Healthchecks.io popularised the inverted model: your job pings them, and they alert when the pings stop. Brilliant for cron jobs — but it relies on your system being honest about its own health, which is exactly what fails during VPN outages.
What Healthchecks.io genuinely does well
- Elegant, simple dead-man's-switch model
- Very affordable at scale
- Open source with a strong self-host community
| Feature | Healthchecks.io | ExitProbe |
|---|---|---|
| Monitoring model | Your server reports in (trust-based) | We actively verify from outside (zero-trust) |
| Works when the monitored box is down | Alerts after silence threshold | Detects on next check interval |
| Tunnel-level visibility | None | Handshake + exit IP per protocol |
| False-negative risk | A hung tunnel that keeps pinging looks healthy | Traffic must actually flow through the tunnel |
| Multi-region verification | Single ingestion endpoint | Distributed probes worldwide |
| VPN config management | N/A | Encrypted vault with AES-256-GCM |
Bottom line
Healthchecks.io answers "did my cron run?". ExitProbe answers "can a user in Frankfurt actually reach the internet through my Tokyo node right now?" Different questions, different tools.
See the difference yourself
Add one VPN server on the free plan and watch real handshakes come in.