Version: 3.x

Deploying a Rasa Assistant using Rasa Ephemeral Installer (REI)

note

Are you unfamiliar with Docker, Kubernetes and Helm? Check out "Understanding Rasa Deployments" on our YouTube channel.

If you would like to deploy Rasa Assistant using the Rasa OSS Helm chart on your machine, you can use Rasa Ephemeral Installer which installs all tools and creates a local Kubernetes cluster that allows you to use the Rasa OSS Helm chart.

We recommend this method as an alternative to docker-compose.

Create a local Kubernetes cluster via REI

The Rasa Ephemeral Install installs the following tools and creates a local Kubernetes cluster using kind.

Tools installed by REI:

  1. Simply execute the following command.

    curl -O https://rei.rasa.com/rei.sh && bash rei.sh -y

    After a few minutes, all components should be installed and a local Kubernetes cluster created.

    tip

    You can use the kubectl cluster-info to verify if all is good.

Deploy Rasa Open Source Assistant

  1. Add the Rasa Helm chart repository to helm.

    helm repo add rasa https://helm.rasa.com
    helm repo update

  2. Deploy Rasa Open Source server by using the rasa helm chart.

    helm install --namespace <NAMESPACE> <RELEASE_NAME> rasa/rasa

    For example:

    helm install --namespace rasa-oss rasa rasa/rasa
    tip

    If a namespace that you'd like to use doesn't exist you can use the --create-namespace to create it.

Access Rasa Open Source Assistant

By default the Rasa deployment is exposed via the rasa (<release name>) service and accessible only within a Kubernetes cluster. You can get the IP address using this command:

export SERVICE_PORT=$(kubectl get --namespace <your namespace> \
-o jsonpath="{.spec.ports[0].port}" services <release name>)
kubectl port-forward --namespace <your namespace> \
svc/<release name> ${SERVICE_PORT}:${SERVICE_PORT} &

Accessing Assistant from your local machine

The example above describe how to access Rasa Open Source Assistant by using kubectl port-forward. The other option is to expose your deployment on NodePort and access it directly.

  1. Prepare configuration that switch the rasa service to NodePort.

    # rasa-values.yaml
    service:
    type: "NodePort"

  2. Upgrade deployment.

    helm upgrade --namespace <NAMESPACE> --reuse-values -f rasa-values.yaml <RELEASE NAME> rasa/rasa

  3. Get the node port and address for the rasa service

    export NODE_PORT=$(kubectl get --namespace <NAMESPACE> -o jsonpath="{.spec.ports[0].nodePort}" services <RELEASE NAME>)
    $ curl http://127.0.0.1:${NODE_PORT}
    Hello from Rasa: 2.8.7

Configuration

Prepare an empty file called rasa-values.yaml which will include all your custom configuration for the installation with Helm.

All available values and more examples you can find in the Rasa helm chart repository.

note

The default configuration of the Rasa chart deploys a Rasa Open Source Server, downloads a model, and serves the downloaded model. Visit the Rasa helm chart repository to check out more examples of configuration.

Loading an initial model

The first time you install Rasa, you may not have a model server available yet, or you may want an lightweight model for testing the deployment. For this purpose, you can choose between training or downloading an initial model. By default, the Rasa chart downloads an example model from GitHub. To use this option, you don't have to change anything.

If you want to define an existing model to download from a URL you define instead, update your rasa-values.yaml with the URL according to the following configuration:

# rasa-values.yaml
applicationSettings:
initialModel: "https://github.com/RasaHQ/rasa-x-demo/blob/master/models/model.tar.gz?raw=true"
note

The URL for the initial model download has to point to a tar.gz file and must not require authentication.

Apply changes by executing the command:

helm upgrade --reuse-values -f rasa-values.yaml \
--namespace <NAMESPACE> <RELEASE_NAME> rasa/rasa

In the rasa helm chart repository you can find more options for loading models.

Configuring Channels

You can enable messaging channels by specifying credentials in rasa-values.yaml in the same way you would define them in credentials.yml when running locally.

For example, to enable the REST channel, update your rasa-values.yaml file with the following channel configuration:

# rasa-values.yaml
applicationSettings:
# (...)
credentials:
# (...)
additionalChannelCredentials:
rest:
note

For the rest channel, no credentials are required. To learn more see: https://rasa.com/docs/rasa/connectors/your-own-website)

Next steps