Hey there, I hope you can help me fixing my problem. I use Rasa X 0.35.0 on a Ubuntu 18.04 server. It was installed with the docker-compose installation script. My config.yml contains the BytePairFeaturizer from Rasa NLU Examples so I edited the docker-compose.yml and changed the worker-container like this:
rasa-worker:
hostname: rasa-worker
extra_hosts:
- "rasa-x:172.28.0.20"
<<: *default-rasa-service
volumes:
- /rasa-extensions/rasa-nlu-examples/:/app/rasa-nlu-examples
environment:
<<: *rasa-credentials
PYTHONPATH: "/rasa-extensions/rasa-nlu-examples/:/app/rasa-nlu-examples"
RASA_ENVIRONMENT: "worker"
DB_DATABASE: "worker_tracker"
MPLCONFIGDIR: "/tmp/.matplotlib"
RASA_MODEL_SERVER: "http://rasa-x:5002/api/projects/default/models/tags/production"
I already tried export PYTHONPATH=/rasa-extensions/rasa-nlu-examples/
and
sudo chgrp -R root /etc/rasa/* && sudo chmod -R 770 /etc/rasa/*
sudo chown -R 1001 /etc/rasa/db && sudo chmod -R 750 /etc/rasa/db
But as soon as I hit “train model” in Rasa X I get the following error (even before adding the BytePairFeaturizer to my config):
2021-01-22 14:47:48 WARNING rasa.utils.common - Failed to write global config. Error: [Errno 13]
Permission denied: '/.config'. Skipping.
Changing the docker-compose.yml back to the normal woker script works fine. But then I can’t use my config.yml