.env file not load with docker installation

[dragnod0]

Hi, I’ve been trying all day with ChatGPT’s help to get AutoGPT running in a Proxmox LXC container where Ubuntu Server 24.04 is installed.

It’s been really tough. I had to completely uninstall AppArmor because Docker wouldn’t start at all with it. The Python variant didn’t work at all.

Although I can start AutoGPT with Docker, I have to manually enter the OpenAI API key each time because the .env file is not found.

The directory ~/AutoGPT exists, and I copied the .env.template file there, renamed it to .env, inserted the API key after OPENAI_API_KEY= and removed the hash to activate it. It still doesn’t work.

I’ve tried accessing Docker’s installation location to adjust the file there, but that doesn’t seem to be straightforward. It’s beyond my knowledge. Unfortunately, ChatGPT hasn’t been helpful in this regard either.

When I start AutoGPT with the manual key entry, I always get the error message
HTTP Request: POST https://api.openai.com/v1/chat/completions "HTTP/1.1 429 Too Many Requests".
I’m guessing this is because I don’t have a Pro account yet.
I really don’t know what to do next. None of the installation guides are working. Maybe someone can give me some tips?

[dragnod0]

Okay, I managed to get it partly working. The API key is now in the docker-compose.yml file. However, I am still getting the error message HTTP Request: POST https://api.openai.com/v1/chat/completions "HTTP/1.1 429 Too Many Requests".

[DaveGreasley]

The Too Many Requests error suggests you have successfully authenticated to OpenAI API but have exceeded the number of requests allowed by the rate limit for the account. See here for more information -

https://help.openai.com/en/articles/5955604-how-can-i-solve-429-too-many-requests-errors
https://platform.openai.com/docs/guides/rate-limits