[Bug]: updating the default_tags causes aws_subnet to be recreated

[vinlp]

Terraform Core Version

1.4.6

AWS Provider Version

5.7.0

Affected Resource(s)

aws_subnet

Expected Behavior

We have default_tags set at the provider level.

Each time the default_tags are updated, we want those tags to be updated on all the resources and especially on the aws_subnet ones. Our aws_subnet resources have their own tags: only Name

Actual Behavior

Each time we update those default_tags, we are having issues with aws_subnet resources trying to be replaced/re-created instead of being updated with the new default_tags + their current tag (Name).

plan output

      ~ tags_all                                       = {
          - "DeployedBy" = "Terraform"
          - "Env"        = "dev"
          - "Name"       = "db-private-subnet"
          - "Project"    = "myproject"
          - "Zone"       = "ew1"
        } -> (known after apply)

this ends up with errors during the apply.

Error: creating EC2 Subnet: InvalidSubnet.Conflict: The CIDR '10.X.XXX.0/22' conflicts with another subnet
│       status code: 400, request id: 24337a43-d090-421b-9831-a6fe3a8252f3

Relevant Error/Panic Output Snippet

Error: creating EC2 Subnet: InvalidSubnet.Conflict: The CIDR '10.X.XXX.0/22' conflicts with another subnet
│       status code: 400, request id: 24337a43-d090-421b-9831-a6fe3a8252f3

Terraform Configuration Files

locals {
  default_tags = {
    Env        = "dev"
    Project    = "my-project"
    DeployedBy = "Terraform"
    Owner      = "ProjectOwner"
    BU         = "BusinessUnit"
  }
}

terraform {
  required_providers {
    aws = {
      source  = "hashicorp/aws"
      version = "5.7.0"
    }
  }

  backend "s3" {
    encrypt = true
  }
}

provider "aws" {
  region              = "eu-west-1"
  allowed_account_ids = [local.config_file_json.infra.aws_account_id]

  default_tags {
    tags = merge(local.default_tags, {
      Zone = "ew1"
    })
  }
}

data "aws_availability_zones" "available" {
  state = "available"
}

resource "aws_vpc" "my_vpc" {
  cidr_block           = "10.2.0.0/16"
  enable_dns_hostnames = true

  tags = {
    Name = "vpc-${var.resource_suffix}"
  }
}

resource "aws_subnet" "my_public_subnets" {
  count = length(["10.2.0.0/22", "10.2.4.0/22", "10.2.8.0/22"])

  vpc_id                  = aws_vpc.my_vpc.id
  cidr_block              = ["10.2.0.0/22", "10.2.4.0/22", "10.2.8.0/22"][count.index]
  availability_zone       = data.aws_availability_zones.available.names[count.index]
  map_public_ip_on_launch = false

  tags = {
    Name = "public-subnet-${data.aws_availability_zones.available.names[count.index]}-${var.resource_suffix}"
    Tier = "Public"
  }
}

Steps to Reproduce

• define default_tags at provider level
• deploy vpc and subnets and set a Name tag on subnets
• tf plan / apply
• update default_tags
• it break on subnets trying to be recreated

Debug Output

No response

Panic Output

No response

Important Factoids

No response

References

No response

Would you like to implement a fix?

None

[github-actions[bot]]

Community Note

Voting for Prioritization

• Please vote on this issue by adding a 👍 reaction to the original post to help the community and maintainers prioritize this request.
• Please see our prioritization guide for information on how we prioritize.
• Please do not leave "+1" or other comments that do not add relevant new information or questions, they generate extra noise for issue followers 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.

[justinretzolk]

Hey @vinlp 👋 Thank you for taking the time to raise this! So that we have the information necessary to look into this, can you supply debug logs (redacted as needed)?

[kristian-lesko]

Hello, I'm seeing a similar issue. It seems at least the following resources are affected:

• aws_cloudwatch_metric_alarm
• aws_default_network_acl
• aws_default_security_group
• aws_secretsmanager_secret_version

Note: in our case, aws_secretsmanager_secret_version depends on a aws_kms_secrets data source, which is reported as re-read during a plan when default_tags are added/changed.

[eml-39502]

Had the issue with AWS provider 5.41.0 and 5.43.0 as well.