atman9

I am hitting this issue as well. Even with `max.poll.records` set as high as 4000, I can't get more than about 80-100 rows per batch. I have 8 partitions across...

I was able to get greater throughput by adjusting the following properties in the `connect-distributed.properties`: ``` # based on 500*180byte message size consumer.fetch.min.bytes=900000 consumer.fetch.wait.max.ms=1000 ```