Andrey

👍

@raychenon thank you for contribution! I have added the issue for the next team alignment meeting, I will notify you about the outcome once it is ready.

@juhovuori thank you very much for reporting! It looks unstable, It passes on travis https://travis-ci.org/zalando/nakadi though. We're aware of a couple of such tests, which has to be fixed. If...

@eupestov thank you for filing an issue. just to clarify `-1` is a valid kafka `retention.ms` https://kafka.apache.org/documentation/#topicconfigs

> Yeah, that why we start using it. But is it to be officially supported by Nakadi? Taking into account it's future support for other messaging backends and other features...

Added a small fix. It worked for me. Maybe you will have time to check this out. https://github.com/zalando-stups/senza/pull/411

@RossierFl I came across similar behaviour, have you found any solutions?

@efeg maybe you could clarify that as well

there is a problem with this approach: client sets 300ms timeout, connection is close by Nakadi due to timeout (300ms), but Kafka Producer continues to retry events for this request...

@gchudnov docs update is great. thank you!