google_container_node_pool create before destroy issue

[dyu-bot]

TL;DR

When trying to change a machine type for a nodepool, using create_before_destroy to upgrade, getting error that the default nodepool already exists.

Expected behavior

Expect a new nodepool with the new machine type to be created (with a new name), old nodepool to be cordoned, drained, and deleted

Observed behavior

│ Error: resource - ***/default-node-pool-7d4f - already exists
│
│   with module.***.google_container_node_pool.pools["default-node-pool"],

Terraform Configuration

Only setting changed was machine-type

Terraform Version

Terraform v1.1.7
+ provider registry.terraform.io/hashicorp/google v3.90.1
+ provider registry.terraform.io/hashicorp/google-beta v3.90.1

Additional information

No response

[bharathkkb]

Hi @dyu-bot Could you post a sample configuration so we can try to repro?

[ee07dazn]

I have been looking into this as well.

I believe this is because we are setting name in the google_container_node_pool instead of name_prefix, and hence, since the node pool already exists, tf throws an error about it. I believe if we switch to using name_prefix instead of name, this should get sorted.

Using beta-private-cluster as an example https://github.com/terraform-google-modules/terraform-google-kubernetes-engine/blob/master/modules/beta-private-cluster/cluster.tf#L421 <-- where name gets set instead of name_prefix

[github-actions[bot]]

This issue is stale because it has been open 60 days with no activity. Remove stale label or comment or this will be closed in 7 days