Monitoring

This repository provides two modular services for system telemetry and monitoring:

Metrics Exporters (services.metrics-exporter)

Defined in modules/services/metrics-exporter.nix. Enabled by default on all hosts via modules/hosts/_common.nix.

• node-exporter: Exposes host OS system metrics (CPU, RAM, disk, systemd unit health) on http://127.0.0.1:9100/metrics.
• smartctl-exporter: Exposes S.M.A.R.T disk statistics and health metrics on http://127.0.0.1:9633/metrics.

Optionally, specific devices can be set using:

services.metrics-exporter.smartctlDevices = [ "/dev/nvme0n1" ];

Monitoring Stack (services.monitoring)

Defined in modules/services/monitoring.nix. Enabled on server/monitoring hosts (currently ixo).

• Prometheus: Listens on http://127.0.0.1:9090 and scrapes node and smartctl exporter targets.
• Grafana: Listens on http://127.0.0.1:3000 for visual dashboarding.