Deploy the hzero-admin helm chart on AWS EKS

Question

Pulumi · Accepted Answer

To deploy the `hzero-admin` helm chart on AWS EKS, you'll need to create an EKS cluster on AWS and then install the helm chart to that cluster. The Pulumi program will include the following steps:

1. **Creating the EKS Cluster**: We will use the `@pulumi/eks` package to create a new EKS cluster. This package provides a high-level interface to simplify the creation of EKS clusters.
   
2. **Deploying the Helm Chart to the Cluster**: Once the EKS cluster is up and running, we will deploy the `hzero-admin` chart on it using the `@pulumi/kubernetes` package, which allows us to operate on Kubernetes objects and Helm charts.

Below is the TypeScript code that accomplishes the above steps. Make sure you have AWS credentials configured for Pulumi to use.

```typescript
import * as eks from "@pulumi/eks";
import * as k8s from "@pulumi/kubernetes";
import * as pulumi from "@pulumi/pulumi";

// Create an EKS cluster.
const cluster = new eks.Cluster("my-cluster", {
    desiredCapacity: 2,
    minSize: 1,
    maxSize: 2,
    storageClasses: "gp2", // Specify any required storage classes
    deployDashboard: false, // Dashboard is not recommended in production environments
});

// Export the cluster's kubeconfig.
export const kubeconfig = cluster.kubeconfig;

// Configuring the Helm chart deployment for `hzero-admin`.
const hzeroAdminChart = new k8s.helm.v3.Chart("hzero-admin-chart", {
    chart: "hzero-admin",                // The name of the chart to deploy
    version: "x.y.z",                    // Replace with the desired chart version
    fetchOpts:{
        repo: "http://helm-repo-url",    // The repository URL where the chart is hosted
    },
    // Add any required values or configurations for the hzero-admin helm chart.
    values: {
        // Values to configure the helm chart. Replace the placeholders with actual values.
        someConfiguration: "value",
        // more configuration here...
    },
}, { provider: cluster.provider }); // Ensure that Helm uses the created EKS cluster as its provider

// When the program runs, Pulumi applies these configurations and creates the EKS cluster,
// then it deploys the Helm chart to the cluster. If you need to customize the configuration
// for `hzero-admin`, you should consult the chart's documentation for the available settings.
```
Let's walk through what each part of the code is doing:

- The `eks.Cluster` class is used to create an EKS cluster in your AWS account. We provide it a name, and specify the desired size parameters, as well as any other specific configurations required. In the storage classes, we set "gp2" which is a general-purpose SSD volume provided by AWS. We turn off the Kubernetes Dashboard by setting `deployDashboard` to `false`, which is a common security practice.

- The `kubeconfig` is exported, which you can use with `kubectl` or other Kubernetes tools to interact with your EKS cluster.

- Next, we define a Helm chart resource using the `k8s.helm.v3.Chart` class. This resource is what will actually deploy `hzero-admin`. It points to the chart name, optionally the version (if you want to pin it to a specific one), and the repository where the Helm chart is hosted.

- The `values` object within the Helm chart configuration allows you to provide custom values to the chart, which is equivalent to what you would put into a `values.yaml` file when using Helm directly.

- Lastly, we use the `{ provider: cluster.provider }` option when creating the Helm chart to ensure that it's installed into the created EKS cluster.

After writing this code to a `index.ts` file, you can deploy the infrastructure by running `pulumi up` from the same directory. Pulumi will then execute the above code to provision the AWS EKS cluster and deploy the `hzero-admin` Helm chart onto it. Make sure to adjust chart version, values, and Helm repository URL as per requirements of the `hzero-admin` chart since these will be specific to the chart's version and your deployment needs.

Remember that managing cloud resources like an EKS cluster can incur costs, so ensure you're familiar with AWS pricing and clean up resources with `pulumi destroy` when they're no longer needed.