AWS upgrades aurora version in the background, breaking deploy

[tdewitt]

When I set this up originally, it create the Aurora cluster with the defined version. It appears that AWS has upgraded it in the background. Since I'm calling this as a module, I can't use lifecycle to ignore the change and move. This will likely prevent any future deploys.

Untested (I might tomorrow) but engine_version supports partials, which should allow AWS to do the forced upgrades without breaking things.

 # module.serverless-static-wordpress.aws_rds_cluster.serverless_wordpress will be updated in-place
  ~ resource "aws_rds_cluster" "serverless_wordpress" {
      ~ engine_version                      = "5.7.mysql_aurora.2.08.3" -> "5.7.mysql_aurora.2.07.1"
        id                                  = "site-serverless-wordpress"
        tags                                = {}
        # (36 unchanged attributes hidden)

        # (1 unchanged block hidden)
    }

terraform-aws-serverless-static-wordpress/rds.tf

Line 47 in abc903d

engine_version = "5.7.mysql_aurora.2.07.1"