Unable to apply empty capacity_reservation_specification.capacity_reservation_target within aws_autoscaling_group

[dankelly-synth]

Terraform and AWS Provider Version

Terraform v1.12.2
on darwin_arm64
+ provider registry.terraform.io/hashicorp/aws v5.100.0

Affected Resource(s) or Data Source(s)

aws_autoscaling_group

Expected Behavior

Empty capacity_reservation_target block is added to capacity_reservation_specification as per plan:

Terraform used the selected providers to generate the following execution
plan. Resource actions are indicated with the following symbols:
  ~ update in-place
Terraform will perform the following actions:
  # module.instance.aws_autoscaling_group.asg will be updated in-place
  ~ resource "aws_autoscaling_group" "asg" {
        id                               = "my-asg"
        name                             = "my-asg"
        # (31 unchanged attributes hidden)
      ~ capacity_reservation_specification {
            # (1 unchanged attribute hidden)
          + capacity_reservation_target {}
        }
        # (14 unchanged blocks hidden)
    }
Plan: 0 to add, 1 to change, 0 to destroy.

Actual Behavior

Unable to apply - provider crashes. See output.

Relevant Error/Panic Output

  ╷
  │ Error: Request cancelled
  │ 
  │   with module.instance.aws_autoscaling_group.asg,
  │   on instance/main.tf line 233, in resource "aws_autoscaling_group" "asg":
  │  233: resource "aws_autoscaling_group" "asg" {
  │ 
  │ The plugin.(*GRPCProvider).ApplyResourceChange request was cancelled.
  ╵
  
  Stack trace from the terraform-provider-aws_v5.100.0_x5 plugin:
  
  panic: interface conversion: interface {} is nil, not map[string]interface {}
  
  goroutine 31 [running]:
  github.com/hashicorp/terraform-provider-aws/internal/service/autoscaling.expandCapacityReservationSpecification(0x1400355c9f0)
        github.com/hashicorp/terraform-provider-aws/internal/service/autoscaling/group.go:3484 +0x17c
  github.com/hashicorp/terraform-provider-aws/internal/service/autoscaling.resourceGroupUpdate({0x11b3dac28, 0x1400355c630}, 0x1400287b380, {0x11b04e5c0, 0x140017f6d00})
        github.com/hashicorp/terraform-provider-aws/internal/service/autoscaling/group.go:1469 +0x5f8
  github.com/hashicorp/terraform-provider-aws/internal/provider.wrapResource.(*wrappedResource).update.interceptedHandler[...].func5(0x1400287b380, {0x11b04e5c0, 0x140017f6d00})
        github.com/hashicorp/terraform-provider-aws/internal/provider/intercept.go:114 +0x2d8
  github.com/hashicorp/terraform-plugin-sdk/v2/helper/schema.(*Resource).update(0x11b3dac28?, {0x11b3dac28?, 0x1400287ca20?}, 0xd?, {0x11b04e5c0?, 0x140017f6d00?})
        github.com/hashicorp/terraform-plugin-sdk/v2@v2.36.1/helper/schema/resource.go:867 +0x64
  github.com/hashicorp/terraform-plugin-sdk/v2/helper/schema.(*Resource).Apply(0x14000406e00, {0x11b3dac28, 0x1400287ca20}, 0x14003180820, 0x1400287ae00, {0x11b04e5c0, 0x140017f6d00})
        github.com/hashicorp/terraform-plugin-sdk/v2@v2.36.1/helper/schema/resource.go:979 +0x660
  github.com/hashicorp/terraform-plugin-sdk/v2/helper/schema.(*GRPCProviderServer).ApplyResourceChange(0x14003502eb8, {0x11b3dac28?, 0x1400287c960?}, 0x14002876230)
        github.com/hashicorp/terraform-plugin-sdk/v2@v2.36.1/helper/schema/grpc_provider.go:1188 +0xaa4
  github.com/hashicorp/terraform-plugin-mux/tf5muxserver.(*muxServer).ApplyResourceChange(0x140033e2c80, {0x11b3dac28?, 0x1400287c690?}, 0x14002876230)
        github.com/hashicorp/terraform-plugin-mux@v0.18.0/tf5muxserver/mux_server_ApplyResourceChange.go:36 +0x184
  github.com/hashicorp/terraform-plugin-go/tfprotov5/tf5server.(*server).ApplyResourceChange(0x14001d3d900, {0x11b3dac28?, 0x1400284db90?}, 0x14000e5c620)
        github.com/hashicorp/terraform-plugin-go@v0.26.0/tfprotov5/tf5server/server.go:866 +0x2a8
  github.com/hashicorp/terraform-plugin-go/tfprotov5/internal/tfplugin5._Provider_ApplyResourceChange_Handler({0x11ae92560, 0x14001d3d900}, {0x11b3dac28, 0x1400284db90}, 0x14004e09500, 0x0)
        github.com/hashicorp/terraform-plugin-go@v0.26.0/tfprotov5/internal/tfplugin5/tfplugin5_grpc.pb.go:611 +0x1c0
  google.golang.org/grpc.(*Server).processUnaryRPC(0x14001165600, {0x11b3dac28, 0x1400315d170}, 0x14002f070e0, 0x14001bc0330, 0x127605df8, 0x0)
        google.golang.org/grpc@v1.72.0/server.go:1405 +0xca8
  google.golang.org/grpc.(*Server).handleStream(0x14001165600, {0x11b4131d8, 0x1400308a820}, 0x14002f070e0)
        google.golang.org/grpc@v1.72.0/server.go:1815 +0x910
  google.golang.org/grpc.(*Server).serveStreams.func2.1()
        google.golang.org/grpc@v1.72.0/server.go:1035 +0x84
  created by google.golang.org/grpc.(*Server).serveStreams.func2 in goroutine 53
        google.golang.org/grpc@v1.72.0/server.go:1046 +0x13c
  
  Error: The terraform-provider-aws_v5.100.0_x5 plugin crashed!
  
  This is always indicative of a bug within the plugin. It would be immensely
  helpful if you could report the crash with the plugin's maintainers so that it
  can be fixed. The output above should help diagnose the issue.
  
  
  exit status 1

Sample Terraform Configuration

dynamic `capacity_reservation_specification` block within `aws_autoscaling_group`, where `capacity_reservation_target` defaults to`{}` if not specified.

Click to expand configuration

  dynamic "capacity_reservation_specification" {
    for_each = length(var.instance.capacity_reservation_specification) > 0 ? [1] : []

    content {
      capacity_reservation_preference = try(var.instance.capacity_reservation_specification.capacity_reservation_preference, null)

      capacity_reservation_target {
        capacity_reservation_ids                 = try(var.instance.capacity_reservation_specification.capacity_reservation_ids, null)
        capacity_reservation_resource_group_arns = try(var.instance.capacity_reservation_specification.capacity_reservation_resource_group_arns, null)
      }
    }
  }

Steps to Reproduce

1. Create aws_autoscaling_group resource without specifying capacity_reservation_specification.
2. Add dynamic capacity_reservation_specification block to return + capacity_reservation_target {} in plan.
3. Apply changes

Debug Logging

Click to expand log output

GenAI / LLM Assisted Development

n/a

Important Facts and References

No response

Would you like to implement a fix?

No

[github-actions]

Community Guidelines

This comment is added to every new Issue to provide quick reference to how the Terraform AWS Provider is maintained. Please review the information below, and thank you for contributing to the community that keeps the provider thriving! 🚀

Voting for Prioritization

• Please vote on this Issue by adding a 👍 reaction to the original post to help the community and maintainers prioritize it.
• Please see our prioritization guide for additional information on how the maintainers handle prioritization.
• Please do not leave +1 or other comments that do not add relevant new information or questions; they generate extra noise for others following the Issue and do not help prioritize the request.

Volunteering to Work on This Issue

• If you are interested in working on this issue, please leave a comment.
• If this would be your first contribution, please review the contribution guide.
• For new resources and data sources, use skaff to generate scaffolding with comments detailing common expectations.

[justinretzolk]

Hey @dankelly-synth 👋 Thank you for taking the time to raise this! While the panic is something that should be addressed, providing the capacity_reservation_target block without either of the arguments within the block wouldn't be valid in the upstream API call, so you'll want to adjust your configuration such that the capacity_reservation_target block isn't defined when it's not needed. This can be done using nested dynamic blocks.

While I don't know exactly how var.instance is shaped, I created a configuration that worked in my testing and seemed to match roughly to what your configuration looks like based on the information I have.

variable "instance" {
  default = {
    capacity_reservation_specification = {
      capacity_reservation_preference = "default"
      capacity_reservation_ids        = ["id"]
    }
  }
}

resource "aws_autoscaling_group" "foo" {
  # ...other config...

  dynamic "capacity_reservation_specification" {
    for_each = length(var.instance.capacity_reservation_specification) > 0 ? [1] : []

    content {
      capacity_reservation_preference = try(var.instance.capacity_reservation_specification.capacity_reservation_preference, null)

      dynamic "capacity_reservation_target" {
        for_each = try(var.instance.capacity_reservation_specification.capacity_reservation_ids, var.instance.capacity_reservation_specification.capacity_reservation_resource_group_arns, null) != null ? [1] : []
        content {
          capacity_reservation_ids                 = try(var.instance.capacity_reservation_specification.capacity_reservation_ids, null)
          capacity_reservation_resource_group_arns = try(var.instance.capacity_reservation_specification.capacity_reservation_resource_group_arns, null)
        }
      }
    }
  }
}

A slightly different approach with a slightly different shape of var.instance.

variable "other_instance" {
  default = {
    capacity_reservation_specification = {
      capacity_reservation_preference = "default"
      capacity_reservation_target = {
        capacity_reservation_ids = ["id"]
      }
    }
  }

  validation {
    condition     = can(var.other_instance.capacity_reservation_specification.capacity_reservation_target) ? length(var.other_instance.capacity_reservation_specification.capacity_reservation_target) == 1 : true
    error_message = "If capacity_reservation_specification.capacity_reservation_target is provided, it must have exactly one key/value pair as its value"
  }
}

resource "aws_autoscaling_group" "foo" {
  # ...other config...

  dynamic "capacity_reservation_specification" {
    for_each = try([var.other_instance.capacity_reservation_specification], [])

    content {
      capacity_reservation_preference = try(capacity_reservation_specification.value.capacity_reservation_preference, null)

      dynamic "capacity_reservation_target" {
        for_each = try([capacity_reservation_specification.value.capacity_reservation_target], [])
        content {
          capacity_reservation_ids                 = try(capacity_reservation_target.value.capacity_reservation_ids, null)
          capacity_reservation_resource_group_arns = try(capacity_reservation_target.value.capacity_reservation_resource_group_arns, null)
        }
      }
    }
  }
}

[dankelly-synth]

@justinretzolk - thanks for these pointers. I'll give them a go and report back. Cheers 🙂

[dankelly-synth]

@justinretzolk - Adding the additional dynamic block worked. Thanks again!

Happy to keep this issue open if you still want to address the panic. Cheers!