[
https://issues.apache.org/jira/browse/FLINK-24668?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17441566#comment-17441566
]
Yufan Sheng commented on FLINK-24668:
-------------------------------------
Hi [~余嘉悦], I can understand what you have said. The behavior for
{{RecordsBySplits}} is totally accepted.
Can you show me some progress on how to reproduce your issues?
> Pulsar consumption data repeated output
> ---------------------------------------
>
> Key: FLINK-24668
> URL: https://issues.apache.org/jira/browse/FLINK-24668
> Project: Flink
> Issue Type: Bug
> Components: Connectors / Pulsar
> Affects Versions: 1.14.0
> Reporter: jiayue.yu
> Assignee: Yufan Sheng
> Priority: Major
>
> Flink-connector-pulsar uses the RecordsBySplits of Flink-connector-base to
> process data, But recordsInCurrentSplit and splitsIterator are not empty when
> nextSplit splitSiterator.hasNext () = false.
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
