearsing it matters.

Rehearse it

• Test a restore monthly so step 2 is routine.
• Keep the runbook updated as your setup changes.
• Store the plan + passwords somewhere reachable even if the server is gone.

The worst time to learn your backups are incomplete is during an actual disaster. Rehearse while it’s calm.

---

Running OrbixPanel? Encrypted offsite backups + a one-command reinstall make DR realistic. OrbixPanel.com → — the free, self-hosted, AI-powered hosting control panel.