Smart Alert System
Get notified when it matters — before problems become outages.
How Alert Rules Work
Define thresholds for any metric — CPU, memory, disk, load, or device-specific values. A background cron job evaluates every rule on a regular cycle, transitioning alerts through states: OK, warning, critical, and firing.
- Threshold-based rules for any metric
- Cron-based evaluation cycle
- Alert states: OK, warning, critical, firing
- Automatic state transitions
Alert Types
Server Scout covers all the critical failure modes — from server resource exhaustion to device-specific issues like switch ports going down or UPS batteries running low.
- CPU, memory, disk, and load thresholds
- Server offline detection
- Service failure alerts (systemd)
- Device alerts: port down, battery low, temperature high
Automatic Recovery Detection
When a metric returns to normal, Server Scout automatically resolves the alert and sends a recovery notification. No manual intervention needed — the full lifecycle from firing to recovered is handled for you.
- Automatic resolution when metrics normalize
- Recovery notifications sent to your team
- Full alert history with timestamps
- No stale alerts cluttering your dashboard
Configurable Thresholds
Every server and device can have its own threshold settings. Set global defaults, then override per-server where needed. A database server might tolerate 80% memory usage, while a web server should alert at 70%.
- Per-server and per-device thresholds
- Global defaults with local overrides
- Separate warning and critical levels
- Easy tuning from the dashboard UI