Metrics

Metrics

The controller exposes Prometheus metrics at GET /metrics (gated by metrics:read, served at the root path, not under /api/v1). This page lists the metric names; for the alerting, scrape, and dashboard artifacts see Observability.

Nodes

Metric	Meaning
rakkr_node_online	Whether a node is currently online.
rakkr_node_offline_alerts_active	Active node-offline alerts.

Audio input quality

Per-node/interface audio quality, derived from meter frames:

Metric	Meaning
rakkr_input_rms_dbfs	Input RMS level (dBFS).
rakkr_input_peak_dbfs	Input peak level (dBFS).
rakkr_input_clipping_ratio	Fraction of samples clipping.
rakkr_input_speech_score	Speech-presence score.
rakkr_input_noise_score	Noise score.
rakkr_input_broadband_noise_score	Broadband-noise score.
rakkr_input_estimated_snr_db	Estimated signal-to-noise ratio (dB).
rakkr_input_intelligibility_score	First-pass intelligibility score.

Live listen monitor

Metric	Meaning
rakkr_listen_monitor_chunk_age_seconds	Age of the latest monitor audio chunk.
rakkr_listen_monitor_chunk_duration_seconds	Duration of the latest chunk.

Recordings & jobs

Metric	Meaning
rakkr_recording_active	Active recordings.
rakkr_recording_duration_seconds	Recording duration.
rakkr_recording_bytes_written	Bytes written for a recording.
rakkr_recording_watchdog_alerts_total	Watchdog alerts raised.
rakkr_device_xruns_total	Audio device xruns.

Uploads

Metric	Meaning
rakkr_upload_queue_depth	Items waiting in the upload queue.
rakkr_upload_queue_oldest_due_seconds	Age of the oldest due queue item.
rakkr_upload_failures_total	Upload failures.

Events

Metric	Meaning
rakkr_audit_events_total	Audit events recorded.
rakkr_health_events_total	Health events recorded.

What’s exported overall

The exposition covers controller availability, node status, recording duration/cache bytes, meter quality, listen-monitor freshness, audit and health-event totals/active counts, watchdog alerts, xruns, and upload queue depth/overdue/failures. The checked Prometheus alert rules, Mimir remote-write example, and Grafana dashboard live under docs/observability/.