When Overwatch modules outrun scale-up speeds it's possible that the Overwatch scale-up requests stack up and cause the cluster to get into an unstable state (usually with long-running init-scripts).
Potential solution:
Wrap the cluster resize API call in a future and validate status before requesting new size.
When Overwatch modules outrun scale-up speeds it's possible that the Overwatch scale-up requests stack up and cause the cluster to get into an unstable state (usually with long-running init-scripts).
Potential solution: Wrap the cluster resize API call in a future and validate status before requesting new size.
Complete with #372