No warning about missing `rest_endpoint` parameter for ACL resources

[m-kruth]

Context

• Upgrading from the deprecated confluentcloud Terraform Provider to version 1.0.0 of the confluent Terraform Provider.
• Existing ACLs had to be reformatted from the old ConfluentCloud format to the new Confluent Provider format.

Problem

Terraform Plan does not show any issues when accidentally omitting the rest_endpoint parameter, this error is only revealed at apply time where the following error is thrown:

Error: error creating Kafka ACLs: one of provider.kafka_rest_endpoint (defaults to KAFKA_REST_ENDPOINT environment variable) or resource.rest_endpoint must be set

  with confluent_kafka_acl.non_prod_experiment_cluster_describe_acl,
  on resources.tf line 75, in resource "confluent_kafka_acl" "non_prod_experiment_cluster_describe_acl":
  75: resource "confluent_kafka_acl" "non_prod_experiment_cluster_describe_acl" {

This led to a situation where existing ACLs were destroying but failed to be recreated, resulting in small period of downtime for existing accounts.

Should the Provider warn about this missing field before allowing the user to run terraform apply?

[linouk23]

@m-kruth thanks for creating an issue!

That's unfortunate that our instructions were not clear enough and it resulted the migration process caused a downtime for existing accounts.

The gotcha is we support 2 configurations now:

# Configure the Confluent Provider
terraform {
  required_providers {
    confluent = {
      source  = "confluentinc/confluent"
      version = "1.0.0"
    }
  }
}

# Option #1 when managing multiple clusters in the same Terraform workspace
provider "confluent" {
  cloud_api_key    = var.confluent_cloud_api_key    # optionally use CONFLUENT_CLOUD_API_KEY env var
  cloud_api_secret = var.confluent_cloud_api_secret # optionally use CONFLUENT_CLOUD_API_SECRET env var
}

# Option #2 when managing a single cluster in the same Terraform workspace
# See https://github.com/confluentinc/terraform-provider-confluent/tree/master/examples/configurations/managing-single-cluster for more details
provider "confluent" {
  cloud_api_key       = var.confluent_cloud_api_key    # optionally use CONFLUENT_CLOUD_API_KEY env var
  cloud_api_secret    = var.confluent_cloud_api_secret # optionally use CONFLUENT_CLOUD_API_SECRET env var

  kafka_rest_endpoint = var.kafka_rest_endpoint        # optionally use KAFKA_REST_ENDPOINT env var 
  kafka_api_key       = var.kafka_api_key              # optionally use KAFKA_API_KEY env var
  kafka_api_secret    = var.kafka_api_secret           # optionally use KAFKA_API_SECRET env var
}
# Create the resources

so rest_endpoint is an optional argument now.

Let me know if it helps.