Deploy the irods-csi-driver helm chart on AWS EKS

Question

Pulumi · Accepted Answer

To deploy the irods-csi-driver helm chart on AWS EKS using Pulumi, we will perform the following steps:

1. **Create an EKS Cluster**: We will start by creating an EKS cluster to run our Kubernetes applications. This requires setting up a cluster and its associated compute infrastructure within AWS.
   
2. **Deploy Helm Chart on EKS**: Once the cluster is ready, we will utilize the `kubernetes.helm.v3.Chart` resource to deploy the irods-csi-driver helm chart on our EKS cluster.

Here is a detailed Pulumi TypeScript program that sets up an EKS cluster and deploys the irods-csi-driver helm chart onto it:

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

// Step 1: Create an EKS cluster
const vpc = new awsx.ec2.Vpc("vpc", {});
const cluster = new eks.Cluster("eks-cluster", {
    vpcId: vpc.id, // Use the VPC we created above
    subnetIds: vpc.publicSubnetIds, // Use the VPC's public subnets
    instanceType: "t2.medium", // Specify the desired EC2 instance type for the worker nodes
    desiredCapacity: 2, // Initial desired number of worker nodes
    minSize: 1, // Minimum number of worker nodes
    maxSize: 3, // Maximum number of worker nodes
    storageClasses: "gp2", // Default storage class for EBS volumes
});

const kubeconfig = pulumi.secret(cluster.kubeconfig);

// Step 2: Create a provider for the cluster created above to host Helm charts
const provider = new k8s.Provider("eks-provider", {
    kubeconfig: cluster.kubeconfig.apply(JSON.stringify) // Serialize the `kubeconfig` to string format as expected by the Provider
});

// Step 3: Deploy the irods-csi-driver helm chart using the Helm Chart resource
const irodsCsiDriverChart = new k8s.helm.v3.Chart("irods-csi-driver", {
    chart: "irods-csi-driver", // The name of the chart
    version: "1.1.1", // Specify the chart version (ensure this is the version you desire)
    namespace: "kube-system", // Specify the namespace to install the chart into
    fetchOpts: {
        repo: "https://cyverse.github.io/irods-csi-driver/", // Helm repository where the chart is hosted
    },
}, { provider }); // Ensure we use the provider that we declared earlier

// Export the Kubeconfig and EKS cluster name
export const clusterName = cluster.eksCluster.name;
export const kubeconfigExport = kubeconfig;

// Run this program using the Pulumi CLI:
// pulumi up

// After deployment, you can interact with your EKS cluster using `kubectl`:
// kubectl --kubeconfig <path-to-kubeconfig> get all --namespace kube-system
```

This program defines a VPC using the `awsx.ec2.Vpc` resource, which creates a new VPC suitable for an EKS cluster, ensuring that the EKS cluster will have a network to operate within. Then it creates the EKS cluster with `eks.Cluster` with a default node group that matches the desired specifications like instance type and scaling options.

Once the EKS cluster is ready, we declare a `k8s.Provider` which uses the kubeconfig from the created EKS cluster to interact with it. This provider configuration will be used when we install the helm chart.

Finally, with the `k8s.helm.v3.Chart` resource, we install the irods-csi-driver Helm chart into the `kube-system` namespace of our cluster specifying the version and repository where the Helm chart is hosted.

To run this Pulumi program, you will need:

- Pulumi CLI installed
- AWS account and the AWS CLI configured
- Your chosen AWS region, user credentials, etc., should already be set up either via the AWS CLI or environment variables.