Flowise + OpenSearch Setup Guide

Flowise + OpenSearch Setup Guide 🚀

This guide provides a step-by-step installation of Flowise with OpenSearch on Ubuntu using Docker.

Simple Demo

📥 Download the Agent

You can download the agent from the official repository:
➡️ GitHub - Flowise OpenSearch Agent

---

Step 1: Install Required Dependencies

Before starting, install Docker if it’s not already installed.

sudo apt update && sudo apt upgrade -y
sudo apt install -y docker.io docker-compose curl

Ensure Docker is running:

sudo systemctl enable --now docker

---

Step 2: Create a Custom Docker Network

To ensure Flowise and OpenSearch can communicate, create a dedicated Docker network:

docker network create flowise-net

---

Step 3: Run OpenSearch

Start OpenSearch: (Create your own strong password - it is 100% required as Opensearch will **refuse to start if weakpassword is used)

docker run -d --name opensearch \
  --network flowise-net \
  -p 9200:9200 -p 9600:9600 \
  -e "discovery.type=single-node" \
  -e "plugins.ml_commons.only_run_on_ml_node=false" \
  -e 'OPENSEARCH_INITIAL_ADMIN_PASSWORD=StrongPass!123' \
  -e "DISABLE_SECURITY_PLUGIN=true" \
  -e "OPENSEARCH_JAVA_OPTS=-Xms2g -Xmx2g" \
  opensearchproject/opensearch:latest

✅ Fixes Applied:

• DISABLE_SECURITY_PLUGIN=true → Disables SSL so OpenSearch works with HTTP. For some reason Flowise refuses to work with HTTPS
• -Xms2g -Xmx2g → Prevents OpenSearch from crashing due to low memory.

Wait 2-3 minutes for OpenSearch to fully start before moving to the next step.

Check if OpenSearch is running:

docker ps

Verify OpenSearch responds:

curl -u 'admin:StrongPass!123' http://localhost:9200

---

Step 4: Run Flowise in the Same Network

Start Flowise:

docker run -d --name flowise \
  --network flowise-net \
  -p 3000:3000 \
  -v ~/.flowise:/root/.flowise \
  flowiseai/flowise:latest

Confirm Flowise is running:

docker ps

---

Step 5: Configure Flowise to Connect to OpenSearch

1. Open Flowise UI at http://localhost:3000
2. Go to Credentials > OpenSearch
3. Enter OpenSearch Details:
   • Host: http://opensearch:9200 (HTTP, not HTTPS)
   • Username: admin
   • Password: StrongPass!123 (use your configured password)
   • Index Name: flowise_vectors
4. Save & Restart Flowise.

---

🚀 Success! Now You Can Start Using Flowise & OpenSearch

Flowise is now connected to OpenSearch and ready for vector-based AI workflows.

👉 Open Flowise at http://localhost:3000 and start building! 🔥

📥 Download the Agent

You can download the agent from the official repository:
➡️ GitHub - Flowise OpenSearch Agent

---

---