[
https://issues.apache.org/jira/browse/FLINK-20653?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17253862#comment-17253862
]
weiyunqing commented on FLINK-20653:
------------------------------------
[~yunta]There is a problem with the code flow here. It should not be 【register
and clear the incremental state first, then store metadata】 but 【store metadata
first, then register and clean up the incremental state】 or 【register the
incremental state first, store metadata secondly, and finally clean up the
referenced incremental state】
> SharedState should be registered after PendingCheckpoint storeCheckpoint
> ------------------------------------------------------------------------
>
> Key: FLINK-20653
> URL: https://issues.apache.org/jira/browse/FLINK-20653
> Project: Flink
> Issue Type: Bug
> Components: Runtime / Checkpointing
> Affects Versions: 1.12.0
> Reporter: weiyunqing
> Priority: Major
> Attachments: image-2020-12-17-20-20-21-246.png
>
>
> SharedState should be registered after PendingCheckpoint storeCheckpoint，If
> the metadata storage fails, the deleted incremental state cannot be recovered。
> !image-2020-12-17-20-20-21-246.png!
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
