[
https://issues.apache.org/jira/browse/FLINK-33079?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Fangliang Liu updated FLINK-33079:
----------------------------------
Summary: The gap between the checkpoint timeout and the interval settings
is too large in the example (was: The gap between the checkpoint timeout and
the interval settings is too large)
> The gap between the checkpoint timeout and the interval settings is too large
> in the example
> --------------------------------------------------------------------------------------------
>
> Key: FLINK-33079
> URL: https://issues.apache.org/jira/browse/FLINK-33079
> Project: Flink
> Issue Type: Improvement
> Components: Documentation
> Affects Versions: 1.19.0
> Reporter: Fangliang Liu
> Priority: Major
> Attachments: image-2023-09-13-14-17-12-718.png,
> image-2023-09-13-14-19-05-493.png
>
>
> The gap between the checkpoint timeout and the interval settings is too large
> https://nightlies.apache.org/flink/flink-docs-master/docs/dev/datastream/fault-tolerance/checkpointing/
> Some users will think that the documentation is the optimal solution and
> refer to this demo setting, and the result is that the actual checkpoint
> interval is not as expected because of the checkpoint-timeout
> !image-2023-09-13-14-17-12-718.png|width=682,height=468!
> The following situation occurs when the checkpoint interval is set to 20s and
> the checkpoint timeout is set to 10 minutes.
> !image-2023-09-13-14-19-05-493.png|width=1637,height=757!
> So lets do some optimization in the checkpoint example(e.g. checkpoint
> interval 60s, checkpoint timeout 60s), or provide more documentation for
> setting up checkpoints config.
>
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
