Monitoring

The monitoring stack provides full observability for the cluster:

• VictoriaMetrics — time-series metrics storage (Prometheus-compatible)
• VictoriaLogs — log storage (Loki-compatible)
• Grafana — dashboards and visualization
• OpenTelemetry Collector — metrics and log collection pipeline
• Metrics Server — resource metrics for kubectl top and HPAs
• argocd-monitor — ServiceMonitors for the Argo CD components

Alertmanager is not separate: it ships inside the VictoriaMetrics k8s-stack chart.

Data Flow

Nodes/Pods
    │
    ├── Container logs ──→ OTel Agent DaemonSet ──→ VictoriaLogs
    ├── Host metrics ────→ OTel Agent DaemonSet ──→ VictoriaMetrics
    ├── kubelet metrics ──→ OTel Agent DaemonSet ──→ VictoriaMetrics
    └── k8s events ─────→ OTel Gateway ──────────→ VictoriaLogs
                                                         │
                                               Grafana queries ◄──── User