provider/aws: Read Elastic Beanstalk stack name

[dharrisio]

Fixes #7300 where an Elastic Beanstalk error or change outside of Terraform modifies the solution_stack_name attribute, which results in a stale value in the state file.

[stack72]

Hi @dharrisio

Just ran through the tests and they look good to me :)

Thanks for this

Paul

[stack72]

Actually, @dharrisio

On second look at the tests, we are getting a plan cycle

=== RUN   TestAccAWSBeanstalkEnv_config
--- FAIL: TestAccAWSBeanstalkEnv_config (401.23s)
    testing.go:255: Step 0 error: After applying this step, the plan was not empty:

        DIFF:

        UPDATE: aws_elastic_beanstalk_environment.tftest
          solution_stack_name: "64bit Amazon Linux running Python" => ""

[dharrisio]

@stack72 The tests fails because the solution_stack_name and template_name are mutually exclusive. When the template_name is used, the solution_stack_name is still read from the environment. When this happens, Terraform will try and correct this by setting solution_stack_name to nothing.

I was able to get around this by setting solution_stack_name to computed, but this means it can no longer use ConflictsWith on the template_name. With this change, if the Terraform document has both attributes, the Elastic Beanstalk API will return an error. I think that will be ok, but wanted to check if there was a better way to handle this.

[stack72]

Hi @dharrisio

If this is Computed, then let's make it so :) Do they both need to be Computed?

P.

[dharrisio]

I think that only the solution_stack_name is computed. This will have a value if solution_stack_name or template_name is used. template_name is optional, and should only have a value if it is explicitly used in the Terraform document.

Running a few more tests to verify, but should have a new pr for this up today.

[ghost]

I'm going to lock this issue because it has been closed for 30 days ⏳. This helps our maintainers find and focus on the active issues.

If you have found a problem that seems similar to this, please open a new issue and complete the issue template so we can capture all the details necessary to investigate further.