#+TITLE: Grey Area - Application Server
#+AUTHOR: Geir Okkenhaug Jerstad
#+DATE: [2025-06-04 Wed]

* Machine Overview

** Role
- **Primary Function**: Multi-purpose application server
- **Character**: Like the Culture GCU =Grey Area= - versatile, handles ambiguous/varied tasks
- **Core Service**: Forgejo (Git hosting and project management)

** Primary Services

*** Forgejo Git Hosting
- Self-hosted Git repositories
- Issue tracking and project management
- CI/CD pipelines
- User and organization management
- Pull request workflows
- Container registry (if needed)

*** Secondary Applications
- Jellyfin or Plex (media server)
- Nextcloud (file synchronization and sharing)
- Grafana (metrics visualization)
- Home Assistant (home automation, if desired)
- Various containerized web applications

** Architecture Notes
- Container-focused deployment (Podman/Docker)
- Headless operation (no desktop environment)
- SSH access for administration
- Reverse proxy integration with =reverse-proxy= machine
- Database support (PostgreSQL for Forgejo)

** Hardware Specifications
- **CPU**: Intel Xeon E5-2670 v3 (24 cores) @ 3.10 GHz
- **Memory**: 31.24 GiB RAM
- **Storage**: SSD for application data and Git repositories
- **Network**: Integration with =sleeper-service= for bulk storage
- **Performance**: Excellent for containerized applications and CI/CD workloads

** Resource Allocation Strategy
- **Forgejo**: 4-6 GB RAM, dedicated CPU cores for Git operations and CI/CD
- **PostgreSQL**: 2-4 GB RAM for database operations
- **Jellyfin**: 4-8 GB RAM for media transcoding (hardware acceleration if available)
- **Containers**: Remaining resources distributed across secondary applications
- **System Reserve**: 8-10 GB RAM kept free for system operations and burst capacity

** Integration Points
- **reverse-proxy**: SSL termination and external routing
- **sleeper-service**: Network storage for large files/backups
- **congenital-optimist**: Development workflow integration
- **All machines**: Centralized Git repository hosting

** Data Management
- Git repositories (local SSD for performance)
- Application databases (PostgreSQL, SQLite)
- Media files (network storage or local cache)
- Configuration backups to =sleeper-service=

** Security Considerations
- Container isolation
- User authentication (LDAP/OIDC integration possible)
- Regular security updates
- Backup and disaster recovery
- Network segmentation

** Performance Capabilities
With 24 cores and 31GB RAM, this machine can handle:
- **High-Concurrency Forgejo**: Multiple simultaneous Git operations, CI/CD builds
- **Media Transcoding**: Real-time 4K transcoding for Jellyfin without performance impact
- **Development Builds**: Parallel compilation of large codebases via CI/CD
- **Container Orchestration**: 10-20+ simultaneous containers without resource constraints
- **Database Performance**: PostgreSQL with excellent query performance and caching
- **Future Expansion**: Room for additional services like Kubernetes, monitoring stack