Allow setting undrainable_node_behavior in upgrade #719
Description
Is there an existing issue for this?
- I have searched the existing issues
Greenfield/Brownfield provisioning
brownfield
Terraform Version
1.14.1
Module Version
11.0
AzureRM Provider Version
4.48.0
Affected Resource(s)/Data Source(s)
azurerm_kubernetes_cluster, azurerm_kubernetes_cluster_node_pool
Terraform Configuration Files
module "aks" {
# ... other required variables ...
undrainable_node_behavior = "Cordon"
}tfvars variables values
N/ADebug Output/Panic Output
# module.cluster.module.aks.azurerm_kubernetes_cluster.main must be replaced
+/- resource "azurerm_kubernetes_cluster" "main" {
......
~ upgrade_settings {
- drain_timeout_in_minutes = 0 -> null
- undrainable_node_behavior = "Cordon" -> null # forces replacement
# (2 unchanged attributes hidden)
}
}
}Expected Behaviour
Terraform attempts to set the undrainable_node_behavior to null for the cluster and node pool upgrade settings because this setting is not supported by the module.
Changing this setting will force the recreation of the cluster even if nothing has actually changed.
Actual Behaviour
Module should support the passing of this variable undrainable_node_behavior to prevent it trying to recreate the cluster or node pools.
Steps to Reproduce
Upgrade the AKS module to 11.0 and run a terraform apply for an existing module.
Important Factoids
No response
References
No response