Deploy the etcd-operator helm chart on Oracle Kubernetes Engine (OKE)

Question

Pulumi · Accepted Answer

To deploy the `etcd-operator` Helm chart on Oracle Kubernetes Engine (OKE), we will use Pulumi's Kubernetes provider to apply a Helm chart. This provider interacts with a Kubernetes cluster's API server to manage the lifecycle of Kubernetes resources.

Before you begin, make sure you have access to an Oracle Kubernetes Engine (OKE) cluster and have the necessary credentials configured in your environment for Pulumi to interact with Oracle Cloud Infrastructure.

Here's a step-by-step guide to deploying the `etcd-operator` using Pulumi with TypeScript:

1. **Set Up a Pulumi Project**: Create a new Pulumi project if you haven't already.
2. **Install Required Packages**: We will use the `@pulumi/kubernetes` package.
3. **Initialize Resources**: We'll define the `kubernetes.Provider` to connect to your OKE cluster and then use `helm.v3.Chart` to deploy the `etcd-operator` from the Helm repository.

Below is a TypeScript program that does just that:

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

// Step 1: Initialize Kubernetes provider.
// The provider uses the kubeconfig file to authenticate with the OKE cluster.
// This file is typically found at `~/.kube/config` or can be set via the `KUBECONFIG` environment variable.
const provider = new k8s.Provider("oke-k8s", {
    kubeconfig: process.env.KUBECONFIG
});

// Step 2: Deploy the etcd-operator Helm chart.
// We specify the chart version to ensure repeatable deployments.
const etcdOperator = new k8s.helm.v3.Chart("etcd-operator", {
    // The chart name
    chart: "etcd-operator",
    // The repository where the chart can be found
    fetchOpts: {
        repo: "https://charts.helm.sh/stable", // Replace with the correct Helm repo for etcd-operator
    },
    // Optionally, we could specify values to customize the Helm chart's defaults.
    // Specifying values here would be equivalent to passing `--set` arguments to `helm install`.
    // values: { ... },
}, { provider });

// Export the endpoint of the etcd operator service
// For the `etcd-operator` chart, review the service names from the chart documentation or by inspecting the generated resources.
export const etcdOperatorEndpoint = etcdOperator.getResourceProperty("v1/Service", "etcd-operator-etcd-operator", "status").apply(status => status.loadBalancer.ingress[0].ip);

```

Here's what the program does:

- We import the necessary Pulumi packages.
- We create a `Provider` resource called `oke-k8s`, which informs the Pulumi Kubernetes SDK how to communicate with the cluster in question.
- Using the `Chart` resource, we deploy the `etcd-operator` Helm chart. The `etcd-operator` chart and its repository may need to be updated if the chart has been moved.
- In the `fetchOpts` property, you would specify the repository that houses the `etcd-operator` Helm chart. Make sure to replace `https://charts.helm.sh/stable` with the actual repository URL if it's different.
- We're using default values for the Helm chart, but you can specify custom `values` if needed, similar to how you would with a `helm install --set` command.
- Finally, we export the IP address of the `etcd-operator` service load balancer so that you can access it from outside the cluster if necessary.

To run this program, you would use the Pulumi CLI:
- `pulumi up` to preview and deploy the changes
- `pulumi stack output etcdOperatorEndpoint` to see the exposed IP address after deployment

Remember, Helm charts often create multiple resources and services. The output we're showing here (`etcdOperatorEndpoint`) is a simplification to illustrate how you might export a useful value from a Helm deployment. Check the ETCD operator documentation for the actual details you need to interact with ETCD once deployed.

Ensure the cluster version, Helm chart name, and values file match the expected inputs for the version of the ETCD operator you wish to deploy. Always consult the official chart documentation for the necessary configuration details.