Install Botfront on servers or clusters

This section will explain how to deploy your Rasa assistant with Botfront on a remote server or Kubernetes / OpenShift cluster

Kubernetes / OpenShift installation

We provide Helm charts to install Botfront and Rasa projects on your cluster.

Please head to http://github.com/botfront/botfront-helm and follow the instructions.

Create a cluster from scratch

If you are looking to create a new cluster for the purpose of hosting Botfront and Rasa projects, you can check the following Terraform project: https://github.com/dialoguemd/botfront-terraform-gcp It is very specific to Google Cloud, but you can adapt it to the cloud provider of your choice.

Single server installation (VM)

An alternative option is to deploy all the services on a single machine. You can get a virtual machine from the Cloud Provider of your choice. We recommend a machine with at least 1 CPU and 2 Gb of RAM

1. Create a virtual machine with Ubuntu installed, and note the external IP address. For this tutorial, we’ll assume the IP address is 123.99.135.3
2. Install Node.js

wget -qO- https://raw.githubusercontent.com/nvm-sh/nvm/v0.35.1/install.sh | bash
export NVM_DIR="$([ -z "${XDG_CONFIG_HOME-}" ] && printf %s "${HOME}/.nvm" || printf %s "${XDG_CONFIG_HOME}/nvm")"
[ -s "$NVM_DIR/nvm.sh" ] && \. "$NVM_DIR/nvm.sh" # This loads nvm
nvm install lts/erbium

3. Install Docker and Docker Compose

sudo apt-get -y update
sudo apt-get -y remove docker docker-engine docker.io
sudo apt -y install docker.io
sudo systemctl start docker
sudo systemctl enable docker
sudo apt install curl
sudo curl -L "https://github.com/docker/compose/releases/download/1.24.1/docker-compose-$(uname -s)-$(uname -m)" -o /usr/local/bin/docker-compose
sudo chmod +x /usr/local/bin/docker-compose

4. Install Botfront

npm install -g botfront
botfront init # create a project

5. Edit the botfront.yml file

nano .botfront/botfront.yml

In the env section, change the root_url to the machine IP address (leave the port 8888 unchanged)

env:
  ...
  root_url: 'http://123.99.135.3:8888'
  ...

6. Launch Botfront

botfront up

7. Open Botfront in your browser (http://123.99.135.3:8888) and setup your project.

Look up the ID of your project. In the example URL http://123.99.135.3:8888/project/mSvL3ccZaz5RimDpF/, mSvL3ccZaz5RimDpF is the ID.

Run botfront set-project <project_id>. Rasa will restart and fetch that project’s configuration.

8. Go to settings/credentials and change the base_url host to the IP address (keep the host unchanged)

rasa_addons.core.channels.webchat.WebchatInput:
  session_persistence: true
  base_url: http://123.99.135.3:5005
  socket_path: "/socket.io/"

9. Botfront is ready to use.

Configuration

Services and Docker images

The table below lists all the services that can be used with Botfront.

Duckling (a structured entity parser developed by Facebook) is not strictly required if your NLU pipeline doesn’t use it.

Also, be very careful with your choice regarding MongoDB. If you decide to just run it as a container, be sure to at least properly mount the volume on a physical disk (otherwise all your data will be gone when the container is destroyed) and seriously consider scheduling back-ups on a regular basis.

Using a hosted service such as MongoDB Atlas is highly recommended, some of them even include a free plan that will be more than enough for small projects.

Environment variables

The following table shows the environment variables required by each service. Be sure to make those available as arguments or in the manifest files of your deployment

The following table list environment variables used to configure logging within botfront and botfront-api. Make those available as arguments or in the manifest files of your deployment

Volumes

Although volumes are technically not required for Botfront to run and work, if you do not mount them your data will be gone when containers are destroyed.

/app/models should be mounted on the same location so when Rasa restarts it can load the latest trained model.

MongoDB database considerations

It is highly recommended (but optional) to provide an oplog url with MONGO_OPLOG_URL. This will improve the reactivity of the platform as well as reduce the network throughput between MongoDB and Botfront.

Indicative minimal technical requirements

Those are the minimal requirements:

Endpoints

Endpoints let you define how your Rasa instance communicates with Botfront and the actions server:

• the Botfront API to query the bot responses (nlg)
• the actions server (action_endpoint)
• the tracker store (tracker_store)

nlg:
  type: 'rasa_addons.core.nlg.GraphQLNaturalLanguageGenerator'
  url: 'http://botfront:3000/graphql'
action_endpoint:
  url: 'http://actions:5055/webhook'
tracker_store:
  store_type:  'rasa_addons.core.tracker_stores.botfront.BotfrontTrackerStore'
  url: 'http://botfront:3000/graphql'