[
https://issues.apache.org/jira/browse/FLINK-4384?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15418570#comment-15418570
]
ASF GitHub Bot commented on FLINK-4384:
---------------------------------------
Github user tillrohrmann commented on a diff in the pull request:
https://github.com/apache/flink/pull/2360#discussion_r74563211
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/rpc/MainThreadExecutor.java
---
@@ -52,4 +52,13 @@
* @return Future of the callable result
*/
<V> Future<V> callAsync(Callable<V> callable, Timeout callTimeout);
+
+ /**
+ * Execute the runnable in the main thread of the underlying RPC
endpoint, with
+ * a delay of the given number of milliseconds.
+ *
+ * @param runnable Runnable to be executed
+ * @param delay The delay after which the runnable will be executed
+ */
+ void scheduleRunAsync(Runnable runnable, long delay);
--- End diff --
That is true. The `TimeUnit` variant would make it a little bit more
explicit for the user that he has to think about the actual delay (unit-wise).
On the other hand, milliseconds are also kind of the default time unit used
also in other places.
> Add a "scheduleRunAsync()" feature to the RpcEndpoint
> -----------------------------------------------------
>
> Key: FLINK-4384
> URL: https://issues.apache.org/jira/browse/FLINK-4384
> Project: Flink
> Issue Type: Sub-task
> Components: Distributed Coordination
> Environment: FLIP-6 feature branch
> Reporter: Stephan Ewen
> Fix For: 1.2.0
>
>
> It is a common pattern to schedule a call to be executed in the future.
> Examples are
> - delays in retries
> - heartbeats,
> - checking for heartbeat timeouts
> I suggest to add a {{scheduleRunAsync()}} method to the {{RpcEndpoint}}.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
