Deployment Environments

Architecture#

Rasa Enterprise allows multiple versions of an assistant to be run simultaneously. By default, two environments are defined:

• production

• worker

Each environment corresponds to a Rasa Open Source server which uses the model with the corresponding tag (e.g. production). Model training always happens on the worker instances to ensure the training does not degrade performance of the bot itself.

A model can be given multiple environment tags, but each environment can only have a single model associated with it. If model A is tagged development and you promote it to production, the model will now have two tags, and the two environments will use the same model. This ensures that you can promote a model to an environment without affecting any others.

The name and number of environments can be customized when deploying Rasa Enterprise.

Workflow#

1. The first step is to move a trained model into the development environment. To do this, first navigate to the models view. Hover over a model, and click the three dots menu which appears on the table row. Now you can select the development environment and click ‘Save’.

This model will now be used when you select the development environment in the Share your Bot link generation screen.

1. Now that your model is running in the development environment, you can do some testing. Navigate to the Conversations screen, open the ‘Share your Assistant’ options and select ‘development’ from the environments dropdown.
2. If you make any improvements, like adding more training data or updating the response templates, you need to use the ‘train’ button to create a new model and promote that model to development.
3. Once you are satisfied with your model’s performance, repeat step 1 but instead selecting the production environment.

Customizing Deployment Environments#

An additional Rasa Open Source server must first be created before it can be made available to Rasa Enterprise. Once it is available, you can add it as a deployment environment to Rasa Enterprise in the user interface on the Models screen.

Starting Additional Rasa Servers#

Any Rasa Open Source deployment can be configured as a deployment environment given the correct settings and credentials. Please follow the instructions for connecting an existing deployment via the event broker to set up event forwarding from your Rasa Open Source server to Rasa Enterprise.

In addition to correctly configuring the event broker, the following settings are required for a Rasa Open Soure server to act as a deployment environment:

1. The environment variable RASA_ENVIRONMENT must be set to the name of the environment e.g. before running the server:

This value must match the target model tag and the name specified in the UI when adding the deployment environment.

2. The model server URL must point to the target model endpoint of your Rasa Enterprise deployment. The URL should be in the format <Rasa-Enterprise-Base-URL>/api/projects/default/models/tags/<target-model-tag> For example:

RASA_X_TOKEN refers to the environment variable set in your Rasa Enterprise deployment.

3. The rasa channel credentials must point to your Rasa Enterprise deployment.

4. The server must be secured with an authentication token. You will need to provide this token to Rasa Enterprise when you add the server as a deployment environment. For example, in the startup command for your Rasa Open Source server:

Configuring Deployment Environments in the UI#

Once you have started and configured a Rasa Open Source server, you can add it as a deployment environment in the Rasa Enterprise UI.

On the models screen, go to Configure Environments:

If you are editing an existing environment, hover over the environment and choose "Edit". If you are adding a new environment, choose Add an Environment:

Add the environment name, url, and authentication token for the Rasa Open Source server the environment should point to. The url field should be set to the base address of the Rasa Open Source server.