Shape

Contributing

Contributing Community

Terratest is an open source project, and contributions from the community are very welcome! Please check out the Contribution Guidelines and Developing Terratest for instructions.

Contribution Guidelines

Contributions to this repo are very welcome! We follow a fairly standard pull request process for contributions, subject to the following guidelines:

  1. Types of contributions
  2. File a GitHub issue
  3. Update the documentation
  4. Update the tests
  5. Update the code
  6. Create a pull request
  7. Merge and release

Types of contributions

Broadly speaking, Terratest contains two types of helper functions:

  1. Integrations with external tools
  2. Infrastructure and validation helpers

We accept different types of contributions for each of these two types of helper functions, as described next.

Integrations with external tools

These are helper functions that integrate with various DevOps tools—e.g., Terraform, Docker, Packer, and Kubernetes—that you can use to deploy infrastructure in your automated tests. Examples:

  • terraform.InitAndApply: run terraform init and terraform apply.
  • packer.BuildArtifacts: run packer build.
  • shell.RunCommandAndGetOutput: run an arbitrary shell command and return stdout and stderr as a string.

Here are the guidelines for contributions with external tools:

  1. Fixes and improvements to existing integrations: All bug fixes and new features for existing tool integrations are very welcome!

  2. New integrations: Before contributing an integration with a totally new tool, please file a GitHub issue to discuss with us if it’s something we are interested in supporting and maintaining. For example, we may be open to new integrations with Docker and Kubernetes tools, but we may not be open to integrations with Chef or Puppet, as there are already testing tools available for them.

Infrastructure and validation helpers

These are helper functions for creating, destroying, and validating infrastructure directly via API calls or SDKs. Examples:

  • http_helper.HttpGetWithRetry: make an HTTP request, retrying until you get a certain expected response.
  • ssh.CheckSshCommand: SSH to a server and execute a command.
  • aws.CreateS3Bucket: create an S3 bucket.
  • aws.GetPrivateIpsOfEc2Instances: use the AWS APIs to fetch IPs of some EC2 instances.

The number of possible such helpers is nearly infinite, so to avoid Terratest becoming a gigantic, sprawling library we ask that contributions for new infrastructure helpers are limited to:

  1. Platforms: we currently only support three major public clouds (AWS, GCP, Azure) and Kubernetes. There is some code contributed earlier for other platforms (e.g., OCI), but until we have the time/resources to support those platforms fully, we will only accept contributions for the major public clouds and Kubernetes.

  2. Complexity: we ask that you only contribute infrastructure and validation helpers for code that is relatively complex to do from scratch. For example, a helper that merely wraps an existing function in the AWS or GCP SDK is not a great choice, as the wrapper isn’t contributing much value, but is bloating the Terratest API. On the other hand, helpers that expose simple APIs for complex logic are great contributions: ssh.CheckSshCommand is a great example of this, as it provides a simple one-line interface for dozens of lines of complicated SSH logic.

  3. Popularity: Terratest should only contain helpers for common use cases that come up again and again in the course of testing. We don’t want to bloat the library with lots of esoteric helpers for rarely used tools, so here’s a quick litmus test: (a) Is this helper something you’ve used once or twice in your own tests, or is it something you’re using over and over again? (b) Does this helper only apply to some use case specific to your company or is it likely that many other Terratest users are hitting this use case over and over again too?

  4. Creating infrastructure: we try to keep helper functions that create infrastructure (e.g., use the AWS SDK to create an S3 bucket or EC2 instance) to a minimum, as those functions typically require maintaining state (so that they are idempotent and can clean up that infrastructure at the end of the test) and dealing with asynchronous and eventually consistent cloud APIs. This can be surprisingly complicated, so we typically recommend using a tool like Terraform, which already handles all that complexity, to create any infrastructure you need at test time, and running Terratest’s built-in terraform helpers as necessary. If you’re considering contributing a function that creates infrastructure directly (e.g., using a cloud provider’s APIs), please file a GitHub issue to explain why such a function would be a better choice than using a tool like Terraform.

File a GitHub issue

Before starting any work, we recommend filing a GitHub issue in this repo. This is your chance to ask questions and get feedback from the maintainers and the community before you sink a lot of time into writing (possibly the wrong) code. If there is anything you’re unsure about, just ask!

Update the documentation

We recommend updating the documentation before updating any code (see Readme Driven Development). This ensures the documentation stays up to date and allows you to think through the problem at a high level before you get lost in the weeds of coding.

The documentation is built with Jekyll and hosted on the Github Pages from docs folder on master branch. Check out Terratest website to learn more about working with the documentation.

Update the tests

We also recommend updating the automated tests before updating any code (see Test Driven Development). That means you add or update a test case, verify that it’s failing with a clear error message, and then make the code changes to get that test to pass. This ensures the tests stay up to date and verify all the functionality in this Module, including whatever new functionality you’re adding in your contribution. The instructions for running the automated tests can be found here.

Update the code

At this point, make your code changes and use your new test case to verify that everything is working. As you work, please make every effort to avoid unnecessary backwards incompatible changes. This generally means that you should not delete or rename anything in a public API.

If a backwards incompatible change cannot be avoided, please make sure to call that out when you submit a pull request, explaining why the change is absolutely necessary.

Note that we use pre-commit hooks with this project. To ensure they run:

  1. Install pre-commit.
  2. Run pre-commit install.

One of the pre-commit hooks we run is goimports. To prevent the hook from failing, make sure to :

  1. Install goimports
  2. Run goimports -w ..

We have a style guide for the Go programming language, in which we documented some best practices for writing Go code. Please ensure your code adheres to the guidelines outlined in the guide.

Create a pull request

Create a pull request with your changes. Please make sure to include the following:

  1. A description of the change, including a link to your GitHub issue.
  2. The output of your automated test run, preferably in a GitHub Gist. We cannot run automated tests for pull requests automatically due to security concerns, so we need you to manually provide this test output so we can verify that everything is working.
  3. Any notes on backwards incompatibility or downtime.

Validate the Pull Request for Azure Platform

If you’re contributing code for the Azure Platform and if you have an active Azure subscription, it’s recommended to follow the below guidelines after creating a pull request. If you’re contributing code for any other platform (e.g., AWS, GCP, etc), you can skip these steps.

Once the Terratest maintainers add Azure tag and Approve the PR, following pipeline will run automatically to perform a full validation of the Azure contribution. You also can run the pipeline manually on your forked repo by following the below guideline.

We have a separate CI pipeline for Azure code. To run it on a forked repo:

  1. Run the following Azure Cli command on your preferred Terminal to create Azure credentials and copy the output:

     az ad sp create-for-rbac --name "terratest-az-cli" --role contributor --sdk-auth
    
  2. Go to Secrets settings page under Settings tab in your forked project, https://github.com/<YOUR_GITHUB_ACCOUNT>/terratest/settings, on GitHub.

  3. Create a new Secret named AZURE_CREDENTIALS and paste the Azure credentials you copied from the 1st step as the value

    AZURE_CREDENTIALS will be stored in your GitHub account; neither the Terratest maintainers nor anyone else will have any access to it. Under the hood, GitHub stores your secrets in a secure, encrypted format (see: GitHub Actions Secrets Reference for more information). Once the secret is created, it’s only possible to update or delete it; the value of the secret can’t be viewed. GitHub uses a libsodium sealed box to help ensure that secrets are encrypted before they reach GitHub.

  4. Create a new Personal Access Token (PAT) page under Settings / Developer Settings, making sure write:discussion and public_repo scopes are checked. Click the Generate token button and copy the generated PAT.

  5. Go back to settings/secrets in your fork and Create a new Secret named PAT. Paste the output from the 4th step as the value

    PAT will be stored in your GitHub account; neither the Terratest maintainers nor anyone else will have any access to it. Under the hood, GitHub stores your secrets in a secure, encrypted format (see: GitHub Actions Secrets Reference for more information). Once the secret is created, it’s only possible to update or delete it; the value of the secret can’t be viewed. GitHub uses a libsodium sealed box to help ensure that secrets are encrypted before they reach GitHub.

  6. Go to Actions tab on GitHub (https://github.com//terratest/actions)

  7. Click ci-workflow workflow

  8. Click Run workflow button and fill the fields in the drop down
    • Repository Info : name of the forked repo (e.g. xyz/terratest)
    • Name of the branch : branch name on the forked repo (e.g. feature/adding-some-important-module)
    • Name of the official terratest repo : home of the target pr (gruntwork-io/terratest)
    • PR number on the official terratest repo : pr number on the official terratest repo (e.g. 14, 25, etc.). Setting this value will leave a success/failure comment in the PR once CI completes execution.

    • Skip provider registration : set true if you want to skip terraform provider registration for debug purposes (false or true)
  9. Wait for the ci-workflow to be finished

    The pipeline will use the given Azure subscription and deploy real resources in your Azure account as part of running the test. When the tests finish, they will tear down the resources they created. Of course, if there is a bug or glitch that prevents the clean up code from running, some resources may be left behind, but this is rare. Note that these resources may cost you money! You are responsible for all charges in your Azure subscription.

  10. PR with the given PR Number will have the result of the ci-workflow as a comment

Merge and release

The maintainers for this repo will review your code and provide feedback. Once the PR is accepted, they will merge the code and release a new version, which you’ll be able to find in the releases page.

Developing Terratest

  1. Running tests
  2. Versioning
  3. Developing For Azure

Running tests

Terratest itself includes a number of automated tests.

Note #1: Some of these tests create real resources in an AWS account. That means they cost money to run, especially if you don’t clean up after yourself. Please be considerate of the resources you create and take extra care to clean everything up when you’re done!

Note #2: In order to run tests that access your AWS account, you will need to configure your AWS CLI credentials. For example, you could set the credentials as the environment variables AWS_ACCESS_KEY_ID and AWS_SECRET_ACCESS_KEY.

Note #3: Never hit CTRL + C or cancel a build once tests are running or the cleanup tasks won’t run!

Prerequisite: The tests expect Terraform, Terragrunt, Packer, and/or Docker to already be installed and in your PATH.

To run all the tests:

go test -v -timeout 30m -p 1 ./...

To run the tests in a specific folder:

cd "<FOLDER_PATH>"
go test -timeout 30m

To run a specific test in a specific folder:

cd "<FOLDER_PATH>"
go test -timeout 30m -run "<TEST_NAME>"

Versioning

This repo follows the principles of Semantic Versioning. You can find each new release, along with the changelog, in the Releases Page.

During initial development, the major version will be 0 (e.g., 0.x.y), which indicates the code does not yet have a stable API. Once we hit 1.0.0, we will make every effort to maintain a backwards compatible API and use the MAJOR, MINOR, and PATCH versions on each release to indicate any incompatibilities.

Developing For Azure

Azure supports multliple cloud environments. In order to properly register the correct environment for you test code, you need to use the Azure SDK Client Factory.

Azure SDK Client Factory

This documentation provides and overview of the client_factory.go module, targeted use cases, and behaviors. This module is intended to provide support for and simplify working with Azure’s multiple cloud environments (Azure Public, Azure Government, Azure China, Azure Germany and Azure Stack). Developers looking to contribute to additional support for Azure to Terratest should leverage client_factory and use the patterns below to add a resource REST client from Azure Go SDK. By doing so, it provides a consistent means for developers using Terratest to test their Azure Infrastructure to connect to the correct cloud and its associated REST apis.

Background

The Azure REST APIs support both Public and sovereign cloud environments (at the moment this includes Public, US Government, Germany, China, and Azure Stack environments). If you are interacting with an environment other than public cloud, you need to set the base URI for the Azure REST API you are interacting with.

Base URI

You must use the correct base URI’s for the Azure REST API’s (either directly or via Azure SDK for GO) to communicate with a cloud environment other than Azure Public. The Azure Go SDK supports this by using the WithBaseURI suffixed calls when creating service clients. For example, when using the VirtualMachinesClient with the public cloud, a developer would normally write code for the public cloud like so:

import (
    "github.com/Azure/azure-sdk-for-go/services/compute/mgmt/2019-07-01/compute"
)

func SomeVMHelperMethod() {
    subscriptionID := "your subscription ID"

    // Create a VM client and return
    vmClient, err := compute.NewVirtualMachinesClient(subscriptionID)

    // Use client / etc
}

However, this code will not work in non-Public cloud environments as the REST endpoints have different URIs depending on environment. Instead, you need to use an alternative method (provided in the Azure REST SDK for Go) to get a properly configured client (all REST API clients should support this alternate method):

import (
    "github.com/Azure/azure-sdk-for-go/services/compute/mgmt/2019-07-01/compute"
)

func SomeVMHelperMethod() {
    subscriptionID := "your subscription ID"
    baseURI := "management.azure.com"

    // Create a VM client and return
    vmClient, err := compute.NewVirtualMachinesClientWithBaseURI(baseURI, subscriptionID)

    // Use client / etc
}

Using code similar to above, you can communicate with any Azure cloud environment just by changing the base URI that is passed to the clients (Azure Public shown in above example).

Lookup Environment Metadata

Developers MUST avoid hardcoding these base URI’s. Instead, they should be looked up from an authoritative source. The AutoRest-GO library (used by the Go SDK) provides such functionality. The client_factory module makes use of the AutoRest EnvironmentFromName(envName string) function to return the appropriate structure. This method and Environment structure is documented on GoDoc here.

To configure different cloud environments, we will use the same AZURE_ENVIRONMENT environment variable that the Go SDK uses. This can currently be set to one of the following values:

Value Cloud Environment
“AzureChinaCloud” ChinaCloud
“AzureGermanCloud” GermanCloud
“AzurePublicCloud” PublicCloud
“AzureUSGovernmentCloud” USGovernmentCloud
“AzureStackCloud” Azure stack

When using the “AzureStackCloud” setting, you MUST also set the AZURE_ENVIRONMENT_FILEPATH variable to point to a JSON file containing your Azure Stack URI details.

Putting it all together

client_factory implements this pattern described above in order to instantiate and return properly configured REST SDK for GO clients so that test implementers don’t have to consider REST API client implementation as long as they have the correct AZURE_ENVIRONMENT env setting. If this environment variable is not set, the client will assume public cloud as the cloud environment to communicate with. We strongly recommend developers creating Terratest helper methods for Azure use this pattern with client factory to create REST API clients. This will reduce effort for Terratest users creating test for Azure resources.

Note the following:

  • TERRAFORM uses ARM_ENVIRONMENT environment variable to set the correct cloud environment.
  • The default behavior of the client_factory is to use the AzurePublicCloud environment. This requires no work from the developer to configure, and ensures consistent behavior with the current SDK code.
Wait, I don’t see the client in client factory for the rest api I want to interact with

If you require a client that is not already implemented in client factory for your helper method, you will need to create a corresponding method that instantiates the client and accepts base URI following the patterns discussed. Below is a walkthrough for adding a client to client factory.

Walkthrough, adding a client to client_factory
Add your client namespace to client factory

In the Azure SDK for GO, each service should have a module that implements that services client. You can find the correct module here. Add that module to the client factory imports. Below is an example for client imports that shows clients for compute, container service and subscriptions.

Loading...
Add your client method to instantiate the client

The next step is to add your method to instantiate the client. Below is an example of adding the method to create a client for Virtual Machines, note that we lookup the environment using getEnvironmentEndpointE and then pass that base URI to the actual method on the Virtual Machines Module to create the client NewVirtualMachinesClientWithBaseURI.

Loading...
Add a unit test to client_factory_test.go

In order to ensure that your CreateClient method works properly, add a unit test to client_factory_test.go. The unit test MUST assert that the base URI is correctly set for your client. Some key points for writing your unit test are:

  • Use table-driven testing to test the various combinations of cloud environments
  • Give the test case a descriptive name so it is easy to identify which test failed.
  • PRs will be rejected if a client is added without a corresponding unit test.

Below is an example of the Virtual Machines client unit test:

Loading...
Use your CreateClient method in your helper

We now can use this client creation method in our helpers to create a Virtual Machines client. Below is an example for how to call into this create method from client_factory:

Loading...

Built by Gruntwork

Your entire infrastructure. Defined as code. In about a day.

Explore Gruntwork.io