Email when the box dies

What this email actually means

That email means the box stopped checking in. Period.

It does not mean Tailscale is healthy. It does not mean the camera app is fine. It does not mean the shop Wi‑Fi is up. One signal. Use it with power/path/auth triage, not instead of it.

Checklist: dead-man that actually shouts

1. Right box. Something you admin and can install a cron / scheduled task on. Cloud cameras ≠ the box.
2. Create a check. New Healthchecks (or equivalent) ping URL. Period that matches how fast you care (e.g. every 5-15 minutes; grace so a short blip doesn’t spam you).
3. Schedule the ping from the box. cron, Task Scheduler, or a tiny script that hits the URL while the machine is alive. Runs as a service user if the box reboots without a desktop login.
4. Wire your email. Confirm you get the “check created / verified” mail. Add a backup address if you have one.
5. Prove-it unplug. Unplug the box (or kill the ping job) on purpose. Wait for the alert. Plug it back in. If the mail never comes, you have hope, not a dead-man.
6. Write it down. Check name, who gets the mail, what “missed ping” means on your one-pager. Placeholders only: your accounts, your URLs.

What people mix up

• Heartbeat down → machine (or the ping job) stopped talking. Start with power / local life.
• Heartbeat up, Tailscale down → box is alive; remote path or auth is the problem (Tailscale after reboot).
• Both fine, cameras dark → different app. Fix that stack; don’t blame Tailscale.

More on sorting dead hardware vs Wi‑Fi vs auth: When the box goes dark.