Best Setup: Use Separate Checks for Uptime, System Load and Drive Health
One tool rarely does all three well. A practical ZimaOS stack is: Uptime Kuma for “is the service reachable?”, Netdata for CPU/RAM/free-space/temperature trends, and Scrutiny for disk SMART history. That gives you useful alerts without pretending that a ping check can predict a failing drive.
Use Uptime Kuma for App and Port Availability
Monitor HTTP endpoints, TCP ports, DNS or ping targets, then attach an SMTP/email notification channel. The Uptime Kuma hardware guide covers storage and monitor scaling. Upstream Uptime Kuma documents notifications and monitor types in its project documentation.
Use Netdata for Resource Thresholds
Free space, sustained CPU load, memory pressure and temperatures are continuous metrics—not simple up/down checks. Netdata's alerting engine is designed for threshold-based health checks. See the Netdata hardware guide and the Netdata notification documentation.
Use Scrutiny for SMART History, but Pass the Actual Drives Through
Scrutiny depends on smartctl seeing the physical disks. In Docker, that means the required --device mappings and capabilities. RAID controllers may need special device-type handling. The official Scrutiny documentation explains those requirements.
Define Alerts That Point to an Action
- Service unreachable for 2–5 minutes → check container/network.
- Filesystem >80–90% full → clean up or expand storage.
- Temperature above your hardware's safe operating target → inspect cooling/workload.
- SMART critical attributes or repeated device errors → back up and replace/diagnose the drive.
For a broader app sizing reference, use the ZimaOS App Store hardware directory.
