Changes to aws_db_parameter_group variable calculations rejected by AWS for not matching capitalization are still persisted to state

[handlerbot]

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

Terraform CLI and Terraform AWS Provider Version

> terraform -v
Terraform v0.13.1
+ provider registry.terraform.io/hashicorp/aws v3.4.0

Affected Resource(s)

• aws_db_parameter_group

Terraform Configuration Files

resource "aws_db_parameter_group" "x" {
  name        = "x"
  family      = "mysql5.7"

  parameter {
    name         = "innodb_buffer_pool_size"
    value        = "{DBInstanceClassmemory*8/19}"
    apply_method = "pending-reboot"
  }
}

Expected Behavior

Changes to the RDS parameter group not accepted by AWS are not persisted in the state file.

Actual Behavior

Changes to the RDS parameter group not accepted by AWS are persisted in the state file, overwriting the prior, still-retained configuration.

Steps to Reproduce

We start with having created the RDS parameter group with the config above:

> terraform state show aws_db_parameter_group.x
# aws_db_parameter_group.x:
resource "aws_db_parameter_group" "x" {
    arn         = "arn:aws:rds:us-west-2:<account_id>:pg:x"
    description = "Managed by Terraform"
    family      = "mysql5.7"
    id          = "x"
    name        = "x"

    parameter {
        apply_method = "pending-reboot"
        name         = "innodb_buffer_pool_size"
        value        = "{DBInstanceClassMemory*8/19}"
    }
}

Now, edit the above resource configuration to change {DBInstanceClassMemory*8/19} to {dBInstanceClassMemory*1/2}.
NOTE THAT THE FIRST LETTER IN THE STRING HAS BEEN LOWERCASED, D -> d

> terraform apply
aws_db_parameter_group.x: Refreshing state... [id=x]

An execution plan has been generated and is shown below.
Resource actions are indicated with the following symbols:
  ~ update in-place

Terraform will perform the following actions:

  # aws_db_parameter_group.x will be updated in-place
  ~ resource "aws_db_parameter_group" "x" {
        arn         = "arn:aws:rds:us-west-2:<account_id>:pg:x"
        description = "Managed by Terraform"
        family      = "mysql5.7"
        id          = "x"
        name        = "x"
        tags        = {}

      + parameter {
          + apply_method = "pending-reboot"
          + name         = "innodb_buffer_pool_size"
          + value        = "{dBInstanceClassMemory*1/2}"
        }
      - parameter {
          - apply_method = "pending-reboot" -> null
          - name         = "innodb_buffer_pool_size" -> null
          - value        = "{dbinstanceclassmemory*8/19}" -> null
        }
    }

Plan: 0 to add, 1 to change, 0 to destroy.

Do you want to perform these actions?
  Terraform will perform the actions described above.
  Only 'yes' will be accepted to approve.

  Enter a value: yes

aws_db_parameter_group.x: Modifying... [id=x]

Error: Error modifying DB Parameter Group: InvalidParameterValue: value does not match pattern
	status code: 400, request id: 0db15cce-235a-4d25-b1ac-556b453b3241

> terraform state list
aws_db_parameter_group.x

> terraform state show aws_db_parameter_group.x
# aws_db_parameter_group.x:
resource "aws_db_parameter_group" "x" {
    arn         = "arn:aws:rds:us-west-2:<account_id>:pg:x"
    description = "Managed by Terraform"
    family      = "mysql5.7"
    id          = "x"
    name        = "x"
    tags        = {}

    parameter {
        apply_method = "pending-reboot"
        name         = "innodb_buffer_pool_size"
        value        = "{dBInstanceClassMemory*1/2}"
    }
}

> terraform plan -refresh=false

No changes. Infrastructure is up-to-date.
[...]

> terraform plan
Refreshing Terraform state in-memory prior to plan...
The refreshed state will be used to calculate this plan, but will not be
persisted to local or remote state storage.

aws_db_parameter_group.x: Refreshing state... [id=x]

------------------------------------------------------------------------

An execution plan has been generated and is shown below.
Resource actions are indicated with the following symbols:
  ~ update in-place

Terraform will perform the following actions:

  # aws_db_parameter_group.x will be updated in-place
  ~ resource "aws_db_parameter_group" "x" {
        arn         = "arn:aws:rds:us-west-2:<account_id>:pg:x"
        description = "Managed by Terraform"
        family      = "mysql5.7"
        id          = "x"
        name        = "x"
        tags        = {}

      + parameter {
          + apply_method = "pending-reboot"
          + name         = "innodb_buffer_pool_size"
          + value        = "{dBInstanceClassMemory*1/2}"
        }
      - parameter {
          - apply_method = "pending-reboot" -> null
          - name         = "innodb_buffer_pool_size" -> null
          - value        = "{dbinstanceclassmemory*8/19}" -> null
        }
    }

Plan: 0 to add, 1 to change, 0 to destroy.

We can confirm this with the AWS console:

The *8/19 value is the actual value, but the state file now thinks otherwise.

I finally wrote the minimalist reproduction case for this in the course of testing out whether #12973 was still a live concern for us.

Interestingly, if you attempt to create an RDS parameter group that does not yet exist, and you use a capitalization of the variable different than what AWS wants, it also get persisted to the state file, but as tainted. ¯\(ツ)/¯

[gdavison]

Thanks for the issue, @handlerbot. This is related to an error in our shared Plugin SDK. hashicorp/terraform-plugin-sdk#476

[github-actions]

This functionality has been released in v3.55.0 of the Terraform AWS Provider. Please see the Terraform documentation on provider versioning or reach out if you need any assistance upgrading.

For further feature requests or bug reports with this functionality, please create a new GitHub issue following the template. Thank you!

[github-actions]

I'm going to lock this issue because it has been closed for 30 days ⏳. This helps our maintainers find and focus on the active issues.
If you have found a problem that seems similar to this, please open a new issue and complete the issue template so we can capture all the details necessary to investigate further.