Monitoring
Monitoring
Prometheus collects recent metrics and Grafana supplies dashboards and email-capable visualization.
| Source | What it contributes |
|---|---|
| Node exporter | Host CPU, memory, filesystem, and operating-system metrics |
| cAdvisor | Container CPU, memory, disk I/O, network, and OOM events |
| Traefik | Edge request and routing metrics |
| CrowdSec | Traffic-screening metrics |
| TrueNAS exporter | Pool health, capacity, disk temperatures, scrub state, vdev errors, and alerts |
| FreePBX exporter | PBX and Asterisk metrics |
| Prometheus | Collection health |
General host and container metrics are collected approximately once a minute. TrueNAS is polled less frequently because its storage API changes more slowly. Prometheus retention is bounded at fourteen days or five gigabytes.
Grafana dashboards cover the overall environment, TrueNAS, CrowdSec, and FreePBX. SMTP delivery is configured, but alert rules and notification policies are not currently provisioned in Git. The present state is therefore monitoring with email capability—not guaranteed end-to-end alerts.