webui not working

[muccio]

installed on macos (apple silicon) with docker desktop in a container.
logs and docker show up and running but browser can't connect to server.
I'm missing something?

[email protected] start
next start
▲ Next.js 13.5.6

• Local: http://localhost:3000
  ✓ Ready in 461ms

[rsxdalv]

Could you please try the primary UI at 7680?

[muccio]

always unreachable

[muccio]

Also not showing ports in the docker desktop interface

[rsxdalv]

Did you run it with docker compose or in another way?

I'm guessing the ports were not exposed during the creation of the container from the image.

[muccio]

I was following the install instructions
docker pull ghcr.io/rsxdalv/tts-generation-webui:main
docker compose up -d

[rsxdalv]

Sounds like a malfunction, 3000 and 7770 should be open. The container tries to use the Nvidia driver, so maybe it fails when it's unavailable? But you should be able to see that in the logs.
I'm assuming other docker containers with ports work.

[muccio]

yes other containers no problem

[rsxdalv]

Thanks, could you share a full log of the container? Maybe it contains the error. Also I'm not sure if it's possible/easy to make a real M1 chip accelerated docker container - this Linux one is most probably not going to be able to use it.

[muccio]

2025-01-14 14:37:04
2025-01-14 14:37:04 ==========
2025-01-14 14:37:04 == CUDA ==
2025-01-14 14:37:04 ==========
2025-01-14 14:37:04
2025-01-14 14:37:04 CUDA Version 11.8.0
2025-01-14 14:37:04
2025-01-14 14:37:04 Container image Copyright (c) 2016-2023, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
2025-01-14 14:37:04
2025-01-14 14:37:04 This container image and its contents are governed by the NVIDIA Deep Learning Container License.
2025-01-14 14:37:04 By pulling and using the container, you accept the terms and conditions of this license:
2025-01-14 14:37:04 https://developer.nvidia.com/ngc/nvidia-deep-learning-container-license
2025-01-14 14:37:04
2025-01-14 14:37:04 A copy of this license is made available in this container at /NGC-DL-CONTAINER-LICENSE for your convenience.
2025-01-14 14:37:04
2025-01-14 14:37:04 WARNING: The NVIDIA Driver was not detected. GPU functionality will not be available.
2025-01-14 14:37:04 Use the NVIDIA Container Toolkit to start this container with GPU support; see
2025-01-14 14:37:04 https://docs.nvidia.com/datacenter/cloud-native/ .
2025-01-14 14:37:04
2025-01-14 14:37:08
2025-01-14 14:37:08 > [email protected] start
2025-01-14 14:37:08 > next start
2025-01-14 14:37:08
2025-01-14 14:37:09 ▲ Next.js 13.5.6
2025-01-14 14:37:09 - Local: http://localhost:3000
2025-01-14 14:37:09
2025-01-14 14:37:09 ✓ Ready in 450ms
2025-01-14 14:37:09 /usr/local/lib/python3.10/dist-packages/gradio/components/dropdown.py:226: UserWarning: The value passed into gr.Dropdown() is not in the list of choices. Please update the list of choices to include: None or set allow_custom_value=True.
2025-01-14 14:37:09 warnings.warn(

[rsxdalv]

Thanks, nothing major up until this point, maybe it's something else. I have previously ran the containers successfully, so I know it's possible, but I can't really guess what is the problem here.
The log indicates that it's going to work anyway, even without GPU.

[rtotalminer]

I get this problem too, I tried with the same PC using Windows but get:

'js' is not recognized as an internal or external command,
operable program or batch file.
Failed to init the app, exiting...
'"' is not recognized as an internal or external command,
operable program or batch file.

[rsxdalv]

I get this problem too, I tried with the same PC using Windows but get:

'js' is not recognized as an internal or external command,
operable program or batch file.
Failed to init the app, exiting...
'"' is not recognized as an internal or external command,
operable program or batch file.

This issue is for Docker containers running Ubuntu on a Mac.
The Windows issue you are reporting usually happens to Windows 11 and some users report that after a restart or some other pause, the error goes away and it just works. It's still not clear exactly what causes it.

[rtotalminer]

Okay, interesting... I am on Windows 10 btw, going to restart my PC if WSL doesn't work -- which it didn't (I am using manual instalation on Windows as Docker didn't work on my Linux)

I select the CPU option BTW.

On WSL I got the dotenv issue.

I got the same error as OP but was using Docker on Linux.

Retrying the manual install on W10, will try restarting if I get the error again.

[rtotalminer]

tts-generation-webui can also be ran inside of a Docker container. To get started, pull the image from GitHub Container Registry:

docker pull ghcr.io/rsxdalv/tts-generation-webui:main

Once the image has been pulled it can be started with Docker Compose:

docker compose up -d
I don't understand how makes a container without specifying the image

I am going to retry it later, I probaly was typing in somemthing wrong.

[rtotalminer]

Also there isn't any venv for python being created?

[rsxdalv]

Docker compose file is at the root of the repository. It's circular because that way everything is within the same repository.

Conda creates venv
The docker container is meant for NVidia GPUs but afaik it works CPU mode too.

[rsxdalv]

On WSL, did you run the .sh file?