TFE: API current state version output errors with "503 Please wait while outputs are parsed from the latest state version" – HashiCorp Help Center

Problem

An API call to show the current state version outputs for a workspace returns a 503 service unavailable error.

Example Request:

$ curl \
  --header "Authorization: Bearer $TOKEN" \
  --header "Content-Type: application/vnd.api+json" \
  https://app.terraform.io/api/v2/workspaces/ws-XXXX/current-state-version-outputs

Error Response:

{
  "errors": [
    {
      "status": "503",
      "title": "service unavailable",
      "detail": "Please wait while outputs are parsed from the latest state version",
      "retry_after_sec": "3"
    }
  ]
}

Prerequisites

Terraform Enterprise (TFE) on Kubernetes, version 1.0.1 or 1.0.2.
The TFE application has been running for more than 24 hours.
You have waited a reasonable amount of time for the outputs to parse, based on the number of resources in the workspace.

Cause

This error is caused by an internal bug where the resources-processed flag in the state version does not update to true.

You can check the value of this flag using the API.

Example Request:

$ curl \
  --header "Authorization: Bearer $TOKEN" \
  --header "Content-Type: application/vnd.api+json" \
  https://app.terraform.io/api/v2/workspaces/<ws-XXXX>/current-state-version

Example Output:

{
  // ...
  "resources-processed": false,
  // ...
}

Solutions

This issue is permanently resolved by upgrading to Terraform Enterprise v1.0.3 or newer. If you cannot upgrade immediately, you may use one of the following workarounds.

Solution 1: Disable Usage Reporting (Preferred Workaround)

In your Terraform Enterprise configuration file, add the environment variable TFE_USAGE_REPORTING_DISABLED=true.

Note: This will disable TFE product usage reporting.

Solution 2: Restart the Terraform Enterprise Pod

You can temporarily resolve the issue by restarting the terraform-enterprise pod.

Note: This workaround must be repeated each time the error reappears.

Solution 3: Restart the Sidekiq Process

Restart the sidekiq process inside the Terraform Enterprise container.

# supervisorctl restart tfe:sidekiq

Note: This workaround must be repeated each time the error reappears.

Solution 4: Upgrade Terraform Enterprise (Permanent Fix)

Upgrade your Terraform Enterprise instance to version 1.0.3 or newer to permanently resolve this bug.

Outcome

After applying a solution, the API call to retrieve workspace outputs will succeed.

Additional Information

API Documentation: Show current state version outputs for a workspace