Terraform CDK
Learn how to use the Doppler Terraform Provider with Terraform CDK.
This guide will show you how to get the Doppler Terraform Provider working with Terraform CDK.
Prerequisites
- The Terraform CLI (1.1+).
- Node.js and npm v16+.
- A current version of Typescript, Python, Go, C#, or Java (this guide will be showing Typescript examples, but examples for other languages are available in Terraform's CDKTF tutorial).
This guide follows the basic outline from Terraform's CDKTF tutorial. If something goes wrong or you need more details, please try checking out their full tutorial!
Install CDKTF
You can install the CDKTF CLI using npm
on most operating systems. It's also available via Homebrew on MacOS systems.
npm install --global cdktf-cli@latest
brew install cdktf
Create a new project
Create a new directory for the test project and then switch to that directory:
mkdir doppler-terraform-cdk
cd doppler-terraform-cdk
Now initialize that project using the cdktf
CLI tool:
# if you're using another language than Typescript, be sure
# to specify the one you're using here.
cdktf init --template=typescript --local
Finally, add the Doppler provider:
cdktf provider add dopplerhq/doppler
We don't currently have a pre-built provider, so the CLI will generate one based on our Terraform provider. You should see output similar to this:
Pre-built provider does not exist for the given constraints.
Adding local provider registry.terraform.io/dopplerhq/doppler with version constraint undefined to cdktf.json
Local providers have been updated. Running cdktf get to update...
β starting...
[2023-01-11T14:04:51.335] [INFO] default - Could not find constraints.json file while filtering: Error: ENOENT: no such file or directory, open '/Users/watsonia
Generated typescript constructs in the output directory: .gen
Tweak Generated Provider
This step isn't strictly required, but will make using the provider easier. Open up the file at .gen/providers/doppler/index.ts
and remove the as X
from each import line. So, you're changing what looks like this:
// generated by cdktf get
export * as config from './config';
export * as environment from './environment';
export * as project from './project';
export * as secret from './secret';
export * as serviceToken from './service-token';
export * as dataDopplerSecrets from './data-doppler-secrets';
export * as provider from './provider';
to something like this:
// generated by cdktf get
export * from './config';
export * from './environment';
export * from './project';
export * from './secret';
export * from './service-token';
export * from './data-doppler-secrets';
export * from './provider';
Edit the Code
Open the main.ts
file in the project and modify it so it looks like this:
import { Construct } from "constructs";
import { App, TerraformStack } from "cdktf";
import * as doppler from "./.gen/providers/doppler";
class MyStack extends TerraformStack {
constructor(scope: Construct, name: string) {
super(scope, name);
new doppler.DopplerProvider(this, "doppler", {
dopplerToken: `${process.env.DOPPLER_ACCESS_TOKEN}`,
});
const testProject = new doppler.Project(this, "terraform-cdk", {
name: "terraform-cdk",
description:
"This is a project for testing terraform-cdk.",
});
new doppler.Environment(this, "test", {
name: "test",
project: testProject.name,
slug: "test",
});
new doppler.Environment(this, "ci", {
name: "ci",
project: testProject.name,
slug: "ci",
});
new doppler.Environment(this, "qa", {
name: "qa",
project: testProject.name,
slug: "qa",
});
}
}
const app = new App();
new MyStack(app, "doppler");
app.synth();
Deploy using cdktf
After doing that, try deploying your changes by running the following:
DOPPLER_ACCESS_TOKEN=$(doppler configure get token --plain) cdktf deploy
This will execute the CLI command using the same token your local Doppler CLI uses. Before this does anything, you'll be presented with a terraform plan that shows what will happen with the option to proceed or not:
Terraform will perform the following actions:
doppler # doppler_environment.ci (ci) will be created
+ resource "doppler_environment" "ci" {
+ id = (known after apply)
+ name = "ci"
+ project = "terraform-cdk"
+ slug = "ci"
}
# doppler_environment.qa (qa) will be created
+ resource "doppler_environment" "qa" {
+ id = (known after apply)
+ name = "qa"
+ project = "terraform-cdk"
+ slug = "qa"
}
# doppler_environment.test (test) will be created
+ resource "doppler_environment" "test" {
+ id = (known after apply)
+ name = "test"
+ project = "terraform-cdk"
+ slug = "test"
}
# doppler_project.terraform-cdk (terraform-cdk) will be created
+ resource "doppler_project" "terraform-cdk" {
+ description = "This is a project for testing terraform-cdk."
+ id = (known after apply)
+ name = "terraform-cdk"
}
Plan: 4 to add, 0 to change, 0 to destroy.
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
Saved the plan to: plan
To perform exactly these actions, run the following command to apply:
terraform apply "plan"
Please review the diff output above for doppler
β― Approve Applies the changes outlined in the plan.
Dismiss
Stop
After applying, you should see the project and associated environments in your Doppler dashboard!
When you deploy this, it creates Terraform state files in the project directory. If you followed the above, then those files will be called
terraform.doppler.tfstate
andterraform.doppler.tfstate.backup
. If you delete the project and want to re-apply from scratch, you would also need to delete these files.
Updated 8 months ago