[
https://issues.apache.org/jira/browse/FLINK-35872?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17867749#comment-17867749
]
dalongliu commented on FLINK-35872:
-----------------------------------
Merged in master: af7d2b3ab0e0ecc4157005704bebac9c767f2e1a
Release-1.20: a3268544636ee120dd6cbed1ba429aafdca2909a
> Fix the incorrect partition generation during period refresh in Full Mode
> -------------------------------------------------------------------------
>
> Key: FLINK-35872
> URL: https://issues.apache.org/jira/browse/FLINK-35872
> Project: Flink
> Issue Type: Sub-task
> Components: Table SQL / API, Table SQL / Gateway
> Affects Versions: 1.20.0
> Reporter: Feng Jin
> Assignee: Feng Jin
> Priority: Blocker
> Labels: pull-request-available
>
> During the period refresh, we currently default to using scheduler time to
> generate the target partition, but we should use the T - 1 partition instead.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
