Configure

This section describes the steps that an administrator from the platform team has to follow to deploy and operate "Ray as Service" on their infrastructure. The high level steps that the administrator has to follow to get this operational on their infrastructure are:

As an Org Admin,

Clone the Ray Template from Rafay's Template Catalog
Configure and Customize the template

As a PaaS Project Admin (or Org Admin),

Create a Service Profile in PaaS Studio for End User Self Service

High Level Steps¶

The sequence diagram below describes the high level steps visually.

sequenceDiagram 
    participant plat as Platform Team 
    participant rafay as Template Catalog
    participant studio as PaaS Admin

    plat->>rafay: Clone Ray Template
    plat->>rafay: Configure Template
    studio->>rafay: Create Service Profile
    rafay-->>plat: Setup Complete

Clone Template from Catalog¶

The step below is typically a one time activity for administrators. Login into your Rafay Org as an Org Admin and follow the steps below:

Click on "System -> Template Catalog"
Click on "Get Started" on the KubeRay template
Follow the wizard i.e. provide a name, version and select the name of the project you created in the previous step (e.g. gpu-paas)
Save as Draft

Ray on Kubernetes

Configure Template¶

When executed, the Rafay agent operating behind the firewall will receive the configured template, the associated Infra as Code (IaC), credentials and policies from the Rafay Platform. It will then execute this code on the behalf of the user.

1. Specify Agent¶

Let's configure the template to be received and executed by the Rafay Agent we created in a prior step.

Under the "Agents" tab, click on "Add Agent"
Select the name of the agent you configured in the prior step
Ensure that the override configuration is specified to "Not Allowed" because we do not want the downstream users to be able to change this
Save your changes

Select Agent

2. Config Context¶

The config context will typically encapsulate credentials and environment variables required for the agent to perform its job. In this case, we will configure the Rafay Agent with credentials so that it can make programmatic (API) calls to the specified Rafay Org.

To get the API Key + Secret for the administrator user,

Navigate to "My Tools -> Manage Keys" and click on "New API Key".
Copy the API Key + Secret combination.

Info

Click here to learn more about API Key & Secret for programmatic access.

Now, we are ready to configure our agent's config context.

Click on the Environment Template → Config Context
Click on the Edit icon in the config context for "kuberay-config-context"
Expand "Environment Variables" and you should see two entries: "API Key & Controller Endpoint"
Click on Edit for API Key
Paste the API Key/Secret string from the above step
Select Override to "Not Allowed" to ensure none of the downstream users have visibility or access to the config context
Save & Continue

Paste API Key

Controller API Endpoint

Info

For self hosted Rafay Controller deployments, the agent will need to be configured to point to your custom domain.

In this guide, we will be using the URL for Rafay's SaaS option i.e. "console.rafay.dev"

Click on Edit for "Controller Endpoint".
Note that Rafay's SaaS Endpoint URL is already configured
Select Override to "Not Allowed" to ensure none of the downstream users have visibility or access to the config context
Save & Continue

Controller Endpoint

Input Variables¶

The Rafay template allows administrators to customize a number of input variables. Please review the table below for a detailed description of the input variables. Typically, the Org Admin only needs to specify/configure the following from the list.

Domain related details
Certificate related details

Note

For testing and evaluation purposes, by default, Rafay automatically programs DNS for the Ray Endpoint URL and configures it for secure access using a trusted digital certificate. For production deployments, administrators should specify both DNS and Certificate details. Customers are recommended to try with the Rafay Defaults first before attempting advanced configuration.

Name	Value Type	Description
cluster_name	Text	Name of the Kubernetes cluster (virtual or dedicated). This is automatically detected when user selects compute instance
Host Project	Expressions	Rafay Project where the compute instance is deployed. This is automatically specified during end user self-service
Environment Name	Expressions	Name of the Rafay environment instance of the template. This is automatically specified during end user self-service
Ingress Domain	Text	The ingress domain for the Ray Dashboard. Defaults to Rafay. Administrators should customize
Kuberay Host Name	Text	Hostname for Ray Dashboard. Required for Custom Ingress Domain
Kuberay Host Cert	Text	Certificate for Ray Dashboard URL. Starts with BEGIN CERTIFICATE, necessary for Custom Ingress Domain
Kuberay Host Key	Text	Key associated with Certificate for Ray Dashboard URL. Starts with BEGIN PRIVATE KEY, necessary for Custom Ingress Domain
Ingress IP	Text	Public Ingress IP of Cluster, necessary for Rafay Ingress Domain. Typically specified in Service Profile
Ingress Namespace	Text	Namespace where ingress controller is installed on the host cluster. Defaults to ingress-nginx
Ingress Class Name	Text	Ingress Class Name. Defaults to nginx
Sub Domain	Expressions	Subdomain for unique Ray Dashboard URL per user
Ingress User	Text	Username to access Ray Dashboard. Defaults to "admin"
Enable Volcano	True/False	Toggle true if Volcano should be installed on the cluster. Ideal for on-premises host clusters where auto scaling is challenging
Kuberay Head Config	HCL	Kuberay head node configurations. Update only if defaults do not work
Kuberay Worker Config	HCL	Kuberay worker node configurations. Update as required
Kuberay Worker Tolerations	HCL	Tolerations for KubeRay worker nodes. Update only if required
Kuberay Worker Node Selector	HCL	Node Selector Config for KubeRay worker nodes. Update only if required
Host server	Expressions	Host Cluster details used to configure Ingress etc on the host cluster using Kubeconfig. Automatically detected
Client Certificate Data	Expressions	Kubeconfig Client Certificate Data. Automatically detected
Client Key Data	Expressions	Kubeconfig Client Key Data. Automatically detected
Certificate Authority Data	Expressions	Kubeconfig Certificate Authority Data. Automatically detected
Kubeconfig	Expressions	Kubeconfig for programmatic access to Host Cluster. Automatically detected