Deployment Overview of North MiniCode on Server¶

Prerequisites and Basic Requirements¶

To ensure the successful deployment of North MiniCode, the server must meet the following requirements:

Operating System: Ubuntu (recommended 22.04 LTS).
Privileges: Root or sudo access is required for package installation and service management.
Hardware Requirement: An NVIDIA GPU is required for hardware acceleration.
Network Ports:
443/TCP: External HTTPS traffic.
8080/TCP: Internal application communication.
11434/TCP: Ollama API service.

The application is accessible via a specific subdomain template based on the server ID.

Parameter	Value
Prefix	`north-mini`
Domain	`hostkey.in`
Full template	`north-mini{Server_ID}.hostkey.in`

The following directories are used for configuration, certificates, and application data:

/root/nginx: Contains the Nginx Docker Compose configuration.
/data/nginx/user_conf.d: Stores Nginx site configuration files.
/data/nginx/nginx-certbot.env: Environment file for Certbot configurations.
open-webui (Docker Volume): Persistent storage for Open WebUI application data.

The installation follows a multi-stage process involving system preparation, driver installation, and container deployment:

System Update: The package cache is updated, and the system is upgraded to the latest distributed version.
Kernel Update: The linux-generic-hwe-22.04 kernel is installed to ensure hardware compatibility.
NVIDIA Driver Setup:
The system checks for NVIDIA GPU presence via lspci.
If a driver is not detected, the recommended NVIDIA drivers are automatically installed using ubuntu-drivers autoinstall.
Ollama Installation:
Ollama is installed via the official installation script: curl -fsSL https://ollama.com/install.sh | sh.
The service is enabled and started via systemd.
The specific model north-mini-code-1.0 is pulled automatically.
Docker Environment: Docker is installed, and the Python Docker module is added to the system.
Container Deployment: The Open WebUI container is launched using the host network mode to communicate with the local Ollama service.

Firewall: Ensure that port 443 is open for external HTTPS access.
Service Isolation: The application runs in a Docker container, isolating the web interface from the core LLM engine (Ollama).
SSL/TLS: SSL certificates are managed via Nginx and Certbot to secure all incoming connections on port 443.

The application utilizes local storage for its data:

Open WebUI Data: Managed within a Docker volume named open-webui located at /app/backend/data inside the container.

The deployment consists of two primary containers:

Image Name: jonasal/nginx-certbot:latest
Ports: Uses host network mode (mapping to port 443 for HTTPS).
Volumes:
nginx_secrets:/etc/letsencrypt
/data/nginx/user_conf.d:/etc/nginx/user_conf.d
Restart Policy: unless-stopped
Purpose: Handles SSL termination and proxies traffic to the application on port 8080.

The following setup actions are performed during the installation process:

Model Download: The script automatically executes ollama pull north-mini-code-1.0 to ensure the required LLM is available immediately after service startup.
Nginx Configuration Injection: A configuration file is dynamically updated in /data/nginx/user_conf.d/ to include a proxy_pass directive pointing to http://127.0.0.1:8080.

To update the main application (Open WebUI), execute the following commands:

docker compose pull && docker compose up -d

Note: This updates the container image while preserving data stored in the open-webui volume.