[
https://issues.apache.org/jira/browse/HUDI-6087?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
ASF GitHub Bot updated HUDI-6087:
---------------------------------
Labels: pull-request-available (was: )
> Hudi streaming-read is not stopping with savepoint correctly
> ------------------------------------------------------------
>
> Key: HUDI-6087
> URL: https://issues.apache.org/jira/browse/HUDI-6087
> Project: Apache Hudi
> Issue Type: Bug
> Reporter: voon
> Assignee: voon
> Priority: Major
> Labels: pull-request-available
>
> Flink supports stopping with savepoint as documented here:
> [https://nightlies.apache.org/flink/flink-docs-master/docs/ops/state/savepoints/#stopping-a-job-with-savepoint]
>
> Stopping with savepoint will invoke these 3 interface functions Flink
> functions.
> # cancel()
> # snapshotState()
> # close()
>
> However, the current implementation of stopping with savepoint will cause an
> *issuedInstant* to be *null* when snapshotState is invoked. This is so as
> cancel() will set the *issuedInstant* to {*}null{*}, causing the
> *snapshotState()* to add a *null* value to the *LinkState* list.
>
> As such, when resuming from a savepoint, there will be:
> # data loss if the LATEST (by not specifying a value for, i.e. default
> value) *_read.start.commit_* is used
> # duplicated data if the EARLIEST *_read.start.commit_* is used
>
>
>
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
