[
https://issues.apache.org/jira/browse/FLINK-6567?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16418717#comment-16418717
]
ASF GitHub Bot commented on FLINK-6567:
---------------------------------------
Github user tillrohrmann commented on a diff in the pull request:
https://github.com/apache/flink/pull/5782#discussion_r178017114
--- Diff:
flink-runtime/src/test/java/org/apache/flink/runtime/executiongraph/ExecutionGraphMetricsTest.java
---
@@ -140,6 +140,9 @@ public void testExecutionGraphRestartTimeMetric()
throws JobException, IOExcepti
assertTrue(currentRestartingTime >=
previousRestartingTime);
previousRestartingTime = currentRestartingTime;
+
+ // add some pause to let the
currentRestartingTime increase
+ Thread.sleep(1L);
--- End diff --
I think we have to keep it here, because otherwise the loop might finish so
fast that we don't see an increase in `previousRestartingTime` because this
value is effectively `System.currentTimeMillis - timestamps[RESTARTING]`.
> ExecutionGraphMetricsTest fails on Windows CI
> ---------------------------------------------
>
> Key: FLINK-6567
> URL: https://issues.apache.org/jira/browse/FLINK-6567
> Project: Flink
> Issue Type: Bug
> Components: Tests
> Affects Versions: 1.3.0, 1.4.0
> Reporter: Chesnay Schepler
> Assignee: Till Rohrmann
> Priority: Blocker
> Labels: test-stability
> Fix For: 1.6.0
>
>
> The {{testExecutionGraphRestartTimeMetric}} fails every time i run it on
> AppVeyor. It also very rarely failed for me locally.
> The test fails at Line 235 if the RUNNING timestamp is equal to the
> RESTARTING timestamp, which may happen when combining a fast test with a low
> resolution clock.
> A simple fix would be to increase the timestamp between RUNNING and
> RESTARTING by adding a 50ms sleep timeout into the
> {{TestingRestartStrategy#canRestart()}} method, as this one is called before
> transitioning to the RESTARTING state.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
