[Bug]: Snowflake_Warehouse Resource Limits Max Cluster Count

[bdumford-OM1]

Terraform CLI Version

1.5.3

Terraform Provider Version

0.93.0

Terraform Configuration

resource "snowflake_warehouse" "warehouse" {
  ...
  max_concurrency_level               = var.warehouse_max_concurrency_level
}

Category

category:resource

Object type(s)

resource:warehouse

Expected Behavior

The documentation for the snowflake_warehouse does say that the max_cluster_count parameter is ranged from 1-10. However, not every Snowflake account is limited to only 10 clusters on a given warehouse. We have a warehouse that has a max cluster count value of 20. I would expect the proper behavior to be that terraform would throw an apply error if the max cluster count exceeded what is allowed on the Snowflake account. I would not expect a plan error, but that's exactly what I'm getting. We did not have this problem in the provider version 0.92.0.

Actual Behavior

The following plan error is thrown:

│ Error: expected max_cluster_count to be in the range (1 - 10), got 20
│ 
│   with snowflake_warehouse.warehouse,
│   on warehouses.tf line 65, in resource "snowflake_warehouse" "warehouse":
│   65:   max_cluster_count                   = var.warehouse_max_cluster_count

Steps to Reproduce

Create a warehouse using the snowflake_warehouse resource using provider version 0.93.0. Set max_cluster_count to a value >10.

How much impact is this issue causing?

Medium

Logs

No response

Additional Information

Our Snowflake account allows >10 max clusters per warehouse. This differs from the Snowflake documentation, but I'm sure there are other users out there that are allowed >10 clusters on their warehouse that will be affected by this.

Would you like to implement a fix?

• Yeah, I'll take it 😎

[sfc-gh-asawicki]

Hey @bdumford-OM1. Thanks for reaching out to us.

You are right; it was once fixed (in #1919), but v0.93.0 introduced the same validation again. We will roll it back in v0.94.0 (expected next week on Wednesday/Thursday).

[bdumford-OM1]

Thanks @sfc-gh-asawicki! I was going to post that I'm not even able to work around this by adding max_cluster_count to a lifecycle -> ignore_changes block. But if it's going to be rolled back next week, I will just wait for that version.

[sfc-gh-asawicki]

Hey @bdumford-OM1, I have relaxed the validations in #2959. It will be part of the v0.94 release (tomorrow morning CEST).

[bdumford-OM1]

Hey @bdumford-OM1, I have relaxed the validations in #2959. It will be part of the v0.94 release (tomorrow morning CEST).

Thanks for taking care of this!