My System
OS VirtualBox Ubuntu 18.04 64 bit on a Windows 10 machine. Docker version: 20.10.12, build e91ed57 Rasa version, I have tried: 3.0.8, 3.0.7 and 1.7.4 I am the root user
I am following the Building an Assistant in Docker documentation here
Issue
When I run:
docker run -v $(pwd):/app rasa/rasa:3.0.8-full init --no-prompt
All of the folders are created, apart from models. The documentation states that the model’s folder should be created when running the above command and for me to check this.
A have had a search around the forum and added -u root.
I have also followed the YouTube Understanding Rasa Deployments - Premade Rasa Containers from the Rasa channel.
The YouTube video encouraged me to try:
docker run -it -p 8080:8080 -v $(pwd):/app rasa/rasa run --enable-api --port 8080
and
docker run -it -p 8080:8080 -v $(pwd):/app rasa/rasa:3.0.8-full run --enable-api --port 8080
This creates an empty models folder.
When I run the second command within the documentation:
docker run -it -v $(pwd):/app rasa/rasa:3.0.8-full shell
I get the following error
Traceback (most recent call last):
File "/opt/venv/bin/rasa", line 8, in <module>
sys.exit(main())
File "/opt/venv/lib/python3.8/site-packages/rasa/__main__.py", line 121, in main
cmdline_arguments.func(cmdline_arguments)
File "/opt/venv/lib/python3.8/site-packages/rasa/cli/shell.py", line 112, in shell
metadata = LocalModelStorage.metadata_from_archive(model)
File "/opt/venv/lib/python3.8/site-packages/rasa/engine/storage/local_model_storage.py", line 75, in metadata_from_archive
cls._extract_archive_to_directory(model_archive_path, temporary_directory)
File "/opt/venv/lib/python3.8/site-packages/rasa/engine/storage/local_model_storage.py", line 84, in _extract_archive_to_directory
with TarSafe.open(model_archive_path, mode="r:gz") as tar:
File "/opt/venv/lib/python3.8/site-packages/tarsafe/tarsafe.py", line 15, in open
return super().open(name, mode, fileobj, bufsize, **kwargs)
File "/usr/lib/python3.8/tarfile.py", line 1592, in open
raise ValueError("nothing to open")
ValueError: nothing to open
I have also tried changing $(pwd) to the path of the folder which contains the rasa files.
I have also tried training a model, as per the setup documentation:
docker run -v $(pwd):/app rasa/rasa:3.0.8-full train --domain domain.yml --data data --out models
I have also set the permission of the folders that contain the code by using:
sudo chmod +777 -R $(pwd)
Any help would be greatly appreciated, as I have been reading around the issue for the past few weeks without any success.