[
https://issues.apache.org/jira/browse/HUDI-6501?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Nicholas Jiang updated HUDI-6501:
---------------------------------
Description: When the last write had failed, HoodieHeartbeatClient would
close with stopping all heartbeat which includes deleting heartbeat file.
Therefore StreamWriteOperatorCoordinator should start heartbeat for lazy failed
writes clean policy to avoid aborting for heartbeat expired when recommitting.
(was: When the last write had failed, HoodieHeartbeatClient would close with
stopping all heartbeat which includes deleting heartbeat file in flink job.
Therefore, it isn't heartbeat expired for the last failed write when
HoodieHeartbeatClient recommits. Fix HoodieHeartbeatClient does not check
whether heartbeat is expired with the last failed writes for recommit.)
> Recommit should not abort for heartbeat expired caused by the last failed
> write
> -------------------------------------------------------------------------------
>
> Key: HUDI-6501
> URL: https://issues.apache.org/jira/browse/HUDI-6501
> Project: Apache Hudi
> Issue Type: Bug
> Components: core
> Affects Versions: 0.14.0
> Reporter: Nicholas Jiang
> Assignee: Nicholas Jiang
> Priority: Major
> Labels: pull-request-available
> Fix For: 0.14.0
>
>
> When the last write had failed, HoodieHeartbeatClient would close with
> stopping all heartbeat which includes deleting heartbeat file. Therefore
> StreamWriteOperatorCoordinator should start heartbeat for lazy failed writes
> clean policy to avoid aborting for heartbeat expired when recommitting.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
