[
https://issues.apache.org/jira/browse/KAFKA-6182?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17110137#comment-17110137
]
Antony Stubbs commented on KAFKA-6182:
--------------------------------------
Ah yes that makes sense. Assuming the exception thrown when the partition
mismatch is found at runtime would cause the entire system to fail fast, I
think yes, this repartition operation solves the issue.
> Automatic co-partitioning of topics via automatic intermediate topic with
> matching partitions
> ---------------------------------------------------------------------------------------------
>
> Key: KAFKA-6182
> URL: https://issues.apache.org/jira/browse/KAFKA-6182
> Project: Kafka
> Issue Type: New Feature
> Components: streams
> Affects Versions: 1.0.0
> Reporter: Antony Stubbs
> Priority: Major
>
> Currently it is up to the user to ensure that two input topics for a join
> have the same number of partitions, and if they don't, manually create an
> intermediate topic, and send the stream #through that topic first, and then
> performing the join.
> It would be great to have Kafka streams detect this and at least give the
> user the option to create an intermediate topic automatically with the same
> number of partitions as the topic being joined with.
> See
> https://docs.confluent.io/current/streams/developer-guide.html#joins-require-co-partitioning-of-the-input-data
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
