Mode: talk, bot not responding v1.1.5

“Mode: talk” and “Share your assistant with Guest Testers” appears broken. I write something, the bot speech bubble shows and starts animating but there is never a response.

“Mode:” Interactive learning appears to work fine.

On an Ubuntu VM on Windows 10 host, prepared and ran docker container using this Dockerfile:

FROM python:3.6-slim

SHELL ["/bin/bash", "-c"]

RUN apt-get update && \
  apt-get install -y --no-install-recommends \
  build-essential \
  wget \
  openssh-client \
  graphviz-dev \
  pkg-config \
  git-core \
  openssl \
  libssl-dev \
  libffi6 \
  libffi-dev \
  libpng-dev \
  libpq-dev \
  curl && \
  apt-get clean && \
  rm -rf /var/lib/apt/lists/* /tmp/* /var/tmp/*

# Install Rasa-X

RUN pip3 install rasa-x --extra-index-url https://pypi.rasa.com/simple

# Init Rasa

RUN mkdir -p /app/rasa/projects/

WORKDIR /app/rasa/projects/

RUN rasa init --no-prompt

# Expose port

EXPOSE 5002

docker forwarded port 5002 to port 5002 on my ubuntu vm accessed via browser on my vm host machine via http://mylocalmachineipaddress:5002/login? and login query parameters provided by rasa x start up messages

i.e. access path is local machine > ubuntu vm > docker > rasa x

launched rasa with just “rasa x” from within the docker container

When you refresh the conversation in “Mode: Talk” is the bot response visible?

Yes, just now I realised there are other topics on this issue a bit older but I can confirm I’m seeing this anomaly with latest rasa x pip install as per Dockerfile provided.

1 Like

Yea, if you hit refresh you can usually get the bots responses to show up. If you are wanting to interact with it you should be able to run rasa from the command line and interact though. It’s been helpful to do that vs Rasa x sometimes. Kind of a pain when were trying to get a few conversations going and in order to get your response you have to hit refresh everytime. There is supposed to be a ‘temp fix’ coming soon I think though

I hope so, it is not a good look for the product. There is a concern people here will lose faith in it.

Just wanted to chime in and say that I’m experiencing exactly the same problem. I updated to the latest Rasa version now in order to see whether the problem persists. However, the NLU has been training the mode for over an hour now already whereas it used to take a few minutes with the old version.

Update: Got the model trained using spacy “en_core_web_sm” instead of “en_core_web_md”. The problem, however, persists.

rasa==1.1.3

rasa-x==0.19.5

rasa-sdk==1.1.0