getumbrel / llama-gpt

A self-hosted, offline, ChatGPT-like chatbot. Powered by Llama 2. 100% private, with no data leaving your device. New: Code Llama support!
https://apps.umbrel.com/app/llama-gpt
MIT License
10.53k stars 666 forks source link

IPv6 support #145

Open JuniperChris929 opened 4 months ago

JuniperChris929 commented 4 months ago

Hi Folks, just deployed (or tried to deploy) the docker-compose files after modifying it to use my external ipv6 network, however, I can never access the UI via IPv6 (connection refused). Is there a quick way to enable IPv6?

Thanks

Chapoly1305 commented 3 months ago

This is not an direct issue of this project. Firstly, access http://[::1]:3000/ and see if you can assess this from your host, I tested without any issue. Ensure you have allowed ufw or iptables to expose port 3000 to network. Especially if you are in a managed network like in company or school, double check if IT has not block the port by default.

If you are sure there is nothing wrong about the firewall, then there are two possible methods,

  1. Use of Nginx or other software as a gateway, when user access a certain port of the host listened by Nginx, it proxy the request to your service, in this case, port 3000. This might be much easier if know Nginx.
  2. Enable the IPv6 support of the Docker by referring https://docs.docker.com/config/daemon/ipv6/

image