Provisioning a VM on Proxmox VE with Pulumi

Question

Pulumi · Accepted Answer

Provisioning a virtual machine (VM) on Proxmox VE (Virtual Environment) with Pulumi would involve several steps, including setting up Pulumi with the appropriate provider, defining the resources in code, and executing the deployment. Support for Proxmox VE is not natively included in Pulumi’s standard providers, so this process would typically require using a custom provider or integrating with existing automation tools that can manage Proxmox resources.

However, it seems that specific support for Proxmox VE is not among the returned Pulumi Registry Results. Without a dedicated Proxmox VE provider, one common way to manage Proxmox resources with Pulumi is to use the Pulumi Command provider to run Proxmox command-line tools or API calls within the Pulumi program.

Below is a tutorial and corresponding TypeScript program demonstrating how you might use the Pulumi Command provider to provision a VM on Proxmox VE. The program utilizes Pulumi’s Command module to interact with the Proxmox VE CLI or API. This method assumes that you have access to a Proxmox VE environment where you have permissions to create VMs.

Prerequisites:

Install Pulumi CLI
Install Node.js
A Proxmox VE installation that you have access to via CLI or API.
Credentials configured for accessing the Proxmox environment.

Tutorial:

First, we set up the Command provider which will help us to run system commands from the Pulumi program. This package is useful when a Pulumi provider for your target system does not exist, as it executes shell commands as part of the resource lifecycle.
Next, we define the VM resources we want to create using command-line syntax or API requests for Proxmox VE. In the code below, the specific VM creation command will depend on Proxmox VE's command-line interface or API, which you would need to replace with the actual commands or API calls.
To create resources on Proxmox VE, we'll encapsulate the actions inside Pulumi resources. Since this might involve multiple steps, you can define each step as a separate Pulumi Command resource, with dependencies between the steps represented by the dependsOn option in each resource.
Finally, the Pulumi program will be executed to perform the necessary command(s) to provision the VM on Proxmox VE.

Here's a TypeScript program that illustrates this process:

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

// Example of Provisioning a VM on Proxmox VE using the Pulumi Command package
// Note: You need to replace the placeholder command strings with actual Proxmox CLI commands or API calls.

// Initialize a Pulumi stack
const stack = new pulumi.Stack("proxmox-vm");

// Create a VM on Proxmox VE using a shell command
const createVm = new command.local.Command("createVm", {
    // Replace this command with the actual command you would use to provision a VM on Proxmox VE
    create: `pvesh create /nodes/{node}/qemu -vmid {vmid} -name {name} -memory {memory} -cores {cores}`,
    // Add additional commands as needed for updates and deletes
    update: `pvesh set /nodes/{node}/qemu/{vmid} -name {newName}`,
    delete: `pvesh delete /nodes/{node}/qemu/{vmid}`,

    // If required, set any environment variables needed for execution
    environment: {
        "PROXMOX_API_TOKEN": "<Your Proxmox API Token>",
    },
}, {dependsOn: [/* any dependencies that need to be provisioned first, if any */]});

// Output the result of the command
export const vmCreationResult = createVm.stdout;

// Export other relevant information like VM ID, Node or any network information as needed

In the program above:

We import the necessary Pulumi libraries.
We define a Command resource that issues commands to Proxmox VE to create, update, and delete a VM.
We specify the environment variable PROXMOX_API_TOKEN which is expected to contain your Proxmox API token for authentication. Ensure to replace <Your Proxmox API Token> with your actual API token.
We export the VM creation result for visibility into what the command returned, which is useful for debugging.
We also mention the possibility to export other relevant pieces of information about the VM, such as IDs or networking information, depending on the actual command outputs.

Remember, the create, update, and delete command strings need to be filled with the real commands you would use for managing VMs in your Proxmox VE environment. This example assumes that the pvesh utility is used to interact with Proxmox VE -- substitute with the appropriate CLI or API calls for your environment.