Unable to restore Snapshot into new Redshift cluster with different configuration

[ghost]

This issue was originally opened by @hdryx as hashicorp/terraform#23720. It was migrated here as a result of the provider split. The original body of the issue is below.

---

Hi,

This is maybe a bug in Terraform when using Redshift. I have created a Snapshot of a Redshift cluster having this config : 8 nodes of dc2.large

Now i'm trying to restore the snapshot into a new cluster with a different config : 2 nodes of dc2.8xlarge but i'm getting this error :

Error: InvalidParameterValue: Snapshot cluster-redshift-dev-snapshot-manual-18122019 does not fit into cluster of node type dc2.8xlarge with 8 node(s) due to technical limitations. Try to restore to a cluster with a different node configuration.
status code: 400, request id: f7b03b91-223e-11ea-b863-2593ae69f79f

Terraform Version

0.12.18

Terraform Configuration Files

This is code of the module creating the cluster : 

resource "aws_redshift_cluster" "redshift" {
  cluster_identifier                  = "${var.cluster_name}"
  database_name                       = var.db_name
  port                                = var.db_port
  master_username                     = var.db_master_username
  master_password                     = random_string.db_password.result
  node_type                           = var.node_type
  cluster_type                        = var.cluster_type
  number_of_nodes                     = var.number_of_nodes
  automated_snapshot_retention_period = var.automated_snapshot_retention_period
  skip_final_snapshot                 = true
  encrypted                           = true
  kms_key_id                          = var.kms_key_id

  publicly_accessible       = var.publicly_accessible
  vpc_security_group_ids    = var.security_groups
  cluster_subnet_group_name = aws_redshift_subnet_group.subnet.id
  iam_roles                 = var.iam_roles

  logging {
    enable        = true
    bucket_name   = var.logs_bucket_name
    s3_key_prefix = "redshift/${var.cluster_name}"
  }

  snapshot_identifier = var.snapshot_identifier != "" ? var.snapshot_identifier : null
  owner_account       = var.owner_account != "" ? var.owner_account : null

  lifecycle { ignore_changes = [tags.DateTimeTag] }
  tags = "${merge(var.resource_tagging, map("Name", "${var.cluster_name}"))}"
}


And this is how i'm calling the module : 

module "redshift_test_restore" {
  source = "../00_modules/redshift"

  cluster_name       = var.redshift_cluster_name_test_restore
  db_name            = var.db_name
  db_port            = var.redshift_cluster_port
  db_master_username = var.db_master_username
  node_type          = "dc2.8xlarge"
  cluster_type       = var.cluster_type
  number_of_nodes    = "2"
  kms_key_id         = ""

  publicly_accessible = "false"
  subnet_ids          = [data.aws_subnet.private_1.id, data.aws_subnet.private_2.id]
  security_groups     = [data.aws_security_group.sg_redshift_cluster.id]
  iam_roles           = [aws_iam_role.redshift_role.arn]

  logs_bucket_name    = data.aws_s3_bucket.logs.id

  snapshot_identifier = "cluster-redshift-dev-snapshot-manual-18122019"
  owner_account       = var.owner_account

  resource_tagging = var.resource_tagging
  environment      = var.environment
  project          = var.project
}

Debug Output

Crash Output

Error: InvalidParameterValue: Snapshot cluster-redshift-dev-snapshot-manual-18122019 does not fit into cluster of node type dc2.8xlarge with 8 node(s) due to technical limitations. Try to restore to a cluster with a different node configuration.
status code: 400, request id: f7b03b91-223e-11ea-b863-2593ae69f79f

Actual Behavior

I was excepecting the creation of a new cluster with the Snapshot restored;

Steps to Reproduce

1. terraform init
2. terraform apply

It looks like that Terraform is trying the apply the value of number_of_nodes stored in the Snapshot instead the value passed to the module.

[hdryx]

I confirm that this is probably a Terraform bug as i tried today to create manually in the AWS console a new cluster based on the Snapshot. The Redshift was created successfully as expected.

[hdryx]

Any news about this issue ?

[jdmchone]

Any update on this? I'm not getting an error message, but when i choose to restore from a snapshot, it is using the snapshot cluster's number_of_instances, instead of the one i set in the terraform config.

[marksergeant]

Confirming, I'm seeing the same issues here as well.

In addition the complaint regarding 'due to technical limitations' is a red herring as it is not an AWS limitation, only a terraform limitation.

[justinretzolk]

Hey y'all 👋 Thank you for taking the time to file this, and for the additional discussion around it. Given that there's been a number of Terraform and AWS provider releases since the last update, can anyone confirm whether you're still experiencing this behavior?

[acavnar-ibm]

Hey y'all 👋 Thank you for taking the time to file this, and for the additional discussion around it. Given that there's been a number of Terraform and AWS provider releases since the last update, can anyone confirm whether you're still experiencing this behavior?

Hi Justin,

I can confirm this is still happening with the latest version of the provider.

[github-actions]

This functionality has been released in v4.9.0 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.