Deploy the redpanda-operator helm chart on Azure Kubernetes Service (AKS)

Question

Pulumi · Accepted Answer

Deploying a Helm chart on an Azure Kubernetes Service (AKS) cluster using Pulumi is a two-step process:

1. Provision an AKS cluster: You will first create an AKS cluster where your applications will run.
2. Deploy the Helm chart: Once you have an AKS cluster, you can deploy the Helm chart for the redpanda-operator or any other applications.

In the following Pulumi TypeScript program, we will accomplish both steps. We'll start by creating a new AKS cluster, and then we'll deploy the redpanda-operator using the Helm chart.

First, make sure you have the Pulumi CLI installed and you're logged in. You should also have configured your Azure credentials appropriately.

### Step 1: Define the AKS Cluster

To define an AKS cluster, we will use the `azure-native` provider, specifically the `ManagedCluster` resource which represents an AKS cluster.

### Step 2: Deploy the redpanda-operator Helm Chart

For deploying the Helm chart, we'll use the `Chart` resource from the `kubernetes` provider. Note that Pulumi's `Chart` resource is a representation of the Helm Chart that should be applied to your cluster.

### Program Explanation and Code

Below is the Pulumi program that fulfills the goal:

```typescript
import * as pulumi from "@pulumi/pulumi";
import * as azureNative from "@pulumi/azure-native";
import * as k8s from "@pulumi/kubernetes";

// Step 1: Provision an AKS cluster
const resourceGroup = new azureNative.resources.ResourceGroup("resourceGroup");

const cluster = new azureNative.containerservice.ManagedCluster("cluster", {
    resourceGroupName: resourceGroup.name,
    agentPoolProfiles: [{
        count: 3,
        maxPods: 110,
        mode: "System",
        name: "agentpool",
        osType: "Linux",
        vmSize: "Standard_DS2_v2",
    }],
    enableRBAC: true,
    dnsPrefix: pulumi.interpolate`${pulumi.getStack()}-kube`,
    kubernetesVersion: "1.18.14",
});

// Export the kubeconfig
export const kubeconfig = pulumi.all([cluster.name, resourceGroup.name]).apply(([clusterName, rgName]) => {
    return azureNative.containerservice.listManagedClusterUserCredentials({
        resourceName: clusterName,
        resourceGroupName: rgName,
    }).then(creds => {
        const encoded = creds.kubeconfigs[0].value;
        return Buffer.from(encoded, 'base64').toString();
    });
});

// Step 2: Deploy the redpanda-operator Helm Chart onto the AKS cluster
const redpandaChart = new k8s.helm.v3.Chart("redpanda-operator", {
    chart: "redpanda-operator",
    version: "0.1.0", // replace with the desired chart version
    fetchOpts: { repo: "https://charts.vectorized.io/" },
}, { provider: new k8s.Provider("k8s-provider", { kubeconfig }) });

// Export the redpanda-operator Helm chart status
export const redpandaOperatorStatus = redpandaChart.status;

```

Here’s what each part of the program is doing:

- The `azureNative.resources.ResourceGroup` resource creates a new resource group to contain the AKS cluster.
- `azureNative.containerservice.ManagedCluster` creates the AKS cluster itself. We specify the size and number of nodes, enable RBAC, and set the Kubernetes version.
- `kubeconfig` is exported, which contains the configuration needed to connect to your AKS cluster with kubectl.
- `k8s.helm.v3.Chart` represents the helm chart for the redpanda-operator. We specify the chart name and version, and provide the repository URL. We also pass the kubeconfig of the AKS cluster to the Kubernetes provider which is necessary for Pulumi to communicate with our cluster.
- Lastly, we export `redpandaOperatorStatus` to reveal the status of the deployed Helm chart.

After executing this program, Pulumi will output the `kubeconfig` you can use to access the cluster and the status of the deployed Helm chart, indicating if it was successful. You can run this program using `pulumi up`, which will execute the deployment.

Please ensure the version of the Helm chart specified is the one you want to deploy, you might need to check for the latest versions on the chart repository.