Hi, Yesterday we ran into problems regarding the distribution of records across Kafka where Flink was used as the producer. So we fixed this and realized that the code to do this would be useful to others.
I put up a Jira ticket and pull request yesterday and it passes all automated tests. Please review. https://issues.apache.org/jira/browse/FLINK-9610 https://github.com/apache/flink/pull/6181 -- Best regards / Met vriendelijke groeten, Niels Basjes
