Error/Bug: aws_s3_bucket_object inconsistent final plan with version_id

[wernerb]

Community Note

• Please vote on this issue by adding a 👍 reaction to the original issue to help the community and maintainers prioritize this request
• 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
• If you are interested in working on this issue or have submitted a pull request, please leave a comment

Terraform CLI and Terraform AWS Provider Version

Terraform v0.13.1
+ provider registry.terraform.io/-/aws v3.4.0
+ provider registry.terraform.io/hashicorp/aws v3.4.0

Affected Resource(s)

• aws_s3_bucket_object

Terraform Configuration Files

provider "aws" {
  version                     = ">= 3.0.0"
}

resource "aws_s3_bucket" "examplebucket" {
  bucket = "sampletestexamplebuckettftest"
  acl    = "private"

  versioning {
    enabled = true
  }
}

variable "content" {
  type = string
}

resource "aws_s3_bucket_object" "firstfile" {
  key        = "someobject"
  bucket     = aws_s3_bucket.examplebucket.id
  content    = var.content
}

resource "aws_s3_bucket_object" "secondfile" {
  key        = "anotherobject"
  bucket     = aws_s3_bucket.examplebucket.id
  content    = "${md5(var.content)} - ${aws_s3_bucket_object.firstfile.version_id}"
}

Debug Output

https://gist.github.com/wernerb/ec3872b69b1093a136fc9dc41880fb28

Panic Output

Expected Behavior

Terraform knows that a new version_id is there and uses that correctly in secondfile

I have submitted a pull request that completely closes this issue.

Actual Behavior

Error thrown during apply:

$ terraform apply -var content=bla2
aws_s3_bucket.examplebucket: Refreshing state... [id=sampletestexamplebuckettftest]
aws_s3_bucket_object.firstfile: Refreshing state... [id=someobject]
aws_s3_bucket_object.secondfile: Refreshing state... [id=anotherobject]

An execution plan has been generated and is shown below.
Resource actions are indicated with the following symbols:
  ~ update in-place

Terraform will perform the following actions:

  # aws_s3_bucket_object.firstfile will be updated in-place
  ~ resource "aws_s3_bucket_object" "firstfile" {
        acl           = "private"
        bucket        = "sampletestexamplebuckettftest"
      ~ content       = "bla" -> "bla2"
        content_type  = "binary/octet-stream"
        etag          = "128ecf542a35ac5270a87dc740918404"
        force_destroy = false
        id            = "someobject"
        key           = "someobject"
        metadata      = {}
        storage_class = "STANDARD"
        tags          = {}
        version_id    = "5_Fxmip9w4DJgZV2or3_SAAn8Rr2QyA5"
    }

  # aws_s3_bucket_object.secondfile will be updated in-place
  ~ resource "aws_s3_bucket_object" "secondfile" {
        acl           = "private"
        bucket        = "sampletestexamplebuckettftest"
      ~ content       = "128ecf542a35ac5270a87dc740918404 - 5_Fxmip9w4DJgZV2or3_SAAn8Rr2QyA5" -> "a1288b1c7e2257a90bad9bdfb7690fbb - 5_Fxmip9w4DJgZV2or3_SAAn8Rr2QyA5"
        content_type  = "binary/octet-stream"
        etag          = "6be52a6a547f72e84fa7cdd36e01c56d"
        force_destroy = false
        id            = "anotherobject"
        key           = "anotherobject"
        metadata      = {}
        storage_class = "STANDARD"
        tags          = {}
        version_id    = "u04WlTsNLJ16GDUVqSxJ5Y0zAwGn7LT8"
    }

Plan: 0 to add, 2 to change, 0 to destroy.

Do you want to perform these actions?
  Terraform will perform the actions described above.
  Only 'yes' will be accepted to approve.

  Enter a value: yes

aws_s3_bucket_object.firstfile: Modifying... [id=someobject]
aws_s3_bucket_object.firstfile: Modifications complete after 0s [id=someobject]

Error: Provider produced inconsistent final plan

When expanding the plan for aws_s3_bucket_object.secondfile to include new
values learned so far during apply, provider
"registry.terraform.io/hashicorp/aws" produced an invalid new value for
.content: was cty.StringVal("a1288b1c7e2257a90bad9bdfb7690fbb -
5_Fxmip9w4DJgZV2or3_SAAn8Rr2QyA5"), but now
cty.StringVal("a1288b1c7e2257a90bad9bdfb7690fbb -
0mNGmhdBL8YmClTknce4vf3eQv5yFfF6").

This is a bug in the provider, which should be reported in the provider's own
issue tracker.

Steps to Reproduce

1.terraform apply -var content=bla

First apply is ok.

Second setup

2. terraform apply -var content=anotherbla

Terraform applies correctly with the updated version_id.

Important Factoids

Changes to content or similar keys that always produce changes to version should be hinted to terraform that it will be a new value. When using a version_id in a following resource when changing the object we get a inconsistent state error message from Terraform regarding the version_id.

Expected behavior is for terraform to be aware a content change is always a version_id change.

Please see connected merge request that fixes this issue.

References

This has been released in version 3.16.0 of the Terraform AWS provider. Please see the Terraform documentation on provider versioning or reach out if you need any assistance upgrading.

For further feature requests or bug reports with this functionality, please create a new GitHub issue following the template for triage. Thanks!

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!