n8n Installation & Setup

Complete guide to installing and configuring n8n for Automatización Lurwis

n8n Installation & Setup

This guide covers installing n8n (self-hosted or cloud) and configuring it for the Automatización Lurwis platform.

Prerequisites

Before installing n8n, ensure you have:

• Node.js 18+ (for self-hosted)
• PostgreSQL 12+ database
• MongoDB 4.4+ for conversation memory
• Redis 6+ for message buffering
• WhatsApp Business API credentials
• Google Gemini API key

---

Installation Options

Self-Hosted (Recommended)

Docker Installation

The recommended approach for production deployments:

1. Create docker-compose.yml

   Create a docker-compose.yml file with the following configuration:

   version: '3.8'
   
   services:
     n8n:
       image: n8nio/n8n:latest
       container_name: n8n-lurwis
       restart: unless-stopped
       ports:
         - "5678:5678"
       environment:
         - N8N_BASIC_AUTH_ACTIVE=true
         - N8N_BASIC_AUTH_USER=${N8N_USER}
         - N8N_BASIC_AUTH_PASSWORD=${N8N_PASSWORD}
         - N8N_HOST=${N8N_HOST}
         - N8N_PORT=5678
         - N8N_PROTOCOL=https
         - WEBHOOK_URL=${WEBHOOK_URL}
         - EXECUTIONS_PROCESS=main
         - N8N_METRICS=true
       volumes:
         - n8n_data:/home/node/.n8n
         - ./backups:/backups
   
   volumes:
     n8n_data:
       driver: local
   
2. Configure environment variables

   Create a .env file:

   N8N_USER=admin
   N8N_PASSWORD=your-secure-password
   N8N_HOST=your-domain.com
   WEBHOOK_URL=https://your-domain.com
   

   Use a strong password and keep your .env file secure. Never commit it to version control.

3. Start n8n

   docker-compose up -d
   

   Access n8n at http://localhost:5678

4. Configure reverse proxy (Production)

   For production, use Caddy or nginx as a reverse proxy with SSL:

   Caddy example:

   your-domain.com {
       reverse_proxy n8n:5678
   }
   

   The workflows use Caddy as shown in the webhook configuration.

NPM Installation

For local development:

npm install -g n8n
n8n start

n8n Cloud

n8n Cloud Setup

1. Create account

   1. Go to n8n.cloud
   2. Sign up for an account
   3. Choose a plan (Pro recommended for production)
2. Configure workspace

   • Set your workspace name: Lurwis Production
   • Note your workspace URL (needed for webhooks)
3. Set environment variables

   In n8n Cloud settings, add environment variables:

   • See Environment Variables page

n8n Cloud automatically handles scaling, backups, and SSL certificates.

---

Workspace Configuration

Import Workflows

After installation, import the two main workflows:

1. Navigate to Workflows

   In n8n interface, click Workflows → Import from File

2. Import Receptor workflow

   Upload: workflows/Picantería Lurwis _ Receptor.json

   This workflow handles:

   • WhatsApp webhook verification
   • Message reception and validation
   • Redis buffering (6-8 second window)
   • Error handling
3. Import Procesador workflow

   Upload: workflows/Picantería Lurwis _ Procesador.json

   This workflow handles:

   • Message classification
   • AI agent routing
   • Order processing
   • Database operations
4. Activate workflows

   Click Active toggle for both workflows

---

Credential Management

Configure credentials for all integrations:

WhatsApp Business API

1. Add WhatsApp credential

   1. Go to Credentials → New
   2. Select WhatsApp
   3. Enter:
      • Access Token: Your Meta access token
      • Phone Number ID: 947279508470714 (from workflow)
2. Configure webhook

   • Webhook URL: https://your-domain.com/webhook/meta-verify
   • Verify Token: meta-verify

PostgreSQL Database

1. Add Postgres credential

   Name: Postgres Lurwis db

   Host: your-postgres-host
   Database: picanteria_db
   User: your-user
   Password: your-password
   Port: 5432
   SSL: Require (production)
   
2. Connection string format

   The workflows use session pooler connection:

   postgresql://user:password@host:5432/picanteria_db?sslmode=require
   

The workflows note mentions “CONNECTION STRING (SESSION POOLER)” with RLS (Row Level Security) enabled.

MongoDB Memory

1. Add MongoDB credential

   Name: Memoria chats Lurwis

   Connection String: mongodb+srv://user:[e-mail]/
   Database: picanteria_db
   
2. Collections used

   The system uses separate collections for conversation memory:

   • historial_clasificador - Message classification
   • historial_detector - Order detection
   • historial_pedidos - Order conversations (25 message window)
   • historial_eventos - Event bookings (15 message window)
   • historial_reservas - Table reservations (15 message window)
   • historial_general - General inquiries (10 message window)

Redis Buffer

1. Add Redis credential

   Name: Buffer Lurwis

   Host: your-redis-host
   Port: 6379
   Password: your-redis-password
   Database: 0
   
2. TTL configuration

   Redis keys use automatic expiration:

   • Buffer messages: 30 seconds
   • Timestamps: 30 seconds
   • Metadata: 120 seconds

Google Gemini AI

1. Add Google Gemini credential

   Name: Modelo Lurwis

   API Key: your-gemini-api-key
   
2. Get API key

   1. Go to Google AI Studio
   2. Create new API key
   3. Copy and save securely

---

Workflow Settings

Execution Settings

Both workflows use executionOrder: v1 for sequential processing.

Error Workflow

The workflows reference an error notification workflow that sends failures to a personal WhatsApp number. Configure this separately for production monitoring.

To create an error workflow:

1. Create new workflow named “Error Notifications”
2. Add Error Trigger node
3. Add WhatsApp node to send error details
4. In main workflows, go to Settings → Error Workflow → Select “Error Notifications”

---

Webhook Configuration

Meta WhatsApp Verification

The Receptor workflow handles Meta’s webhook verification:

• Path: /webhook/meta-verify
• Method: GET and POST
• Verify Token: meta-verify
• Response Mode: Response Node (for proper Meta acknowledgment)

1. Configure in Meta Developer Console

   1. Go to WhatsApp Business API settings
   2. Add webhook URL: https://your-domain.com/webhook/meta-verify
   3. Verify token: meta-verify
   4. Subscribe to: messages field
2. Test webhook

   Meta will send a verification request. The workflow returns:

   Status: 200
   Body: hub.challenge value
   

---

Performance Tuning

Execution Limits

EXECUTIONS_DATA_PRUNE=true
EXECUTIONS_DATA_MAX_AGE=336  # 14 days
N8N_CONCURRENCY=10

Memory Settings

For high-traffic production:

NODE_OPTIONS=--max-old-space-size=4096

---

Next Steps

Environment Variables

Configure all required environment variables

Production Checklist

Review production deployment requirements

Database Schema

Understand the database structure

Troubleshooting

Common issues and solutions