Open
Description
Hello,
This issue is maybe related to hashicorp/terraform-provider-kubernetes-alpha#41, I create an issue here to track this behavior and share information.
We manage our kubernetes cluster via terraform, to enforce maintenability and collaboration (using gitlab and Merge requests).
To deploy CRD we are using kubernetes-alpha provider from hashicorp.
Some parameters are not idempotents, and when we run apply two times, the second one is failing cause of dry-run failure.
resource "kubernetes_manifest" "arangodb-db" {
provider = kubernetes-alpha
manifest = {
"apiVersion" = "database.arangodb.com/v1alpha"
"kind" = "ArangoDeployment"
"metadata" = {
"name" = "test"
}
"spec" = {
"mode" = "Single"
"bootstrap" = {
"passwordSecretNames" = {
"root" = "Auto"
}
}
}
}
}
Some other behaviors results to a failure, first time apply the previous hcl, then change it to
resource "kubernetes_manifest" "arangodb-db" {
provider = kubernetes-alpha
manifest = {
"apiVersion" = "database.arangodb.com/v1alpha"
"kind" = "ArangoDeployment"
"metadata" = {
"name" = "test"
}
"spec" = {
"mode" = "Single"
"bootstrap" = {
"passwordSecretNames" = {
"root" = "Auto"
}
}
"tls" = {
"caSecretName" = "None"
}
}
}
}
It's failure, but expected to have a force replace scenario
Best regards,
Gaël
Metadata
Metadata
Assignees
Labels
No labels