🐛 Plannig via annotation or UI leaves Module Object in old state

[arne21a]

Operator Version, Kind and Kubernetes Version

• Operator version: v2.9.2
• Kind: Workspace | Module
• Kubernetes version: v1.32.6-eks-931bdca

YAML Manifest File

apiVersion: app.terraform.io/v1alpha2
kind: Workspace
metadata:
  annotations:
    workspace.app.terraform.io/run-new: "false"
    workspace.app.terraform.io/run-type: refresh
...
status:
  defaultProjectID: prj-redacted
  observedGeneration: 1
  plan:
    id: run-UTJXBvCPKaNnmV4q <-- correct latests run id
    status: planned_and_finished
    terraformVersion: 1.10.4
  runStatus:
    configurationVersion: cv-zaySUX6qZKaRk8iL
    id: run-TQMd4v5fbPQPPVp3
    outputRunID: run-TQMd4v5fbPQPPVp3
    status: applied
  terraformVersion: 1.10.4
  updateAt: 1755854482

---
apiVersion: app.terraform.io/v1alpha2
kind: Module
metadata:
  annotations:
  name: my-module
  namespace: tfc-operator-configuration
spec:
    ...
status:
  configurationVersion:
    id: cv-zaySUX6qZKaRk8iL
    status: uploaded
  observedGeneration: 1
  run:
    configurationVersion: cv-zaySUX6qZKaRk8iL
    id: run-s4NjHXVcJ3EXqo3y <-- old (first) run ID
    status: errored
  workspaceID: ws-AuNESoWvTWy2V2i7

Output Log

Output of relevant kubectl commands

Steps To Reproduce

• Create a Workspace and Module (in my case the first run failed. I do not know if that is relevant)
• Annotate the workspace with

    workspace.app.terraform.io/run-new: "true"
    workspace.app.terraform.io/run-type: refresh

• Wait for refresh to finish

Expected Behavior

What should have happened?
The module should show the latest run ID and update the status from errored to applied

Actual Behavior

What actually happened?
The module still shows the first run ID and does not update its status.

Additional Context

Add any other context about the problem here.

References

Community Note

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

[arne21a]

I found a workaround to trigger module reconciliation manually:

You can force a reconciliation by patching the spec.restartedAt field with the current UTC timestamp using the following command:

kubectl patch module <NAME> \
  --type=merge \
  --patch '{"spec": {"restartedAt": "'`date -u -Iseconds`'"}}'

as stated in the docs:
https://github.com/hashicorp/hcp-terraform-operator/blob/main/docs/module.md

Even though this workaround exists, I would expect the operator to handle reconciliation automatically when changes occur, without requiring manual intervention.