If the client triggers more operations after an initiated cluster shutdown, we would not wait for that operation to complete, or for the result to be served to the client. The client would likely see a timeout while polling for the result. I am ok with such semantics. Alternatively, we can answer the requests against the `TriggerHandler` with status code 503. What do you think?
[ Full content available at: https://github.com/apache/flink/pull/6785 ] This message was relayed via gitbox.apache.org for [email protected]
