How ankit-goyal-git/fincore-monitoring-project works
Project Type and Purpose
• This is a full-stack monitoring & observability solution for a Dockerized Flask-based banking microservices application.
• It uses Prometheus to scrape metrics (application + system via Node Exporter), Grafana for dashboards and alerting, Alertmanager (via Grafana) for notifications, and Terraform to provision AWS infrastructure (EC2, VPC, remote state in S3).
File-Structure Analysis
• Top level:
• app.py + Dockerfile – Flask banking app container
• prometheus.yml – Prometheus scrape configuration
• grafana/ – Grafana config + dashboards + provisioning rules
• terraform-fincore/ – Terraform modules for AWS EC2, VPC, security groups, remote state (S3 backend)
• Patterns:
• Containerized microservices (Flask app, Prometheus, Node Exporter, Grafana) running on a single EC2 host via Docker
• IaC with Terraform and remote state in S3
• Grafana provisioning of dashboards, datasources, and alerts via mounted JSON/YAML
Core Components & Responsibilities
• Infrastructure (AWS)
• VPC, public subnet, Internet Gateway, Routing
• Security Group (ports 22, 3000, 5000, 9090, 9100)
• EC2 instance (user-data script installs Docker, clones the repo, starts containers)
• S3 bucket (Terraform remote state backend)
• Docker Host (EC2)
• Flask Banking App container (exposes /metrics, endpoints /balance, /transfer, /transactions)
• Node Exporter container (exposes host-level metrics on 9100)
• Prometheus container (scrapes app & node exporter on 9090)
• Grafana container (listens on 3000, auto-loads dashboards, alert rules, data source)
• Observability Pipeline
• Prometheus scrapes metrics and stores time-series data
• Grafana queries Prometheus via Prometheus-datasource
• Alert rules (configured in Grafana provisioning) trigger notifications via SMTP/Slack
Interactions and Data Flow
1. Terraform CLI → AWS APIs → Provision VPC, Subnet, SG, EC2, S3 backend
2. EC2 user-data → installs Docker & Git → clones repository → launches containers
3. Node Exporter & Flask App → /metrics endpoints → scraped by Prometheus
4. Prometheus data → queried by Grafana dashboards
5. Grafana alert rules → evaluate Prometheus metrics → send emails via SMTP (or Slack)
6. Terraform remote state → stored and versioned in S3
Architectural Patterns & Design Principles
• Separation of concerns: infrastructure (Terraform) vs. application code vs. observability tools
• Immutable infrastructure: Docker containers launched by user-data script
• Infrastructure as Code: Terraform modules, variables, remote state
• Continuous provisioning: Grafana auto-provisioning of dashboards & alerts
Final Checklist
• [ ] All major components identified & labeled
• [ ] Architectural patterns annotated (IaC, containerization, observability pipeline)
• [ ] Detail Terraform remote state and provisioning sequence