aws_availability_zones started proposing changes on each apply

[vladimirtiukhtin]

Terraform CLI and Terraform AWS Provider Version

Terraform v0.13.0-rc1
+ provider registry.terraform.io/-/aws v2.70.0
+ provider registry.terraform.io/-/random v2.3.0
+ provider registry.terraform.io/hashicorp/aws v2.70.0
+ provider registry.terraform.io/hashicorp/random v2.3.0

Affected Resource(s)

• aws_availability_zones

Debug Output

 <= data "aws_availability_zones" "available"  {
        group_names = [
            "eu-west-2",
        ]
      ~ id          = "2020-07-30 12:47:26.4286656 +0000 UTC" -> "2020-07-30 12:47:47.9707994 +0000 UTC"

Expected Behavior

No changes proposed

[bflad]

Hi @vladimirtiukhtin 👋 Thank you for submitting this and sorry you ran into trouble here. Can you please provide the relevant data source configuration?

[vladimirtiukhtin]

Thanks for the prompt response @bflad. My configuration is incredibly simple

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

[bflad]

This should fix it: #14412

I'm guessing there is going to be other instances of this unexpected behavior with other data sources in Terraform 0.13 as well, but its going to be a week or two before we can easily gain more visibility into this type of issue.

[bflad]

The fix for this has been merged and will release with version 3.1.0 of the Terraform AWS Provider, next week. 👍

[ghost]

This has been released in version 3.1.0 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!

[vladimirtiukhtin]

@bflad I'm afraid the issue still exists in 3.1.0. It looks like the root cause is in ID which became dependant on current data/time

[vladimirtiukhtin]

Should we open a new issue or re-open this one?

[jurgenweber]

This still seems to be broken:

$ tf version
Terraform v0.13.0
+ provider registry.terraform.io/hashicorp/aws v3.1.0
+ provider registry.terraform.io/hashicorp/helm v1.2.4
+ provider registry.terraform.io/hashicorp/kubernetes v1.12.0
+ provider registry.terraform.io/hashicorp/local v1.4.0
+ provider registry.terraform.io/hashicorp/null v2.1.2
+ provider registry.terraform.io/hashicorp/random v2.3.0
+ provider registry.terraform.io/hashicorp/template v2.1.2
+ provider registry.terraform.io/hashicorp/tfe v0.20.0

and:

 <= data "aws_availability_zones" "available"  {
        group_names = [
            "ap-southeast-2",
        ]
      ~ id          = "2020-08-14 00:15:16.884539377 +0000 UTC" -> "2020-08-14 00:15:48.674203932 +0000 UTC"
        names       = [
            "ap-southeast-2a",
            "ap-southeast-2b",
            "ap-southeast-2c",
        ]
        state       = "available"
        zone_ids    = [
            "apse2-az3",
            "apse2-az1",
            "apse2-az2",
        ]
    }

it also seems to be that many data providers are suffering from this; I also see it from:

<= data "aws_kms_secrets" "stuff"  {
      ~ id        = "2020-08-14 00:15:19.986658155 +0000 UTC" -> "2020-08-14 00:15:48.477072138 +0000 UTC"
        plaintext = {

and its dumping my secrets to console as plain text!!!!!

also:

#14649

[bflad]

Hi folks 👋

There was some unexpected changes with data source handling that occurred in the Terraform CLI version 0.13.0 release that are causing this additional data source output in addition to the unstable id attribute in the output. There are now larger tracking issues for both sides of this so to consolidate efforts and discussions:

• Handle semantically equivalent changes from the legacy SDK when planning data sources terraform#25812
• Data sources with non-deterministic IDs show perpetual diff with Terraform 0.13 (R015, R016, R017 linters) #14579

Please upvote and subscribe to those for further updates. 👍 I'm going to lock this issue to encourage any discussion in those.