[Bug]: modifying Redshift Cluster (redshift-cluster-01): InvalidParameterCombination: No modifications were requested

[Aussie007]

Terraform Core Version

1.4.1

AWS Provider Version

4.52.0

Affected Resource(s)

aws_redshift_cluster

Expected Behavior

Should perform in-place update

Actual Behavior

Fails with error:

Error: modifying Redshift Cluster (dw-dev-redshift-cluster-01): InvalidParameterCombination: No modifications were requested
│ status code: 400, request id: 51c15156-5ca5-4de0-a431-xxxxxxxxxxxx

Relevant Error/Panic Output Snippet

Terraform plan output:

Terraform used the selected providers to generate the following execution plan. Resource actions are indicated with the following symbols:
  ~ update in-place

Terraform will perform the following actions:

  # aws_redshift_cluster.rs_cluster_01 will be updated in-place
  ~ resource "aws_redshift_cluster" "rs_cluster_01" {
      + default_iam_role_arn                 = "arn:aws:iam::XXXXXXXXXXXX:role/redshift"
        id                                   = "redshift-cluster-01"
        tags                                 = {
            "AppID"   = "xx"
            "AppRole" = "Redshift"
            "Env"     = "dev"
            "Name"    = "redshift-cluster-01"
            "VPC"     = "dev"
        }
        # (35 unchanged attributes hidden)

        # (1 unchanged block hidden)
    }

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




Full error:

aws_redshift_cluster.rs_cluster_01: Modifying... [id=redshift-cluster-01]
╷
│ Error: modifying Redshift Cluster (redshift-cluster-01): InvalidParameterCombination: No modifications were requested
│       status code: 400, request id: 51c15156-5ca5-4de0-a431-xxxxxxxxxxxx
│
│   with aws_redshift_cluster.rs_cluster_01,
│   on redshift-clusters.tf line 6, in resource "aws_redshift_cluster" "rs_cluster_01":
│    6: resource "aws_redshift_cluster" "rs_cluster_01" {

Terraform Configuration Files

.

Steps to Reproduce

Run terraform apply and try to apply proposed actions/changes

Debug Output

No response

Panic Output

No response

Important Factoids

No response

References

No response

Would you like to implement a fix?

None

[github-actions]

Community Note

Voting for Prioritization

• Please vote on this issue by adding a 👍 reaction to the original post to help the community and maintainers prioritize this request.
• Please see our prioritization guide for information on how we prioritize.
• 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.

Volunteering to Work on This Issue

• If you are interested in working on this issue, please leave a comment.
• If this would be your first contribution, please review the contribution guide.

[hagen-h]

From what I can see, the issue is when a piece of meta info is updated, and nothing relating cluster needs modification.

how I can replicate this issue is when I change the database_name parameter and nothing that actually changes the cluster.

if you change another parameter along with database_name then that OTHER property gets modified and the error does not occur, but another bug occurs that still the database_name remains not modified, then next time you run plan you will see that the database_name needs to be modified and then upon apply you get the error described in this issue

[rhussain1]

I'm getting this same issue when I try creating a VPC endpoint, get hit with the 'No modifications were requested' error. In the end, had to apply a minor change to my cluster and create a vpc_endpoint together for the change to get picked up.

[stephanieleevillanueva]

I'm still running into this issue when trying to update the cluster_parameter_group_name property. I've read the workaround from folks where you have to also apply a cluster property change. I've tried several combinations but can't seem to get it to work still, any tips on which property I can use to piggyback my change? Thanks.

[armen-tractatus]

I'm running into the same issue when attempting to change only database_name. If I also change cluster_identifier, then a new cluster is created without the error message. That seems like a significant change once in production, but not too bad during early development stages.

[bashokku]

I'm also facing a similar issue. Any workaround for this error?