FaultPilot home
FaultPilot documentation
Uptime monitors
Configure public checks, confirmation thresholds, timeouts, maintenance, and reliable alert delivery.
Open in FaultPilot1
Choose the correct check
- Use HTTP or website monitoring for public web endpoints.
- Use SSL and domain-expiration checks for certificate and registration lifecycle visibility.
- Monitor a lightweight readiness endpoint when possible instead of a slow business transaction.
2
Configure reliability
- Set a timeout that reflects the endpoint's normal worst-case response time.
- Use consecutive failure and recovery thresholds to avoid alerts from one transient result.
- Treat expected authentication responses explicitly; avoid repeatedly testing a throttled login endpoint.
- FaultPilot suppresses customer-down alerts when its runner or outbound network cannot provide trustworthy evidence.
3
Connect operations
- Attach active email, Slack, or webhook destinations.
- Connect customer-visible monitors to a status page.
- Add planned work to a maintenance window before checks begin failing.
4
Interpret state
- UP means the latest confirmed state is healthy.
- DOWN requires the configured number of consecutive counted failures.
- UNKNOWN means FaultPilot cannot currently produce trustworthy evidence and should not blame the customer target.
- DEGRADED is reserved for partial or policy-defined impairment, not one arbitrary failed check.