I have already read the comment for this method, yes, it will cause the test program to become unresponsive. But each PR will verify this test, why it passed very many times, but this PR, I committed twice, it failed on this use case. Therefore, I suspect that it is caused by the kafka 2.0 version. But with your reminder, it looks like it's a probabilistic problem, and it can fail in existing code.
[ Full content available at: https://github.com/apache/flink/pull/6703 ] This message was relayed via gitbox.apache.org for [email protected]
