azurerm_redhat_openshift_cluster - when deleting a cluster, if its already in the state of deleting it should wait and watch the delete to finish.

[paulczar]

Is there an existing issue for this?

• I have searched the existing issues

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 "me too" comments, 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 and review the contribution guide to help.

Terraform Version

1.4.6

AzureRM Provider Version

3.92.0

Affected Resource(s)/Data Source(s)

azurerm_redhat_openshift_cluster

Terraform Configuration Files

.

Debug Output/Panic Output

.

Expected Behaviour

when running an operation that would delete the cluster, either deleting it or doing a version change, if the cluster is already in a deleting state it should wait for the delete operation to finish and then update its state and move to the next resource.

Actual Behaviour

To perform exactly these actions, run the following command to apply:
    terraform apply "aro.plan"
terraform apply aro.plan
azurerm_redhat_openshift_cluster.cluster: Destroying... [id=/subscriptions/c5545383-1a94-45fe-b501-7ebdf43e5d7a/resourceGroups/aro-xx-rg/providers/Microsoft.RedHatOpenShift/openShiftClusters/aro-pczarkow]
╷
│ Error: deleting Provider Open Shift Cluster (Subscription: "c5545383-1a94-45fe-b501-7ebdf43e5d7a"
│ Resource Group Name: "aro-xx-rg"
│ Open Shift Cluster Name: "aro-xx"): performing Delete: unexpected status 400 with error: RequestNotAllowed: Request is not allowed in provisioningState 'Deleting'.
│
│ deleting Provider Open Shift Cluster (Subscription: "c5545383-1a94-45fe-b501-7ebdf43e5d7a"
│ Resource Group Name: "aro-xx-rg"
│ Open Shift Cluster Name: "aro-xx"): performing Delete: unexpected status 400 with error: RequestNotAllowed:
│ Request is not allowed in provisioningState 'Deleting'.
╵
make: *** [Makefile:18: create] Error 1

Steps to Reproduce

No response

Important Factoids

No response

References

No response

[teowa]

@paulczar , thanks for submitting this.
From the source code, when we delete the azurerm_redhat_openshift_cluster resource, the provider is supposed wait until the response code changed from 200 to 404. Is it possible the deletion is triggered outside of terraform or in previous Terraform apply? Could you please share the debug log containing API https flow with TF_LOG=debug (please hide sensitive info in the log).

terraform-provider-azurerm/internal/services/redhatopenshift/redhat_openshift_cluster_resource.go

Line 517 in 2d02a31

if err := client.DeleteThenPoll(ctx, *id); err != nil {

terraform-provider-azurerm/vendor/github.com/hashicorp/go-azure-sdk/resource-manager/redhatopenshift/2023-09-04/openshiftclusters/method_delete.go

Line 65 in 2d02a31

if err := result.Poller.PollUntilDone(ctx); err != nil {

terraform-provider-azurerm/vendor/github.com/hashicorp/go-azure-sdk/sdk/client/pollers/poller.go

Line 124 in 2d02a31

p.latestResponse, p.latestError = p.poller.Poll(ctx)

terraform-provider-azurerm/vendor/github.com/hashicorp/go-azure-sdk/sdk/client/resourcemanager/poller_delete.go

Lines 88 to 102 in 2d02a31

	if resp.Response != nil {
	switch resp.StatusCode {
	case http.StatusNotFound:
	{
	result.Status = pollers.PollingStatusSucceeded
	return
	}
	case http.StatusOK:
	{
	result.Status = pollers.PollingStatusInProgress
	return
	}
	}