Taiga on AWS — Production Infrastructure

Production deployment of Taiga (open-source Agile PM) on AWS with Docker Swarm Multi-AZ, Terraform, Ansible, full CI/CD pipeline and Prometheus/Grafana monitoring.

AWSTerraformAnsibleDocker SwarmGitLab CI/CDTraefikPrometheusGrafanaPostgreSQL

Live ↗ GitHub ↗ GitLab ↗

Overview

Automated deployment of Taiga on AWS with high availability architecture across 2 availability zones.

Architecture

3 Masters running Traefik (reverse proxy + TLS) and Prometheus/Grafana
6 Workers running the Taiga application stack (Docker Swarm)
RDS PostgreSQL Multi-AZ for the database
EFS for shared static/media files across workers
NLB for cross-zone load balancing
S3 for automated daily backups

CI/CD Pipeline

dev → main       →  Integration tests (pytest) + SonarCloud analysis
main → staging   →  Docker build → push :staging → deploy to staging
staging → prod   →  Docker build → push :latest  → deploy to production

Key Features

One-command deployment: ./deploy.sh prod
Automated backups to S3 (daily cron, 30-day retention)
Full monitoring: CPU, RAM, disk, HTTP metrics, container metrics
Automatic TLS certificate via ACM + Route53
SSH access via bastion with custom port (security hardening)