resource/aws_ecs_task_definition: Support task scoped docker volume configurations

[camdenfullmer]

Recently support was added for Docker volume configurations for ECS task definitions: #5727

Unfortunately, if you want a task scoped docker volume configuration this implementation doesn't work resulting in the following error:

Error: Error applying plan:

1 error(s) occurred:

* aws_ecs_task_definition.service: 1 error(s) occurred:

* aws_ecs_task_definition.service: ClientException: When 'scope' parameter is 'task', 'autoprovision' must either not be specified or set to 'True'.
	status code: 400, request id: <REQUEST_ID>

Terraform does not automatically rollback in the face of errors.
Instead, your Terraform state file has been partially updated with
any resources that successfully completed. Please address the error
above and apply again to incrementally change your infrastructure.

With the following configuraiton:

volume {
    name      = "scratch"
    docker_volume_configuration {
        scope         = "task"
        driver        = "local"
    }
}

It seems that not specifying a value results in autoprovision being set to false. Setting autoprovison to true results in an error when the ECS agent tries to run the task.

AWS documentation states that autoprovision is only used for shared scope: https://docs.aws.amazon.com/AmazonECS/latest/developerguide/docker-volumes.html

Changes proposed in this pull request:

• Set Autoprovision to nil when using a task scoped docker volume and autoprovision is set to false.

Questions:

• I tried removing the Default value of false, but false was still picked up as the value. Is there a way to check for no value and set it to nil instead?

[FernandoMiguel]

also being affected by this. couldnt find a workaround for false

[camdenfullmer]

@bflad Is there anything else I need to add to this PR to get it merged into a release?

[bflad]

@camdenfullmer are you able to implement a separate acceptance test instead of modifying the existing one? Duplicating (with minor adjustments) the existing test and test configuration is perfectly fine. We should perform regression testing on both scenarios. 👍

[camdenfullmer]

Yep, I can do that 👍

[camdenfullmer]

@bflad Added a new test instead of modifying the existing one.

[bflad]

Looks great, @camdenfullmer! 🚀

--- PASS: TestAccAWSEcsTaskDefinition_withScratchVolume (4.72s)
--- PASS: TestAccAWSEcsTaskDefinition_withTaskScopedDockerVolume (5.00s)
--- PASS: TestAccAWSEcsTaskDefinition_arrays (5.09s)
--- PASS: TestAccAWSEcsTaskDefinition_withDockerVolume (5.09s)
--- PASS: TestAccAWSEcsTaskDefinition_constraint (5.12s)
--- PASS: TestAccAWSEcsTaskDefinition_withDockerVolumeMinimalConfig (5.12s)
--- PASS: TestAccAWSEcsTaskDefinition_ExecutionRole (12.62s)
--- PASS: TestAccAWSEcsTaskDefinition_basic (13.06s)
--- PASS: TestAccAWSEcsTaskDefinition_changeVolumesForcesNewResource (13.16s)
--- PASS: TestAccAWSEcsTaskDefinition_Inactive (13.76s)
--- PASS: TestAccAWSEcsTaskDefinition_Fargate (17.93s)
--- PASS: TestAccAWSEcsTaskDefinition_withNetworkMode (20.57s)
--- PASS: TestAccAWSEcsTaskDefinition_withTaskRoleArn (21.22s)
--- PASS: TestAccAWSEcsTaskDefinition_withEcsService (63.25s)

[ghost]

This has been released in version 1.39.0 of the provider. Please see the Terraform documentation on provider versioning or reach out if you need any assistance upgrading. As an example:

provider "aws" {
	version = "~> 1.39.0"
}
# ... other configuration ...

[FernandoMiguel]

Thank you both

[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 feel this issue should be reopened, we encourage creating a new issue linking back to this one for added context. Thanks!