Deploy the twitter-app helm chart on Rancher

Question

Pulumi · Accepted Answer

To deploy a Helm chart on a Rancher-managed Kubernetes cluster using Pulumi, you will first need to have access to a Rancher server and have a Kubernetes cluster provisioned and managed by Rancher. Also, you must have access to the Helm chart that you want to deploy; in this case, the twitter-app chart.

To accomplish the deployment, you will use the Pulumi Rancher2 provider, which allows you to interact with Rancher resources. The deployment will consist of adding the Helm chart as a catalog in Rancher and then deploying it to a specific project within the Kubernetes cluster.

Below is a program in TypeScript that demonstrates how to deploy the twitter-app Helm chart on Rancher. This program assumes you have already set up Pulumi and have access to a Rancher environment with an existing Kubernetes cluster.

First, we configure Pulumi to use the Rancher2 provider. We then create a catalog for the Helm chart, assuming that the twitter-app Helm chart is hosted in a Git repository accessible via a URL. Finally, we deploy the application by creating a App resource, specifying the details of the Helm chart and the settings you want to configure.

import * as rancher2 from "@pulumi/rancher2";
import * as pulumi from "@pulumi/pulumi";

// This is the main Pulumi program entry point.
const main = async () => {

    // Initialize the Rancher2 provider.
    // Ensure you have the Rancher API credentials and Rancher server URL set up in your environment.
    const provider = new rancher2.Provider("rancherProvider", {
        apiURL: "https://your-rancher-api-url", // Replace with your Rancher API URL.
        accessToken: "your-access-token",       // Replace with your Rancher API Access Token.
        secretKey: "your-secret-key",           // Replace with your Rancher API Secret Key.
    });

    // Create a catalog for the Helm chart if it's not part of the default catalogs.
    // The catalog is a repository that holds Helm chart definitions.
    const twitterAppCatalog = new rancher2.CatalogV2("twitterAppCatalog", {
        // The Rancher Cluster ID where you want to deploy.
        // You can get the cluster ID from the Rancher UI or API.
        clusterId: "c-xxxxx", // Replace with your actual cluster ID.
        gitRepo: "https://github.com/<user>/twitter-helm-charts", // Replace with your Git repository URL.
        gitBranch: "main", // The default branch where the chart is located.
        // Specify the path if the chart is in a subdirectory, e.g., "path/to/chart/".
    }, { provider: provider });

    // Deploy the twitter-app Helm chart using the created catalog.
    const twitterApp = new rancher2.AppV2("twitterApp", {
        // The namespace where you want to deploy your app. 
        // If the namespace doesn't exist, Rancher will create it.
        namespace: "default",
        // Reference the ID of the catalog you created earlier.
        repoName: twitterAppCatalog.id,
        // Specify the name of the chart you want to deploy.
        chartName: "twitter-app",
        // Specify the version of the chart you're deploying.
        chartVersion: "1.0.0", // Replace with the actual version you wish to deploy.
        // The values for the Helm chart.
        // These settings depend on what the twitter-app chart requires.
        // You can use the `valuesYaml` property to specify values in YAML format, or `values` as an object.
        valuesYaml: `replicaCount: 2`, // This is an example of setting the replica count to 2.
    }, { provider: provider });

    // Export the app endpoint, usually by getting the service URL or IP from the deployed resources.
    return {
        appUrl: pulumi.output(twitterApp.status).apply(status => status?.appUrl), // This is an example key; adjust according to actual output.
    };
};

// Run the main function and export the stack outputs.
main().then(outputs => pulumi.output(outputs));

In this example:

The rancher2.CatalogV2 resource creates a new catalog entry pointing to the Git repository where the Helm chart is located. Ensure to replace placeholder values with actual details like the Git repository URL, branch, and the Rancher Cluster ID.
The rancher2.AppV2 resource represents the Helm chart deployment. It requires the name of the namespace to deploy into (default in this example), the repository name (which we obtain from the catalog resource ID), the chart name (twitter-app), and optionally, the chart version.
valuesYaml is used to pass a string containing the YAML-formatted values you want to set for the Helm chart, such as replicaCount. Replace the placeholder with actual configuration relevant to the twitter-app chart. You can also use the values property if you prefer to specify your values as a JavaScript object.
The appUrl stack export is just a placeholder demonstrating how you might export a relevant endpoint or value from the deployed app. You will need to adjust the property path based on the actual outputs provided by the Helm chart.

Please replace placeholder values (anything with "your-" prefix or surrounded by angle brackets <>) with your actual values.

Make sure that you have appropriate access to the Rancher API through the configured credentials before running this program. In practice, sensitive information such as access tokens and secret keys should be managed securely, for example, by using Pulumi Secrets.

Additionally, specific values for chart configuration (valuesYaml or values) depend entirely on the content of the Helm chart. You will need to refer to the documentation or source of the twitter-app chart to determine the correct values.

Remember to install the required Pulumi package for Rancher2 and set up your Pulumi project before running this code. You can do so by running pulumi new typescript to create a new Pulumi TypeScript project and then use npm or yarn to install the @pulumi/rancher2 package.

Run pulumi up to preview and deploy your changes. After reviewing the preview, you can proceed to confirm the deployment if everything looks correct.