Core Service Stack

• N8N (Root /): The primary workflow automation tool, accessible at the root path of the public IP. It connects seamlessly with the local Ollama instance for AI tasks.
• Ollama: The engine for serving LLMs (like Llama 3) directly on the VM's GPU, providing low-latency inference for both n8n and Open WebUI.
• Open WebUI : A modern, web-based chat interface used to interact directly with the Ollama-served LLMs.
• Postgres Database: Provides persistent, scalable data storage for the n8n application.
• Nginx Reverse Proxy: Manages TLS termination (HTTPS on port 443) and routes traffic securely to all services.

FABRIC Networking Configuration

• FabNetv4Ext (Public Access): Connects the slice to the external network, providing the public IP address necessary to access the Nginx proxy via port 443.
• FabNetv4 (Internal Interoperability): Ensures the VM has direct access to the standard FABRIC Measurement and Control Plane (MCP) network, enabling seamless connectivity to future FABRIC resources deployed on this subnet.
• Access URL: All services are accessed through the Nginx endpoint: https://<VM_PUBLIC_IP>:443/.

GPU Acceleration

The VM is specifically provisioned with a GPU to meet the computational demands of the Ollama service. This ensures rapid loading and execution of LLMs, making the overall automation stack highly responsive and efficient for AI-related tasks.