terraform-provider-google: Upgrade to 4.34.0 breaks on `node_pool_auto_config` block with `google-beta` provider

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 me too comments, 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.
If an issue is assigned to the modular-magician user, it is either in the process of being autogenerated, or is planned to be autogenerated soon. If an issue is assigned to a user, that user is claiming responsibility for the issue. If an issue is assigned to hashibot, a community member has claimed the issue already.

Terraform Version

Terraform v1.2.8
on linux_amd64
+ provider registry.terraform.io/hashicorp/external v2.2.2
+ provider registry.terraform.io/hashicorp/google v4.34.0
+ provider registry.terraform.io/hashicorp/google-beta v4.34.0
+ provider registry.terraform.io/hashicorp/kubernetes v2.13.1
+ provider registry.terraform.io/hashicorp/null v3.1.1
+ provider registry.terraform.io/hashicorp/random v3.3.2
+ provider registry.terraform.io/hashicorp/time v0.8.0

Affected Resource(s)

google_container_cluster

Terraform Configuration Files

We’re using a module within a module within a module to create the cluster (terraform-google-modules/kubernetes-engine/google//modules/safer-cluster-update-variant) - I can try to produce a minimal example with just the resource if necessary, but I’m unsure whether it’s relevant, maybe this is already enough.

Debug Output

Panic Output

Expected Behavior

Plan shows that changes outside of Terraform were made:

  # module.project.module.gke.module.gke.google_container_cluster.primary has changed
  ~ resource "google_container_cluster" "primary" {
        id                          = "projects/***/locations/europe-west4/clusters/***"
        name                        = "***"
        # (28 unchanged attributes hidden)

      + node_pool_auto_config {
        }

        # (22 unchanged blocks hidden)
    }

With this change to be rolled back:

  # module.project.module.gke.module.gke.google_container_cluster.primary will be updated in-place
  ~ resource "google_container_cluster" "primary" {
        id                          = "projects/***/locations/europe-west4/clusters/***"
        name                        = "***"
        # (28 unchanged attributes hidden)

      - node_pool_auto_config {
        }

        # (22 unchanged blocks hidden)
    }

I would expect this to either succeed, or the change outside of Terraform to not happen in the first place.

Actual Behavior

Apply fails with error

Error: Provider produced inconsistent final plan

When expanding the plan for
module.project.module.gke.module.gke.google_container_cluster.primary to
include new values learned so far during apply, provider
"registry.terraform.io/hashicorp/google-beta" produced an invalid new value
for .node_pool_auto_config: block count changed from 0 to 1.

This is a bug in the provider, which should be reported in the provider's own
issue tracker.

Steps to Reproduce

terraform plan
terraform apply

Important Factoids

References

I believe this was introduced by https://github.com/GoogleCloudPlatform/magic-modules/pull/6133

About this issue

Original URL
State: closed
Created 2 years ago
Reactions: 19
Comments: 19

Commits related to this issue

downgrade to google prodiver 4.33.0 due to https://github.com/hashicorp/terraform-provider-google/issues/12422 — committed to TakeoffTech/netbox-gcp-deployment by scheuk 2 years ago

Most upvoted comments

4.38.0 has been released, and the issue still persists on both node_pool_defaults and node_pool_auto_config. This has not been resolved

Apply output:


│ Error: Provider produced inconsistent final plan
│ 
│ When expanding the plan for module.cluster.google_container_cluster.cluster to include new values learned so far during apply, provider
│ "registry.terraform.io/hashicorp/google-beta" produced an invalid new value for .node_pool_defaults: block count changed from 0 to 1.
│ 
│ This is a bug in the provider, which should be reported in the provider's own issue tracker.
╵
╷
│ Error: Provider produced inconsistent final plan
│ 
│ When expanding the plan for module.cluster.google_container_cluster.cluster to include new values learned so far during apply, provider
│ "registry.terraform.io/hashicorp/google-beta" produced an invalid new value for .node_pool_auto_config: block count changed from 0 to 1.
│ 
│ This is a bug in the provider, which should be reported in the provider's own issue tracker.

thameezb on Sep 30, 2022

Thanks for the feedback! I was able to reproduce this error locally and I’m working on a fix now.

shuyama1 on Sep 13, 2022

Glad it’s solved! Thanks for the update!

shuyama1 on Oct 6, 2022

Agree with @thameezb. I have the same issue on the latest release i-e 4.38.0 on both node_pool_defaults and node_pool_auto_config. The block count changed without having it specified.

dani2819 on Sep 30, 2022

The PR has not been merged and released yet. You could try to pin the provider version to 4.33.0 to prevent issues in the short term until the fix is released. The PR is targeting the next release.

shuyama1 on Sep 20, 2022

@shuyama1 can you take a look at this?

edwardmedia on Sep 12, 2022

The provider needs to be fixed, but as a work around we added this to our GKE module:

variable "node_pool_network_tags" {
  description = "Network tags to use in autopilot or auto provisioned node pools"
  type = list(string)
  default = []
}

  dynamic "node_pool_auto_config" {
    for_each = var.cluster_autoscaling.enabled || var.enable_autopilot ? [""]: []
    content {
      dynamic "network_tags" {
        for_each = length(var.node_pool_network_tags) == 0 ? [] : [var.node_pool_network_tags]
        iterator = taglist
        content {
          tags = taglist
        }
      }
    }
  }

philip-harvey on Aug 30, 2022