[
https://issues.apache.org/jira/browse/FLINK-8085?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16280031#comment-16280031
]
ASF GitHub Bot commented on FLINK-8085:
---------------------------------------
Github user GJL commented on a diff in the pull request:
https://github.com/apache/flink/pull/5087#discussion_r155207330
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/executiongraph/ExecutionGraph.java
---
@@ -965,11 +965,20 @@ public void cancel() {
// we build a future that is complete
once all vertices have reached a terminal state
final ConjunctFuture<Void> allTerminal
= FutureUtils.waitForAll(futures);
- allTerminal.thenAccept(
- (Void value) -> {
- // cancellations may
currently be overridden by failures which trigger
- // restarts, so we need
to pass a proper restart global version here
-
allVerticesInTerminalState(globalVersionForRestart);
+ allTerminal.whenCompleteAsync(
--- End diff --
Does it have to run asynchronously? If yes, does it make sense to specify a
thread pool? Now it can run on `ForkJoinPool.commonPool()`.
> Thin out the LogicalSlot interface
> ----------------------------------
>
> Key: FLINK-8085
> URL: https://issues.apache.org/jira/browse/FLINK-8085
> Project: Flink
> Issue Type: Improvement
> Components: Distributed Coordination
> Affects Versions: 1.5.0
> Reporter: Till Rohrmann
> Assignee: Till Rohrmann
> Priority: Minor
> Labels: flip-6
>
> The {{LogicalSlot}} interface contains method which we don't strictly need
> (e.g. {{isCanceled}}, {{isReleased}}). Moreover, we should decouple the
> {{LogicalSlot}} from the {{Execution}} by only setting the
> {{ExecutionAttemptID}} instead of {{Execution}}.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
