I have created python script which basically turn off the autoscale of GKE cluster , then stop the underlying nodes in each MIG(Managed Instance Group) present zone wise. Now the problem is while stopping the instances , instances being recreated and status is getting changed from UNKNOWN to READY state immediate after stopping the instance. Though I turned off autoscale group before instance being stopped. on the other hand this is not the case while I manually stop the instance from console itself in each MIG per ZONE. why its happening ?can anyone suggest what needs to be checked or what more I need to apply on the code.