[
https://issues.apache.org/jira/browse/FLINK-21133?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17290206#comment-17290206
]
Piotr Nowojski commented on FLINK-21133:
----------------------------------------
I agree about that we might need to change the API somehow, but maybe there are
less invasive changes. Calling `notifyCheckpointComplete()` after closing will
be very invasive/affecting silently a lot of built in and user
functions/operators. I would hope there are other options. For example IMO
renaming `close()` to `flush()` or `finish()` combined with a contract that
`notifyCheckpointComplete()` (and other checkpointing calls) can be called
between `finish()` and `dispose()` would be better. It would be just as
invasive, but as at least user code would not brake silently. But keep in mind
that at the moment on the user functions level we even don't have `dispose()`
:/ Let's maybe move this particular discussion to FLIP-147 discussion thread?
> FLIP-27 Source does not work with synchronous savepoint
> -------------------------------------------------------
>
> Key: FLINK-21133
> URL: https://issues.apache.org/jira/browse/FLINK-21133
> Project: Flink
> Issue Type: Bug
> Components: API / Core, API / DataStream, Runtime / Checkpointing
> Affects Versions: 1.11.3, 1.12.1
> Reporter: Kezhu Wang
> Priority: Critical
> Fix For: 1.11.4, 1.13.0, 1.12.3
>
>
> I have pushed branch
> [synchronous-savepoint-conflict-with-bounded-end-input-case|https://github.com/kezhuw/flink/commits/synchronous-savepoint-conflict-with-bounded-end-input-case]
> in my repository. {{SavepointITCase.testStopSavepointWithFlip27Source}}
> failed due to timeout.
> See also FLINK-21132 and
> [apache/iceberg#2033|https://github.com/apache/iceberg/issues/2033]..
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
