Github user srdo commented on the issue:
https://github.com/apache/storm/pull/1825
Hi. STORM-2077 is being resolved here
https://issues.apache.org/jira/browse/STORM-2087. Sorry for the inconvenience.
I'm not sure I understand why adding manual partition management is
necessary. Kafka has recently added heartbeating from a background thread, so
it's no longer necessary to poll often in order to avoid rebalances (see
http://kafka.apache.org/documentation.html#upgrade_1010_notable, search for
"heartbeat"). That change decouples necessary poll frequency from the time it
would take to detect a crashed consumer, so the `max.poll.interval.ms` can be
bumped if a rebalance every 5 minutes is unacceptable (and then only if poll
isn't called in that interval).
Could you describe a use case?
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
