DocumentDB change of instance_class is reported as done, but it neither done nore planned for maintenance

[Porkepix]

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 0.12.29
Provider version 3.3.0

Affected Resource(s)

• aws_docdb_cluster_instance

Terraform Configuration Files

resource "aws_docdb_cluster_instance" "foo" {
  count              = 1
  identifier         = "foo"
  cluster_identifier = aws_docdb_cluster.foo.id
  instance_class     = "db.t3.medium"

  tags = {
……
  }
}

Debug Output

  # module.unifygroup.aws_docdb_cluster_instance.docdb[0] will be updated in-place
  ~ resource "aws_docdb_cluster_instance" "docdb" {
        arn                          = "arn:aws:rds:eu-west-1:account_id:db:staging-foo-front-0"
        auto_minor_version_upgrade   = true
        availability_zone            = "eu-west-1a"
        ca_cert_identifier           = "rds-ca-2015"
        cluster_identifier           = "staging-foo-front-cluster"
        db_subnet_group_name         = "staging-foo-front"
        dbi_resource_id              = "foobar"
        endpoint                     = "staging-foo-front-0.bar.eu-west-1.docdb.amazonaws.com"
        engine                       = "docdb"
        engine_version               = "3.6.0"
        id                           = "staging-foo-front-0"
        identifier                   = "staging-foo-front-0"
      ~ instance_class               = "db.r5.large" -> "db.t3.medium"
        port                         = 27017
        preferred_backup_window      = "03:00-05:00"
        preferred_maintenance_window = "sun:02:44-sun:03:14"
        promotion_tier               = 0
        publicly_accessible          = false
        storage_encrypted            = false
        tags                         = {
…
        }
        writer                       = true

        timeouts {}
    }

This isn't a debug output as there's not really an error, so I'm not sure anything useful would be in here. However I can get such output if needed with the required level, if it's not too verbose as I'd have to go through the whole output to sanitize it.

Expected Behavior

The instance should have been replaced from a r5.large to a t3.medium, or planified to do so in the next maintenance window. At the very least an explanation should be given about why this isn't done, rather than reporting it as done, while actually doing nothing.

Actual Behavior

Changed was reported as done, while that wasn't the case

Steps to Reproduce

1. terraform apply

[theahura]

Wondering if anyone has had a chance to look at this, we just ran into this bug in our org as well.

[ShajeeOnCloud]

just got the same issue, raised it as a problem with AWS.
However, even if it is a problem on AWS side, terraform should raise an error or warning that the update did not went through. On-the contrary it says updated successfully while in reality the updgrade/downgrade of the instance did not happen.

[martimconstantino]

The same thing happens to me, not sure if I should wait at least a day or try to change manually (beating the purpose of Terraform)

[Jake7D]

Just had the same thing happen to us. We did have some pending maintenance which we wondered whether it was the scheduled update, but it was just OS updates. We've had to change it manually this time around

[ShpilerX]

Would be great to address this since also had the same experience - a successful tf apply and no change on docdb side.

[fire2]

Same thing happening here.

[ianuragsingh]

Facing same issue , Instance type is not upgrading.