pnowojski commented on a change in pull request #8602: [FLINK-12313] Add
workaround to avoid race condition in SynchronousCheckpointITCase test
URL: https://github.com/apache/flink/pull/8602#discussion_r293255730
##########
File path:
flink-streaming-java/src/test/java/org/apache/flink/streaming/runtime/tasks/SynchronousCheckpointITCase.java
##########
@@ -108,6 +110,9 @@ public void initializeLatchesAndError() {
error.set(null);
}
+ @Rule
+ public final Timeout timeoutPerTest = Timeout.seconds(7);
Review comment:
Just saying from my experience: I would set the timeout to 60 seconds.
Travis can often stall for many seconds. I've never investigated the reasons
why (swapping? gc? vm failover?), but the problem with travis is not that it's
2 or 3 times slower, but that it can randomly stall from time to time. For
example a test might usually execute in 2 seconds, but from time to time it
will take 30 seconds to complete.
On the other hand, is there any benefit of timeouting those tests in 7s
instead of 60s? On travis I think there is none. Locally there is benefit only
if you are debugging this particular test failure for shorter feedback loop,
which is also a rare case and one can manually and temporarily lower the
timeout. I think tests that are setting some lower timeouts are actively
waiting on it and that creates a motivation to lower the timeout as low as
possible (to speed up the tests).
(If you insist on trying this 7s timeout, I won't block it)
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
