How shanmugapriya-st/slo-error-budget-demo works
This is a small SLO/error-budget monitoring demo: generated traffic exercises a deliberately unreliable Flask service, Prometheus scrapes its metrics and evaluates burn-rate alerts, and Grafana presents reliability and latency signals. Its other major capability is deploying the stack to AWS with Terraform. The README documents the main runtime links; configuration details and Terraform internals were not sampled, so those relationships are shown only where documented.
Main entry points: prometheus.yml Prometheus · load_gen.py Load generator · slo-dashboard.json Grafana dashboard · main.tf AWS deployment
Components 9
Traffic generation
- Load generator
app/load_gen.py
Demo service
- Flask service
app/app.py
Metrics and alerts
- Prometheus
prometheus/prometheus.yml - Burn-rate rules
prometheus/alert_rules.yml
Dashboards and deployment
- Grafana dashboard
grafana/dashboards/slo-dashboard.json - Grafana provisioning
grafana/provisioning - AWS deployment
terraform/main.tf
Other components
- Operator
- AWS EC2
Connections 8
- Operator → Load generator: starts traffic
README.md - Load generator → Flask service: sends requests
README.md - Prometheus → Flask service: scrapes metrics
README.md - Prometheus → Burn-rate rules: evaluates rules
README.md - Grafana dashboard → Prometheus: queries metrics
README.md - Grafana provisioning → Grafana dashboard: provisions dashboard
README.md - Operator → AWS deployment: runs deployment
README.md - AWS deployment → AWS EC2: provisions instance
README.md