Drift Detection for AppSync Schema Changes

[AlexEshoo]

Community Note

• Please vote on this issue by adding a 👍 reaction to the original issue to help the community and maintainers prioritize this request
• Please do not leave "+1" or other comments that do not add relevant new information or questions, they generate extra noise for issue followers and do not help prioritize the request
• If you are interested in working on this issue or have submitted a pull request, please leave a comment

Description

The documentation for the aws_appsync_graphql_api notes that there is no drift detection for schema changes:

schema - (Optional) The schema definition, in GraphQL schema language format. Terraform cannot perform drift detection of this configuration.

This causes some problems/confusion when changes are made in the AWS console but not on the local state.

It looks like the AWS API provides an endpoint for getting the current schema for a graphql API in appsync: https://docs.aws.amazon.com/appsync/latest/APIReference/API_GetIntrospectionSchema.html

Would this allow drift detection to be added to the terraform provider?

New or Affected Resource(s)

• aws_appsync_graphql_api

Potential Terraform Configuration

resource "aws_appsync_graphql_api" "my_graphql_api" {
  authentication_type = "AMAZON_COGNITO_USER_POOLS"
  name = "myAPI"
  schema = data.local_file.graphql_schema.content
  user_pool_config {
    default_action = "ALLOW"
    user_pool_id = local.cognito_user_pool_id
  }
}

[nikunjundhad]

This feature is more required when you have multiple contributor with console and terraform code access for your appsync api. Person A directly change the schema in console for quick testing purpose and then forgot to remove those unintentional changes. And these changes are not showing as a difference during terraform planning and execution cycle and we lost track of change management. When Person B need to change schema which is required this will start creating confusion and he/she don't know what is the correct schema to keep. One showing in console or another which is in terraform infra code. Please add this feature to save this pain.

[ryancausey]

I just ran into this. I was assuming I could make test changes in the console in our nonprod environment, and then re-run the Terraform pipeline to reset those changes. It turns out this is not the case. It looks like this issue has been open for 4 years so far. What is the likelihood of this being picked up and fixed?