High Level Architecture

To fully deploy Coder Enterprise, you will need to:

1. Prepare a Kubernetes environment on which you will deploy Coder Enterprise
2. Install Coder Enterprise onto your Kubernetes environment
3. Set up and configure your Coder Enterprise deployment

Architecture

Before proceeding, it's helpful to have a high-level overview of how the Coder Enterprise ecosystem works.

Coder Enterprise is made up of the following components:

• Manager: Acts as the central authority; provides authentication and supports the Dashboard and an API which you can use to create and interact with Environments
• Envproxy: Acts as the websocket proxy to an environment's editor and terminal
• PostgreSQL: Provides data storage for session tokens, environment information, etc.

Every component runs within its Kubernetes pod in a basic Coder deployment.

Kubernetes NGINX Ingress

By default, Coder Enterprise deploys an NGINX Kubernetes ingress controller to allocate and route requests to the appropriate service. You can disable this controller in the helm chart values if you would prefer to use your ingress or gateway.

Deployment Options

There are two primary ways to deploy Coder Enterprise:

1. The default installation, which is a non-air-gapped option, using the Kubernetes provider of your choice; you should be able to freely access Coder resources from this environment
2. A secured, [air-gapped option](doc:air-gapped-deployments); you can choose to limit access and deploy Coder by first pulling in all of the required resources, or you can choose to whitelist the URLs/IP addresses needed to access Coder Enterprise resources

Please note that Coder Enterprise's auto-generated licenses do not permit the use of air-gapped environments.