Skip to content

Latest commit

 

History

History
132 lines (92 loc) · 3.9 KB

UPGRADE-6.0.md

File metadata and controls

132 lines (92 loc) · 3.9 KB

Upgrade from v5.x to v6.x

Please consult the examples directory for reference example configurations. If you find a bug, please open an issue with supporting configuration to reproduce.

List of backwards incompatible changes

  • Minimum supported version of Terraform AWS provider updated to v5.45 to support latest resources
  • Minimum supported version of Terraform raised to v1.3
  • logging block within the aws_redshift_cluster resource has been replaced with a standalone resource. After upgrade, a new resource for logging will be created.
  • snapshot_copy block within the aws_redshift_cluster resource has been replaced with a standalone resource. After upgrade, to prevent errors due to already existing snapshot_copy configurations, import of the new resource is required.

Additional changes

Added

  • aws_redshift_logging has been added to replace logging block in aws_redshift_cluster
  • aws_redshift_snapshot_copy has been added to replace snapshot_copy block in aws_redshift_cluster

Modified

  • None

Removed

  • logging block in aws_redshift_cluster
  • snapshot_copy block in aws_redshift_cluster

Variable and output changes

  1. Removed variables:
  • Cluster
    • var.logging.enable has been removed
  1. Renamed variables:
  • None
  1. Added variables:
  • Snapshot Copy
    • var.snapshot_copy.manual_snapshot_retention_period
  1. Removed outputs:
  • None
  1. Renamed outputs:
  • None
  1. Added outputs:
  • None

Upgrade Migration

Before v5.x Example

module "redshift" {
  source  = "terraform-aws-modules/redshift/aws"
  version = "~> 5.0"

  snapshot_copy = {
    destination_region = "us-east-1"
    grant_name         = aws_redshift_snapshot_copy_grant.useast1.snapshot_copy_grant_name
  }

  logging = {
    enable        = true
    bucket_name   = module.s3_logs.s3_bucket_id
    s3_key_prefix = local.s3_prefix
  }
}

After v6.x Example

module "redshift" {
  source  = "terraform-aws-modules/redshift/aws"
  version = "~> 6.0"

  snapshot_copy = {
    destination_region = "us-east-1"
    grant_name         = aws_redshift_snapshot_copy_grant.useast1.snapshot_copy_grant_name
  }

  logging = {
    bucket_name   = module.s3_logs.s3_bucket_id
    s3_key_prefix = local.s3_prefix
  }
}

Diff of Before vs After

  # module.redshift.aws_redshift_logging.this[0] will be created
  + resource "aws_redshift_logging" "this" {
      + bucket_name        = "ex-complete20240414012816938100000003"
      + cluster_identifier = "ex-complete"
      + id                 = (known after apply)
      + s3_key_prefix      = "redshift/ex-complete/"
    }

  # module.redshift.aws_redshift_snapshot_copy.this[0] will be created
  + resource "aws_redshift_snapshot_copy" "this" {
      + cluster_identifier               = "ex-complete"
      + destination_region               = "us-east-1"
      + id                               = (known after apply)
      + manual_snapshot_retention_period = (known after apply)
      + retention_period                 = (known after apply)
      + snapshot_copy_grant_name         = "ex-complete-us-east-1"
    }

The aws_redshift_logging can be applied or imported. If setting the log_destination_type, an apply following an import will be required to clear the remaining diff. The aws_redshift_snapshot_copy resource requires importing if an existing snapshot_copy configuration exists.

State Move Commands

None required

State Import Commands

During the migration to v6.x of this module, logging and snapshot_copy resources will be created by this module if those settings are configured. In order to guarantee the best experience and prevent data loss, you will need to import them into terraform state using commands like these:

terraform import 'module.redshift.aws_redshift_logging.this[0]' <cluster-id>
terraform import 'module.redshift.aws_redshift_snapshot_copy.this[0]' <cluster-id>