Appendix A - Setting Up Alert Channels

Overview

The observability stack gives you full visibility into the cluster and application health. Three tools work together:

Prometheus          - collects and stores metrics
    ↓
Alertmanager        - receives alerts from Prometheus, routes to channels
    ↓
Grafana             - visualizes metrics as dashboards

All three are deployed via kube-prometheus-stack Helm chart - one install, everything included.

What Gets Monitored

• Cluster level - node CPU, memory, disk, network
• Kubernetes level - pod restarts, deployment health, HPA scaling events
• Application level - FinTrack-specific alerts (backend down, high memory, crash loops)

Accessing the Tools

Grafana

kubectl port-forward svc/kube-prometheus-stack-grafana -n monitoring 3000:80

Open: http://localhost:3000

• Username: admin
• Password: fintrack-grafana-2025

Useful dashboards to explore:

• Kubernetes / Compute Resources / Namespace (Pods) - filter to fintrack