[
https://issues.apache.org/jira/browse/FLINK-22684?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17346225#comment-17346225
]
Roman Khachatryan commented on FLINK-22684:
-------------------------------------------
# I'm assuming you are referring to
CheckpointConfig+ExecutionCheckpointingOptions vs
SavepointConfigOptions+SavepointRestoreSettings. The former one seems more
suitable as we indeed don't have any in-flight data in savepoints currently.
And if we add it in the future, it can be a cross-cutting concern, similar to
PREFER_CHECKPOINT_FOR_RECOVERY
# I think we should currently have ALL and NONE with the ability to add more
values in the future
# IIUC, it shouldn't be an issue: we have to load _metadata file anyways and
deserialize all the handles in it
# As this is an exceptional situation, I think the setting should apply only
to a checkpoint from which the job is started manually. If later it is
restarted automatically from the SAME checkpoint, it should also apply. And if
later it is restarted from a different checkpoint, the setting should NOT
apply. I didn't look into the code though and I'm not sure whether it's
possible and worth the effort to differentiate all these cases
> Add the ability to ignore in-flight data on recovery
> ----------------------------------------------------
>
> Key: FLINK-22684
> URL: https://issues.apache.org/jira/browse/FLINK-22684
> Project: Flink
> Issue Type: Improvement
> Reporter: Anton Kalashnikov
> Priority: Major
>
> The main case:
> * We want to restore the last unaligned checkpoint.
> * In-flight data of this checkpoint is corrupted.
> * We want to ignore this corrupted data and restore only states.
> The idea is having new configuration parameter('ignoreInFlightDataOnRecovery'
> or similar). and If it set to true, ignore the metadata of in-flight data on
> the Checkpoint Coordinator side.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
