Skip to content

QuickSight Dataset "UseAs" Support and Terraform Limitations | Automation limitation for QuickSight Rules datasets

0

Reaching out for a customer who is a big user of Amazon QuickSight datasets. As per the documentation for the new "UseAs" key for datasets, there is only couple of options available when creating a dataset, either AWS Console or the CLI. Even then, the available paths seem limited to either: Creating the dataset directly as an RLS rules dataset, or Duplicating an existing dataset directly as a rules dataset. This has become a problem for the Cx because they rely heavily on Terraform to manage and maintain their infrastructure. At the moment, Terraform does not support the UseAs attribute for QuickSight datasets. There is already an open issue in the HashiCorp repository requesting this feature, but it has been open for quite some time without progress, so it’s unclear when (or if) this will be supported.

https://github.com/hashicorp/terraform-provider-aws/issues/43103 Additionally, the update-data-set CLI command also does not support setting or updating the UseAs key, which further limits our automation options. Looking for a recommended workarounds for customers managing QuickSight via IaC. Can anyone share roadmap or future support for this feature (either in Terraform or the AWS APIs) ? Any guidance or additional context you can share would be greatly appreciated.

AWS

asked 10 months ago146 views

1 Answer
0

Based on the available information, the "UseAs" parameter for Amazon QuickSight datasets currently has a valid value of "RLS_RULES" (Row-Level Security Rules). This parameter indicates the usage purpose of the dataset.

You're correct that there are limitations in how this parameter can be set and managed through infrastructure as code. The current AWS CLI and API documentation doesn't show support for setting or updating the "UseAs" attribute through the update-data-set command. The CreateDataSet API operation example provided doesn't explicitly show how to set this parameter either.

For customers heavily relying on Terraform for infrastructure management, this presents a challenge since the Terraform AWS provider doesn't currently support the UseAs attribute for QuickSight datasets, as you mentioned.

As for workarounds, you might consider:

  1. Using a hybrid approach where most of the dataset configuration is managed through Terraform, but the specific "UseAs" attribute is set manually or through custom scripts using the AWS CLI or SDK after the initial creation.

  2. Creating custom scripts that use the AWS SDK to create datasets with the required "UseAs" attribute, which can be integrated into your deployment pipeline.

  3. Using AWS CloudFormation with custom resources that can invoke Lambda functions to set these specific attributes that aren't directly supported.

Unfortunately, I don't have information about the roadmap for supporting this feature in Terraform or future AWS API enhancements. I would recommend continuing to follow the GitHub issue you referenced and reaching out to AWS Support for more specific guidance on the roadmap for this feature.
Sources
DataSet - Amazon Quick Sight
DataSetSummary - Amazon Quick Sight
CreateDataSet - Amazon Quick Sight

answered 10 months ago

You are not logged in. Log in to post an answer.

A good answer clearly answers the question and provides constructive feedback and encourages professional growth in the question asker.