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 datablock for aws_db_subnet_group should support lookup by tag. If you create the group without a name, it gets a terraform generated ID which is not friendly, particularly in multi-region, multi-account envs.

This usually leads you to create a big if/else block enumerating all the subnet group names. Additionally, you can't move an RDS instance to another db subnet group in the same VPC - once you start using the group it's difficult to migrate.

If we could simply look the aws_db_subnet_group up using a filter similar to this data source we could pretty easily mitigate this issue by adding tags to the group (funny enough, the Name tag is likely what we'd use) then using a data block like:

data "aws_db_subnet_group" "my_group" {
  filter {
    name   = "tag:Name"
    values = ["yakdriver"]
  }
}

Name should also be added as an output attribute to the data source, it currently isn't. I assume that is because you don't need the lookup if you already know the name.

New or Affected Resource(s)

• aws_db_subnet_group data block

Potential Terraform Configuration

data "aws_db_subnet_group" "my_group" {
  filter {
    name   = "tag:Name"
    values = ["yakdriver"]
  }
}

resource "aws_db_instance" "my_instance" {
     ...
     db_subnet_group_name = data.aws_db_subnet_group.my_group.name
     ...
}

References

• #0000