Github user zentol commented on the issue:
https://github.com/apache/flink/pull/6352
yay travis is green.
---
Github user tillrohrmann commented on the issue:
https://github.com/apache/flink/pull/6352
Interesting. I thought I would have executed `flink-yarn-tests` locally
with all tests. But it looks as if `YARNHighAvailabilityITCase` suffers from
the same problem as the `YARNSessionFIFOITcas
Github user zentol commented on the issue:
https://github.com/apache/flink/pull/6352
The `YARNHighAvailabilityITCase` still has the same problem.
---
Github user tillrohrmann commented on the issue:
https://github.com/apache/flink/pull/6352
The failing test case seems to be caused by
`YARNSessionFIFOITCase#testJavaAPI` not calling `yarnCluster.shutdownCluster`.
I've fixed this problem in master and the release branch. Rebasing onto
Github user zentol commented on the issue:
https://github.com/apache/flink/pull/6352
will look into them tomorrow
---
Github user zentol commented on the issue:
https://github.com/apache/flink/pull/6352
the test failures may highlight tests that weren't shutting down the last
application properly; previously this would've succeeded since the check was
done in `@Before`.
---