Distributed Deployment & Load Balancing Guide

Overview

Gojang supports distributed deployment and load balancing to achieve high availability and horizontal scaling. Key requirements:

• PostgreSQL (not SQLite)
• Persistent session storage (Redis or DB)
• Load balancer (Nginx/HAProxy/Cloud LB)

Architecture Patterns

1. Simple load balancing (multiple instances on same server)
2. Multi-server deployment (different machines)
3. Container orchestration (Kubernetes)

Steps

• Configure shared PostgreSQL
• Use Redis or Postgres for sessions
• Configure load balancer (Nginx/HAProxy)
• Run multiple instances (systemd units or Docker compose scaled services)
• Add health check endpoint (/health)

Session Storage

Recommended: Redis (fast, battle-tested). Alternative: Postgres-backed sessions using scs/postgresstore.

Load Balancer Examples

• Nginx upstream configuration
• HAProxy config example
• Cloud load balancer guidance for ALB/GCP

Testing

• Verify all instances respond to /health
• Test session persistence across instances
• Test failover by stopping an instance

Best Practices

• Use the same SESSION_KEY and CSRF_SECRET across instances
• Avoid local file storage (use S3 or shared storage)
• Use connection pooling and monitor DB connections

---

This guide helps you scale Gojang horizontally and operate in production with multiple instances and a load balancer.