Setting Up Weaviate with Docker on macOS

This guide explains how to set up and run Weaviate in a Docker container on macOS. It also covers installing dependencies and verifying the setup.

Prerequisites

Ensure you have the following installed on your macOS system:

Docker (Download Here)
Python 3.8+ (Download Here)

To check if Docker is installed, run:

docker --version

If Docker is not installed, download and install it from the link above.

Step 1: Start Weaviate Using Docker

Run the following command to pull and start Weaviate with OpenAI text vectorization:

docker run -d --name weaviate \
  -e PERSISTENCE_DATA_PATH=/var/lib/weaviate \
  -e OPENAI_APIKEY= "<your-openai-key>" \
  -e ENABLE_MODULES=text2vec-openai \
  -e QUERY_DEFAULTS_LIMIT=25 \
  -e AUTHENTICATION_ANONYMOUS_ACCESS_ENABLED=true \
  -p 8080:8080 -p 50051:50051 semitechnologies/weaviate:latest

Replace <your-openai-key> with your OpenAI API key.

To verify Weaviate is running, execute:

docker ps | grep weaviate

You should see a running Weaviate container.

Step 2: Install Python Dependencies

To interact with Weaviate, install the required Python packages:

pip install weaviate-client

To verify installation:

python -c "import weaviate; print('Weaviate client installed successfully')"

Step 3: Test Connection to Weaviate

Create a Python script (test_weaviate.py) to verify connectivity:

import weaviate

client = weaviate.Client("http://localhost:8080")

if client.is_ready():
    print("✅ Weaviate is ready!")
else:
    print("❌ Weaviate is not running. Check Docker container.")

Run the script:

python test_weaviate.py

If successful, you should see:

✅ Weaviate is ready!

Step 4: Stop and Restart Weaviate

To stop Weaviate, run:

docker stop weaviate

To restart Weaviate, run the original docker run command again.

Troubleshooting

1. Docker is not running

Run:

open -a Docker

Wait for Docker to fully start, then retry running Weaviate.

2. Port 8080 is in use

Check if another process is using port 8080:

lsof -i :8080

If occupied, stop the process or run Weaviate on a different port (-p 9090:8080).

Clean up Docker

How to Clean Up Docker When You're Done

If you're finished with Weaviate or other Docker containers, follow these steps to clean up your system and free up disk space.

🧹 Step 1: Stop and Remove the Weaviate Container

Find the running Weaviate container:
```
docker ps
```
- Look for a container with the name or image related to Weaviate (e.g., weaviate/weaviate).
Stop the Weaviate container:
```
docker stop <container_id>
```
- Replace <container_id> with the actual container ID from the docker ps command.
Remove the stopped Weaviate container:
```
docker rm <container_id>
```
- This removes the container instance but not the image.

🗑️ Step 2: Remove Unused Docker Images

If you want to remove the Weaviate image, run:

docker images

Find the Weaviate image ID and remove it:

docker rmi <image_id>

🛠️ Step 3: Remove Docker Networks & Volumes (Optional)

To delete any Docker networks associated with Weaviate:

docker network prune

To delete any Docker volumes associated with Weaviate:

docker volume prune

🚀 Step 4: Full Cleanup (Only If Needed)

If you want to completely clean up all stopped containers, unused images, networks, and volumes, run:

docker system prune -a

⚠️ Warning: This will remove all unused containers, images, volumes, and networks, so use it with caution!

✅ Step 5: Verify Cleanup

To check that Weaviate is gone:

docker ps -a  # Should not list Weaviate
docker images  # Should not list Weaviate image

Now, your system is clean and free of Docker leftovers!

Next Steps

You now have Weaviate running in Docker on macOS. Next, you can:

Load data into Weaviate
Perform vector searches
Integrate with AI models for semantic retrieval

For more details, visit Weaviate Documentation.

FilesExpand file tree

docker_weaviate_readme.md

Latest commit

History