Ubuntu Deployment Guide

MariaDB AI RAG - Ubuntu Native Deployment Guide

Deploy MariaDB AI RAG .deb package directly on Ubuntu (without Docker)

---

Quick Start

# 1. Install MariaDB
sudo apt update && sudo apt install -y mariadb-server mariadb-client
sudo systemctl start mariadb && sudo systemctl enable mariadb

# 2. Secure MariaDB (set root password during setup)
sudo mysql_secure_installation

# 3. Create database
sudo mariadb -u root -p <<EOF
CREATE DATABASE kb_chunks CHARACTER SET utf8mb4 COLLATE utf8mb4_unicode_ci;
EXIT;
EOF

# 4. Install MariaDB AI RAG
sudo apt install -y ./ai-nexus.deb

# 5. Configure (update GEMINI_API_KEY)
cp /opt/rag-in-a-box/config/config.env.template /path/to/config.env
nano /path/to/config.env

# 6. Start services in their own terminals
/opt/rag-in-a-box/bin/rag-api --config /path/to/config.env
CONFIG_FILE=/path/to/config.env /opt/rag-in-a-box/bin/mcp-server

# 7. Verify
curl http://localhost:8000/health

---

Prerequisites

System Requirements

• OS: Ubuntu 22.04 LTS or 24.04 LTS (x86_64)

• CPU: 4+ cores (8+ recommended)

• RAM: 8+ GB (16+ recommended)

• Storage: 20+ GB free

• Access: Root/sudo privileges

Required

• Google Gemini API Key: Get from https://makersuite.google.com/app/apikey

Verify System

# Check Ubuntu version
lsb_release -a

# Check disk space
df -h /

# Check ports are free
sudo netstat -tuln | grep -E ':(8000|8002|3306)'
# No output = ports available

---

Step 1: Install MariaDB

# Update package lists
sudo apt update

# Install MariaDB
sudo apt install -y mariadb-server mariadb-client

# Start and enable MariaDB
sudo systemctl start mariadb
sudo systemctl enable mariadb

# Verify running
sudo systemctl status mariadb

---

Step 2: Secure MariaDB

sudo mysql_secure_installation

Follow prompts:

• Enter current password for root: [Press Enter]

• Switch to unix_socket authentication? n

• Change the root password? Y

  • New password: [Choose a secure password]

  • Re-enter: [Same password]

• Remove anonymous users? Y

• Disallow root login remotely? Y

• Remove test database? Y

• Reload privilege tables? Y

---

Step 3: Create Database

# Login to MariaDB
sudo mariadb -u root -p
# Enter password: [your_password]

In MariaDB shell:

-- Create database
CREATE DATABASE kb_chunks CHARACTER SET utf8mb4 COLLATE utf8mb4_unicode_ci;

-- Verify
SHOW DATABASES;

-- Exit
EXIT;

---

Step 4: Configure MariaDB

# Edit configuration
sudo nano /etc/mysql/mariadb.conf.d/50-server.cnf

Add under [mysqld] section:

character-set-server = utf8mb4
collation-server = utf8mb4_unicode_ci
innodb_page_size = 16k
innodb_default_row_format = dynamic
max_connections = 100
innodb_buffer_pool_size = 2G

Save and restart:

# Save: Ctrl+X, Y, Enter
sudo systemctl restart mariadb

---

Step 5: Install MariaDB AI RAG Package

# Navigate to directory with .deb file
cd ~/

# Install package
sudo apt install -y ./ai-nexus.deb

# If dependency issues:
sudo dpkg -i ai-nexus.deb
sudo apt-get install -f

Verify installation:

# Check installed files
dpkg -L ai-nexus | head -20

# Expected locations:
# /opt/rag-in-a-box/bin/rag-api
# /opt/rag-in-a-box/bin/mcp-server
# /opt/rag-in-a-box/config/config.env.template

# Check binaries exist
ls -lh /opt/rag-in-a-box/bin/

---

Step 6: Configure MariaDB AI RAG

# Edit configuration file
cp /opt/rag-in-a-box/config/config.env.template /path/to/config.env
nano /path/to/config.env

Update these essential settings:

# ===== DATABASE CONFIGURATION =====
DB_HOST=localhost
DB_PORT=3306
DB_USER=root
DB_PASSWORD=your_secure_database_password
DB_NAME=kb_chunks

# ===== API KEYS (REQUIRED - UPDATE THIS!) =====
GEMINI_API_KEY=your_actual_gemini_api_key_here

# ===== SECURITY KEYS (MUST BE IDENTICAL) =====
SECRET_KEY=your_generated_secret_key_must_be_same_for_all_three
JWT_SECRET_KEY=your_generated_secret_key_must_be_same_for_all_three
MCP_AUTH_SECRET_KEY=your_generated_secret_key_must_be_same_for_all_three

# ===== SERVER CONFIGURATION =====
APP_HOST=0.0.0.0
APP_PORT=8000
MCP_HOST=0.0.0.0
MCP_PORT=8002

# ===== EMBEDDING & LLM =====
EMBEDDING_PROVIDER=gemini
embedding_model=text-embedding-004
LLM_PROVIDER=gemini
LLM_MODEL=gemini-2.0-flash

# ===== TABLE NAMES =====
DOCUMENTS_TABLE=documents_DEMO_gemini
VDB_TABLE=vdb_tbl_DEMO_gemini

# ===== MCP CONFIGURATION =====
MCP_ENABLE_AUTH=true
MCP_ENABLE_VECTOR_TOOLS=true
MCP_ENABLE_DATABASE_TOOLS=true
MCP_ENABLE_RAG_TOOLS=true
MCP_READ_ONLY=false
MCP_LOG_LEVEL=INFO

# ===== PROCESSING =====
CHUNK_SIZE=512
CHUNK_OVERLAP=128
DOCUMENT_PROCESSING_BATCH_SIZE=5
EMBEDDING_BATCH_SIZE=32

Save: Ctrl+X, Y, Enter

---

Step 7: Start Services in their own terminals

# Start RAG API
/opt/rag-in-a-box/bin/rag-api --config /path/to/config.env

# Start MCP Server
CONFIG_FILE=/path/to/config.env /opt/rag-in-a-box/bin/mcp-server

Step 8: Verify Deployment

Check listening ports:

sudo netstat -tuln | grep -E ':(8000|8002)'

Should show LISTEN on both ports

Test Health Endpoints

# Test RAG API
curl http://localhost:8000/health
# Expected: {"status":"healthy","database":"connected"}

# Test MCP Server
curl http://localhost:8002/health
# Expected: {"status":"healthy"}

# Test API info
curl http://localhost:8000/

View Logs

Expected log messages:

INFO:     Started server process
INFO:     Waiting for application startup.
INFO:     Application startup complete.
INFO:     Uvicorn running on http://0.0.0.0:8000

---

Step 9: Test Functionality

Generate Authentication Token

# Generate token
curl -X POST "http://localhost:8000/token" \
  -H "Content-Type: application/json" \
  -d '{"username":"admin","password":"your_password"}'

# Save token for next commands
export TOKEN="eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9..."

Test Document Upload

# Create test document
echo "This is a test document for MariaDB AI RAG RAG system. It contains sample text for testing." > test_document.txt

# Upload document
curl -X POST "http://localhost:8000/documents/ingest" \
  -H "Authorization: Bearer $TOKEN" \
  -F "file=@test_document.txt"

# Expected output:
# {"document_id":1,"filename":"test_document.txt","chunks_created":1,"status":"success"}

Test RAG Query

# Query the document
curl -X POST "http://localhost:8000/orchestrate/generation" \
  -H "Authorization: Bearer $TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"query":"What is this document about?"}'

# Expected: AI-generated response with sources

Verify Database

# Login to MariaDB
mariadb -u root -p kb_chunks
# Enter password: [your_password]

In MariaDB:

-- Show tables
SHOW TABLES;

-- Check documents
SELECT id, filename, created_at FROM documents_DEMO_gemini;

-- Check embeddings
SELECT COUNT(*) FROM vdb_tbl_DEMO_gemini;

-- Exit
EXIT;

---

Access Points

After successful deployment:

• RAG API Swagger UI: http://<server-ip>:8000/docs

• RAG API Health: http://<server-ip>:8000/health

• MCP Server: http://<server-ip>:8002/mcp

• MCP Health: http://<server-ip>:8002/health

Get server IP:

hostname -I

---

Troubleshooting

Services Won't Start

Check logs in the terminal windows

Common causes:

1. MariaDB not running

sudo systemctl status mariadb
sudo systemctl start mariadb

1. Configuration errors

nano /path/to/config.env
# Check for typos, missing values

1. Port already in use

sudo lsof -i :8000
sudo lsof -i :8002
# Stop conflicting service or kill process

1. Permission issues

sudo chmod +x /opt/rag-in-a-box/bin/rag-api
sudo chmod +x /opt/rag-in-a-box/bin/mcp-server
sudo chmod 640 /opt/rag-in-a-box/config/config.env.template

Database Connection Fails

# Test MariaDB connection
mariadb -u root -p -e "SELECT 1;"

# Check MariaDB status
sudo systemctl status mariadb

# Restart MariaDB
sudo systemctl restart mariadb

# Check credentials in config
sudo grep DB_ /path/to/config.env

# View MariaDB logs
sudo tail -f /var/log/mysql/error.log

Authentication Fails

# Verify all three secret keys are identical
sudo grep SECRET_KEY /path/to/config.env

# Should show same value for:
# SECRET_KEY=...
# JWT_SECRET_KEY=...
# MCP_AUTH_SECRET_KEY=...

# If different, fix and restart
nano /path/to/config.env

API Key Invalid

# Test Gemini API key
API_KEY="YOUR_KEY"
curl -s "https://generativelanguage.googleapis.com/v1beta/models?key=$API_KEY"

# If invalid, update config
nano /path/to/config.env
# Update: GEMINI_API_KEY=...

# Restart services
/opt/rag-in-a-box/bin/rag-api --config /path/to/config.env
CONFIG_FILE=/path/to/config.env /opt/rag-in-a-box/bin/mcp-server

Port Already in Use

# Find process using port
sudo lsof -i :8000
sudo lsof -i :8002

# Kill process (if safe)
sudo kill <PID>

Out of Memory

# Check memory
free -h
top

# Add swap if needed (4GB example)
sudo fallocate -l 4G /swapfile
sudo chmod 600 /swapfile
sudo mkswap /swapfile
sudo swapon /swapfile

# Make permanent
echo '/swapfile none swap sw 0 0' | sudo tee -a /etc/fstab

---

Maintenance

Daily Operations

# Check service status
sudo systemctl status mariadb

# Test RAG API
curl http://localhost:8000/health
# Expected: {"status":"healthy","database":"connected"}

# Test MCP Server
curl http://localhost:8002/health
# Expected: {"status":"healthy"}

# Test API info
curl http://localhost:8000/

# Monitor disk space
df -h

Backup Database

# Backup
sudo mysqldump -u root -p kb_chunks > kb_chunks_backup_$(date +%Y%m%d).sql

# Compress backup
gzip kb_chunks_backup_$(date +%Y%m%d).sql

# Restore (if needed)
mariadb -u root -p kb_chunks < kb_chunks_backup_20241013.sql

Update Configuration

# Edit config
nano /path/to/config.env

# Restart services to apply changes
/opt/rag-in-a-box/bin/rag-api --config /path/to/config.env
CONFIG_FILE=/path/to/config.env /opt/rag-in-a-box/bin/mcp-server

Update MariaDB AI RAG

# Install new version
sudo apt install -y ./ai-nexus-new-version.deb

# Start services in their own terminals
/opt/rag-in-a-box/bin/rag-api --config /path/to/config.env
CONFIG_FILE=/path/to/config.env /opt/rag-in-a-box/bin/mcp-server

# Verify
curl http://localhost:8000/health

Optimize Database

mariadb -u root -p <<EOF
USE kb_chunks;
OPTIMIZE TABLE documents_DEMO_gemini;
OPTIMIZE TABLE vdb_tbl_DEMO_gemini;
EXIT;
EOF

---

Uninstall

# Remove package
sudo apt remove --purge rag-in-a-box

# Remove configuration (optional)
sudo rm -rf /opt/rag-in-a-box/

# Remove database (optional - ⚠️ deletes all data)
mariadb -u root -p -e "DROP DATABASE kb_chunks;"

---

Security Best Practices

Change Default Passwords

# Change MariaDB root password
sudo mariadb -u root -p

ALTER USER 'root'@'localhost' IDENTIFIED BY 'your_new_secure_password';
FLUSH PRIVILEGES;
EXIT;

Update config:

nano /path/to/config.env
# DB_PASSWORD=your_new_secure_password
/opt/rag-in-a-box/bin/rag-api --config /path/to/config.env
CONFIG_FILE=/path/to/config.env /opt/rag-in-a-box/bin/mcp-server

Generate New Secret Keys

# Generate secure key
python3 -c "import secrets; print(secrets.token_urlsafe(64))"

# Use same value for all three keys in config
nano /path/to/config.env

Configure Firewall

# Install UFW
sudo apt install -y ufw

# Allow SSH (IMPORTANT!)
sudo ufw allow 22/tcp

# Allow RAG API
sudo ufw allow 8000/tcp

# Allow MCP Server
sudo ufw allow 8002/tcp

# Enable firewall
sudo ufw enable

# Check status
sudo ufw status

Restrict Database Access

# Create dedicated database user
sudo mariadb -u root -p

CREATE USER 'rag_user'@'localhost' IDENTIFIED BY 'your_secure_password';
GRANT ALL PRIVILEGES ON kb_chunks.* TO 'rag_user'@'localhost';
FLUSH PRIVILEGES;
EXIT;

Update config:

nano /path/to/config.env
# DB_USER=rag_user
# DB_PASSWORD=your_secure_password

---

Quick Reference

Essential Commands

# Start services
/opt/rag-in-a-box/bin/rag-api --config /path/to/config.env
CONFIG_FILE=/path/to/config.env /opt/rag-in-a-box/bin/mcp-server

# Test health
curl http://localhost:8000/health

# Edit config
nano /path/to/config.env

# Database access
mariadb -u root -p kb_chunks

File Locations

/opt/rag-in-a-box/bin/rag-api                   # RAG API binary
/opt/rag-in-a-box/bin/mcp-server                # MCP Server binary
/opt/rag-in-a-box/config/config.env.template    # Configuration file
/var/log/mysql/error.log                        # MariaDB logs

Service Dependencies

MariaDB (Port 3306)
    ↓
RAG API (Port 8000)
    ↓
MCP Server (Port 8002)

Start order: MariaDB → RAG API → MCP Server Stop order: MCP Server → RAG API → MariaDB

---

Architecture Overview

Ubuntu System (Native)
├── MariaDB Service (systemd)
│   └── Database: kb_chunks (Port 3306)
├── RAG API Service (systemd)
│   └── FastAPI Server (Port 8000)
└── MCP Server Service (systemd)
    └── FastAPI Server (Port 8002)

---

Performance Tuning

MariaDB Optimization

sudo nano /etc/mysql/mariadb.conf.d/50-server.cnf

[mysqld]
# Adjust based on available RAM
innodb_buffer_pool_size = 4G  # 50-70% of RAM
max_connections = 200
innodb_log_file_size = 512M
query_cache_size = 0
query_cache_type = 0

System Resources

# Monitor resources
htop
# Or
top

# Check disk I/O
iostat -x 1

# Check network
iftop

---

Deployment Complete! 🎉

Your MariaDB AI RAG is now running natively on Ubuntu.

Next Steps:

1. Access Swagger UI: http://<server-ip>:8000/docs

2. Generate authentication token

3. Upload test documents

4. Start querying with RAG

For support:

• Check logs

• Verify config: nano /path/to/config.env

• Test health: curl http://localhost:8000/health

_{This page is: Copyright © 2025 MariaDB. All rights reserved.}