Skip to content
This repository has been archived by the owner on Apr 10, 2024. It is now read-only.

stolostron/hub-of-hubs-console

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

@open-cluster-management/console

License Build Coverage Maintainability Rating Security Rating Vulnerabilities

What is console?

The console provides a user interface for cluster related items such as:

  • Credentials management (For example, Cloud Providers)
  • Cluster management
    • Creating
    • Importing
    • Upgrading
    • Detaching
    • Destroying

Go to the Contributing guide to learn how to get involved.

Prerequisites

  • Node.js v14.x
    • node -v
  • NPM

Workspaces

The console repository is setup to use NPM 7 workspaces. This allows the projects with the console repository to reference each other. This enables the easy sharing of a common libraries. It also enables the decoupling of different projects but allows them to pulled together into a final product.

Notes:

  • Workspaces are defined in the package.json at the root
  • While workspaces allow the sequential running of commands across all workspaces, console is using a project workspaces-run to run command across all workspaces concurrently.

Running

  1. Clone repository

  2. Install dependencies

    npm ci
    
  3. Setup environment

    You need:

    • to be connected to a OpenShift 4.x.x cluster
    • to have Advanced Cluster Management installed on the cluster
    npm run setup
    

    This will create a .env file in the backend directory containing the environment variables.

  4. Start the development services

    npm start
    

    This will start the frontend and the backend in parallel. (It may take up to 30 seconds for the UI to appear)

    The frontend will proxy requests to the backend using react scripts.

    The backend will proxy requests to the kubernetes cluster specified by CLUSTER_API_URL in backend/.env.

Backend

Method Path Description
ALL /api Proxy request to cluster /api
ALL /apis Proxy request to cluster /apis
GET /watch Server side event stream of kubernetes resources
POST /upgrade Upgrade a cluster

Proxy Routes

The backend proxies api calls to the kubernetes cluster api. Proxy calls will use the clients api token stored in acm-access-token-cookie cookie. This enforces user RBAC for proxy calls.

Watch Route

The backend /watch route streams resource changes using server side events.

Frontend

The frontend is using react scripts to simplify dependencies. The react scripts internally handle all the webpacking of the frontend. The goal here is fewer dependencies of the frontend.

Chrome

To develop with self signed certificates goto chrome://flags/and enable

  • Allow invalid certificates for resources loaded from localhost.
  • Insecure origins treated as secure.

Authentication

Frontend has a cookie acm-access-token-cookie that contains the user's token.

If the backend responds with a 401 Unauthorized the frontend starts an OAuth flow to authorize with the cluster.

  1. Frontend redirects to the backend /login endpoint.
  2. Backend redirects to the cluster /authorize endpoint.
  3. Cluster OAuth redirects back to the backend /login/callback endpoint.
  4. Backend redirects to the frontend and sets the acm-access-token-cookie.

Optional Features

In some cases there are development preview (Dev Preview) features or technical preview (Tech Preview) features that can optionally be enabled and used in the product. These features are enabled using a FeatureGate CR. The following features are available:

Single node OpenShift

Enabling this feature will allow the user to create a cluster that only contains a single master node. This option is only available for providers AWS, Azure, GCP, OpenStack and VMware when the OpenShift release image is version 4.8 or higher.

To enable this feature, oc apply -f features/single-node-openshift.yaml.

References

console is part of the open-cluster-management community. For more information, visit: open-cluster-management.io

Debugging

To restart the console pods, run:

kubectl delete pods -n open-cluster-management -l component=console