Terraform-provider-helm: Any plans as of the morning of Dec 16th with helm_releases in EKS attempt to change version of helm release to 0.2.2

[ajcann]

As strange as it sounds, across 8 clusters with 0 local or remote state changes and with no module changes (confirmed with three engineers), any attempt to create a plan that should result in nothing attempt to change versions of all installed helm_releases from whatever their current version is to 0.2.2.

[doug-ba]

This exact problem is affecting us, too. We're running Terraform v0.12.21, terraform-provider-helm_v1.0.0_x4 (but also tried latest).
I use a local plugin cache, so the plugin did not change.

We made no changes to our terraform code. This worked normally yesterday.

This is the relevant section of my debug log:
2020/12/16 17:01:52 [WARN] Provider "registry.terraform.io/-/helm" produced an invalid plan for helm_release.ordered_helm_release[0], but we are tolerating it because it is using the legacy plugin SDK. The following problems may be the cause of any confusing errors from downstream operations: - .version: planned value cty.StringVal("0.2.2") does not match config value cty.StringVal("1.33.1") nor prior value cty.StringVal("1.33.1")
2020/12/16 17:01:53 [WARN] Provider "registry.terraform.io/-/helm" produced an invalid plan for helm_release.openstack_external_dns[0], but we are tolerating it because it is using the legacy plugin SDK. The following problems may be the cause of any confusing errors from downstream operations: - .version: planned value cty.StringVal("0.2.2") does not match config value cty.StringVal("2.19.1") nor prior value cty.StringVal("2.19.0")

[jharris-]

I am also seeing this on a few non-managed kube clusters.

2020/12/16 19:11:40 [WARN] Provider "registry.terraform.io/-/helm" produced an invalid plan for helm_release.ordered_helm_release[0], but we are tolerating it because it is using t
he legacy plugin SDK.
    The following problems may be the cause of any confusing errors from downstream operations:
      - .version: planned value cty.StringVal("0.2.2") does not match config value cty.StringVal("1.33.1") nor prior value cty.StringVal("1.33.1")

[ajcann]

Okay I think the issue might be that the default repository ( https://github.com/helm/charts ) is now deprecated and creating unspecified behavior. Updating the repository field to an actively supported repo with your charts is likely the fix here. Testing now.

[doug-ba]

Looks like you may be right. I switched my "stable" repository to point to https://charts.helm.sh/stable as outlined here: https://helm.sh/blog/new-location-stable-incubator-charts/ and am no longer seeing the incorrect version. I haven't actually applied the changes yet.

[yasra002]

I tested updating the repository field and it worked for us and problem was solved.

[AndreaGiardini]

I can confirm that updating the repository field worked for us. No redeploy needed

[jrhouston]

Yeah this will likely have been caused by the the stable repository being sunset. We haven't put out a new release of the provider yet. See their documentation for more details.

[andreabolandrina]

that "0.2.2" version everywhere seemed a bit weird.
In my case, I had to replace https://kubernetes-charts.storage.googleapis.com with https://charts.helm.sh/stable/.

[aareet]

Closing this issue since the solution is discussed above.