Closed r4hulgupta closed 2 years ago
With the field named initial_node_count
as is, changing the count up/down would imply the cluster was provisioned w/ a different initial count. I don't believe that's a change we want to make. https://registry.terraform.io/providers/hashicorp/google/latest/docs/resources/container_node_pool#node_count is the field to manually update the # of nodes in a pool.
I'm going to lock this issue because it has been closed for 30 days ⏳. This helps our maintainers find and focus on the active issues. If you have found a problem that seems similar to this, please open a new issue and complete the issue template so we can capture all the details necessary to investigate further.
Community Note
Description
Changing the value of initial_node_count forces a node pool destroy and create. In general when doing such an operation using
gcloud container clusters resize <CLUSTER_NAME> --num-nodes <> --node-pool <>
does not really destroy and create the node pool.The current resource implementation has ForceNew set to true and it maybe as simple as setting it to false?
I can assume there was some history behind the current implementation but I would like to propose this change and understand why if it can't be done.
The reason behind changing initial_node_count to scale up and down a node-pool manually is so that users can adjust their node pool minimum size which only changes during a scale up event by the autoscaler even if the current node pool size is <= min_node_count. So if a user wants their min_node_count to be increased it won't trigger a scale up until autoscaler sees the need to scale up, and in that case changing initial_node_count could help get to the desired number of nodes without destroying and creating the node pool.
New or Affected Resource(s)
resource_container_cluster_node_pool resource_container_cluster
Potential Terraform Configuration
Proposed changes do not make changes to terraform configuration. It could be simply changing ForceNew to False.
References