PostgreSQL with the PGvector extension for RAG: Added Docker Compose configuration

docker-compose/withPostgresPGVectorExtension/.env

@@ -0,0 +1,6 @@
+POSTGRES_USER=changeUser
+POSTGRES_PASSWORD=changePassword
+POSTGRES_DB=n8n
+
+POSTGRES_NON_ROOT_USER=changeUser
+POSTGRES_NON_ROOT_PASSWORD=changePassword

docker-compose/withPostgresPGVectorExtension/README.md

@@ -0,0 +1,70 @@
+# n8n with PostgreSQL and pgvector Extension
+
+Starts n8n with PostgreSQL as database, including the PGvector extension for vector similarity searches.
+
+## Features
+
+- PostgreSQL database for n8n
+- PGvector extension for storing and querying vector embeddings
+
+## What's Possible?
+
+This Docker Compose setup delivers a ready-to-use PostgreSQL database with **PGvector** extension, enabling powerful vector-based AI capabilities right out of the box.
+
+### Example: RAG Workflow with pgvector
+
+The diagram below shows a sample n8n workflow that demonstrates Retrieval-Augmented Generation (RAG) capabilities using the PostgreSQL/pgvector backend:
+
+![AI-Powered Semantic Search Workflow](./rag_pg_vector_workflow_example.png)
+
+This example workflow illustrates:
+- Document embeddings storage using Google Gemini embeddings 001
+- Semantic similarity searches against your vector database
+- AI reasoning with Google Gemini Flash 2.0 as the LLM
+- A complete RAG pipeline implementation with minimal configuration
+
+## Start
+
+To start n8n with PostgreSQL and pgvector extension, simply run docker-compose by executing the following command in the current folder.
+
+**IMPORTANT:** But before you do that change the default users and passwords in the [`.env`](.env) file!
+
+```
+docker-compose up -d
+```
+
+To stop it execute:
+
+```
+docker-compose stop
+```
+
+## Configuration
+
+The default name of the database, user and password for PostgreSQL can be changed in the [`.env`](.env) file in the current directory.
+
+## PGVector Store Information
+
+This setup automatically creates:
+
+1. The pgvector extension in your PostgreSQL database
+2. An embeddings table with the following schema:
+
+```sql
+CREATE TABLE embeddings (
+  id SERIAL PRIMARY KEY,
+  embedding vector,
+  text text,
+  created_at timestamptz DEFAULT now()
+);
+```
+
+You can use this table to store vector embeddings from AI models for semantic search, similarity comparison, and other vector-based operations in your n8n workflows.
+
+## Accessing PostgreSQL with pgvector
+
+The PostgreSQL instance is exposed on port 15432 (to avoid conflicts with any local PostgreSQL installations). You can connect to it using:
+
+```
+psql -h localhost -p 15432 -U [POSTGRES_USER] -d [POSTGRES_DB]
+```

docker-compose/withPostgresPGVectorExtension/docker-compose.yml

@@ -0,0 +1,55 @@
+version: '3.8'
+
+volumes:
+  pgvector_storage:
+  n8n_storage:
+
+networks:
+  n8n-network:
+    driver: bridge
+
+services:
+  postgres-pgvector:
+    image: pgvector/pgvector:pg17
+    container_name: postgres-pgvector
+    ports:
+      - 15432:5432
+    restart: always
+    environment:
+      - POSTGRES_USER
+      - POSTGRES_PASSWORD
+      - POSTGRES_DB
+      - POSTGRES_NON_ROOT_USER
+      - POSTGRES_NON_ROOT_PASSWORD
+    volumes:
+      - pgvector_storage:/var/lib/postgresql/data
+      - ./init-data.sh:/docker-entrypoint-initdb.d/init-data.sh
+    healthcheck:
+      test: ['CMD-SHELL', 'pg_isready -h localhost -U ${POSTGRES_USER} -d ${POSTGRES_DB}']
+      interval: 5s
+      timeout: 5s
+      retries: 10
+    networks:
+      - n8n-network
+
+  n8n:
+    image: docker.n8n.io/n8nio/n8n
+    restart: always
+    environment:
+      - DB_TYPE=postgresdb
+      - DB_POSTGRESDB_HOST=postgres-pgvector
+      - DB_POSTGRESDB_PORT=5432
+      - DB_POSTGRESDB_DATABASE=${POSTGRES_DB}
+      - DB_POSTGRESDB_USER=${POSTGRES_NON_ROOT_USER}
+      - DB_POSTGRESDB_PASSWORD=${POSTGRES_NON_ROOT_PASSWORD}
+    ports:
+      - 5678:5678
+    links:
+      - postgres-pgvector
+    volumes:
+      - n8n_storage:/home/node/.n8n
+    depends_on:
+      postgres-pgvector:
+        condition: service_healthy
+    networks:
+      - n8n-network

docker-compose/withPostgresPGVectorExtension/init-data.sh

@@ -0,0 +1,31 @@
+#!/bin/bash
+set -e;
+
+
+if [ -n "${POSTGRES_NON_ROOT_USER:-}" ] && [ -n "${POSTGRES_NON_ROOT_PASSWORD:-}" ]; then
+	psql -v ON_ERROR_STOP=1 --username "$POSTGRES_USER" --dbname "$POSTGRES_DB" <<-EOSQL
+		CREATE USER ${POSTGRES_NON_ROOT_USER} WITH PASSWORD '${POSTGRES_NON_ROOT_PASSWORD}';
+		GRANT ALL PRIVILEGES ON DATABASE ${POSTGRES_DB} TO ${POSTGRES_NON_ROOT_USER};
+		GRANT CREATE ON SCHEMA public TO ${POSTGRES_NON_ROOT_USER};
+	EOSQL
+
+	# Create vector extension and embeddings table
+	echo "Creating vector extension and embeddings table..."
+	if ! psql -v ON_ERROR_STOP=1 --username "$POSTGRES_USER" --dbname "$POSTGRES_DB" <<-EOSQL
+		CREATE EXTENSION IF NOT EXISTS vector;
+
+		CREATE TABLE IF NOT EXISTS embeddings (
+		  id SERIAL PRIMARY KEY,
+		  embedding vector,
+		  text text,
+		  created_at timestamptz DEFAULT now()
+		);
+	EOSQL
+	then
+		echo "ERROR: Failed to create vector extension or embeddings table"
+		exit 1
+	fi
+	echo "Vector extension and embeddings table created successfully"
+else
+	echo "SETUP INFO: No Environment variables given!"
+fi