Implementing lightweight interface monitoring

Personal projects have few APIs but occasionally fail silently. Manually checking daily isn't feasible, so I automated it as a scheduled task.

Tasks request the target URL at fixed intervals, recording status codes and response times. Alerts are triggered only after consecutive failures to avoid false positives from transient fluctuations.

Notifications use standard channels and include the address, failure count, and last response, so you can troubleshoot without logging into the server to check logs.

This approach doesn't aim for comprehensive coverage but ensures the most critical issues are caught early, while keeping maintenance costs low.