Monitoring

The monitoring stack runs as a Docker Compose profile and provides metrics collection, visualization, and alerting.

---

Starting the Stack

docker compose --profile monitoring up -d

This starts Prometheus, Grafana, Alertmanager, cAdvisor, Node Exporter, Redis Exporter, and Gotify (for push notifications).

---

Custom Metrics

The platform exposes 12 custom cm_* Prometheus metrics:

• API request rates and latencies
• BullMQ queue sizes (email, SMS, video scheduling)
• Active canvass sessions
• External service health gauges

---

Grafana Dashboards

Three pre-configured dashboards auto-provisioned from configs/grafana/:

• API Performance — request rates, latencies, error rates
• Application Overview — queue sizes, active sessions, service health
• System Health — container resources, host metrics, Redis stats

---

Alertmanager

Alert rules in configs/prometheus/alerts.yml cover:

• API downtime and high error rates
• Queue backlogs
• Service connectivity failures
• Resource utilization thresholds

---

Admin Routes

• /app/observability — embedded Grafana dashboards, alert status, and service health (3 tabs)

Direct Access

• Grafana: localhost:3005 or grafana.DOMAIN
• Prometheus: localhost:9090
• Alertmanager: localhost:9093