[
https://issues.apache.org/jira/browse/FLINK-33498?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17791986#comment-17791986
]
Zakelly Lan commented on FLINK-33498:
-------------------------------------
I notice that the incremental checkpoint is configured by
'state.backend.incremental' and the CheckpointCoordinator does not know the
original value of this (It is configured by 'execution.checkpointing.xxx').
Even so, the coordinator sending the strategy of sharing files to state
backends and the state backends take this only a suggestion. It is a weak
contract between them and it is not easy to fail checkpoint triggering request
based on incremental/full option in runtime.
I suggest we disable the incremental request from REST API first, and consider
a clearer division of responsibilities between coordinator and state backends.
> Do not allow manually triggering incremental checkpoints with full checkpoint
> configured
> ----------------------------------------------------------------------------------------
>
> Key: FLINK-33498
> URL: https://issues.apache.org/jira/browse/FLINK-33498
> Project: Flink
> Issue Type: Improvement
> Components: Runtime / Checkpointing
> Reporter: Zakelly Lan
> Assignee: Zakelly Lan
> Priority: Major
> Fix For: 1.19.0
>
>
> Currently, when a job is configured to run with incremental checkpoint
> disabled, user manually triggers an incremental checkpoint actually
> triggering a full checkpoint. That is because the files from full checkpoint
> cannot be shared with an incremental checkpoint. So we'd better throw some
> exception somewhere around {{CheckpointCoordinator}} and fail the request in
> this case.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
