Monitoring
Metrics, health states, intervals and manual refresh.
What is collected
CPU usage, memory usage, disk usage and uptime, via read-only commands appropriate to each OS. If a value cannot be read, it is shown as a dash; unknown is never zero.
Health states
- Healthy: reachable and within normal thresholds.
- Attention: something is elevated and worth a look.
- Critical: a metric crossed a critical threshold.
- Offline: the server can't be reached right now.
Cadence
Each server refreshes automatically on its configured interval, with retries for transient failures. The refresh button on each card answers “now”, immediately.
