README.md

Create Run Action

Requires that the workspace already exists within HCP Terraform.

Performs an API driven run in HCP Terraform, using a configuration version and the workspace's current variables. If a configuration version is not provided, will default to the workspace's most recently used version. HCP Terraform Create-Run API.

This action will wait until the run status has reached a terminal/final state.

HCP Terraform API-driven Run Workflow

Behaviors / Expected Outcomes

• If the Plan reaches a failured state, will return Error status and status code of 1.
• If the Workspace is locked, will return a Error status and status code of 1.
• If the Plan timesout once TF_MAX_TIMEOUT has been reached, will return a Timeout and status code of 1. Does not mean the Plan will fail, however the action will stop waiting for the Plan to complete.
• A successful Plan will return a Success and status code of 0.

Action Inputs / Outputs

To view all available inputs and outputs, see ./action.yml file.

Variables

TF_VAR_* Environment variable(s)

HCP Terraform API runs allow you to specify a list of variables using key and value attributes. These variables can be configured using environment variables with the TF_VAR_* naming convention. These variables take precedence over variables with the same key applied to the workspace(e.g., variable sets). For greater flexibility and features, it's recommended to use HCP Terraform workspace variables and variable sets.

Note

All values must be expressed as an HCL literal in the same syntax you would use when writing Terraform code. Variables expressed as string will need to be escaped to include double quotations like, TF_VAR_image_id="\"ami-my-custom-image\"" or TF_VAR_image_id='"ami-my-custom-image'". Read more on Terraform complex typed values

In your GitHub Action workflow file, you can specify variables using the env keyword at either the job or workflow level.

env:
  TF_VAR_image_id="\"ami-abc123\"" ## string variable
  TF_VAR_user_map='{"1":{"name":"jason bourne","size":2}}'` # map of objects variable
  TF_VAR_availability_zone_names='["us-east-1a","us-west-1c", "us-west-2b"]'` # list of strings variable

Your *.tf configuration

variable "image_id" {
  type = string
}

variable "availability_zone_names" {
  type    = list(string)
  default = ["us-west-1a"]
}


variable "user_map" {
  type = map(object({
    name = string
    size = number
  }))
}

All values must be expressed as an HCL literal in the same syntax you would use when writing Terraform code

Example Usage

Like with the other actions, you can specify hostname, token, and organization as inputs OR as environment variables.

This specific example does not pass a configuraiton version id, so the run defaults to the workspace's most recently used version. Read more about create run behavior in the HCP Terraform API

- uses: hashicorp/tfc-workflows-github/actions/[email protected]
  id: run
  with:
    workspace: "my-workspace"
    plan_only: true
    message: "Plan Run from GitHub Actions"
    ## Can specify hostname,token,organization as direct inputs
    hostname: ""
    organization: ${{ vars.TF_CLOUD_ORGANIZATION }} # Configured as GitHub configuration variable
    token: ${{ secrets.TF_API_TOKEN }} # Configured as GitHub Secret
    ## OR
    # env:
      # TF_CLOUD_ORGANIZATION: ${{ vars.TF_CLOUD_ORGANIZATION }} # Configured as GitHub configuration variable
      # TF_API_TOKEN: ${{ secrets.TF_API_TOKEN }} # Configured as GitHub Secret

Creating a new run with recently uploaded configuration in a previous step

- uses: hashicorp/tfc-workflows-github/actions/[email protected]
  id: upload
  with:
    workspace: ${{ env.TF_WORKSPACE }}
    directory: ${{ env.TF_DIRECTORY }}

- uses: hashicorp/tfc-workflows-github/actions/[email protected]
  id: run
  with:
    workspace: ${{ env.TF_WORKSPACE }}
    configuration_version: ${{ steps.upload.outputs.configuration_version_id }}