[
https://issues.apache.org/jira/browse/FLINK-21218?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Yun Tang closed FLINK-21218.
----------------------------
Resolution: Information Provided
Since Flink already use new API EmbeddedRocksDBStateBackend to replace
RocksDBStateBackend in state backend loader, this problem has gone.
> "state.checkpoints.dir" should be required only when
> "execution.checkpointing.interval" is specified
> ----------------------------------------------------------------------------------------------------
>
> Key: FLINK-21218
> URL: https://issues.apache.org/jira/browse/FLINK-21218
> Project: Flink
> Issue Type: Improvement
> Components: Runtime / State Backends
> Reporter: Dongwon Kim
> Priority: Minor
>
> Users have to specify "state.checkpoints.dir" even when to use a state
> backend as an unreliable per-key state storage.
> Thread in user ML :
> http://apache-flink-user-mailing-list-archive.2336050.n4.nabble.com/Setting-quot-unreliable-quot-RocksDB-state-backend-w-o-quot-execution-checkpointing-interval-quot-ans-td41003.html
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
