Github user tillrohrmann commented on a diff in the pull request:
https://github.com/apache/flink/pull/5437#discussion_r168525466
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/dispatcher/MiniDispatcher.java
---
@@ -100,6 +101,18 @@ public MiniDispatcher(
return acknowledgeCompletableFuture;
}
+ @Override
+ public CompletableFuture<JobResult> requestJobResult(JobID jobId, Time
timeout) {
+ final CompletableFuture<JobResult> jobResultFuture =
super.requestJobResult(jobId, timeout);
+
+ if (executionMode == ClusterEntrypoint.ExecutionMode.NORMAL) {
+ // terminate the MiniDispatcher once we served the
first JobResult successfully
+ jobResultFuture.thenRun(this::shutDown);
--- End diff --
I think you're right. We should also shut down the `MiniDispatcher` in case
of a failure.
---
