[
https://issues.apache.org/jira/browse/FLINK-5601?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16671781#comment-16671781
]
Jiayi Liao commented on FLINK-5601:
-----------------------------------
[~aljoscha]
Thanks for pointing it out. I've done this on my own business and it works as
what I expect.
To avoid the problem of compatibiliy, perhaps we can add a V2 version of
watermarks operator or making the checkpointing optional by adding parameters
into the constructor method. But I'm not sure if this is appropriate.
Tell me if you have better ideas and I'll submit the PR according to your
advice.
> Window operator does not checkpoint watermarks
> ----------------------------------------------
>
> Key: FLINK-5601
> URL: https://issues.apache.org/jira/browse/FLINK-5601
> Project: Flink
> Issue Type: Improvement
> Components: State Backends, Checkpointing
> Affects Versions: 1.5.0, 1.6.0, 1.7.0
> Reporter: Ufuk Celebi
> Assignee: Jiayi Liao
> Priority: Critical
> Fix For: 1.8.0
>
>
> During release testing [[email protected]] and I noticed that
> watermarks are not checkpointed in the window operator.
> This can lead to non determinism when restoring checkpoints. I was running an
> adjusted {{SessionWindowITCase}} via Kafka for testing migration and
> rescaling and ran into failures, because the data generator required
> determinisitic behaviour.
> What happened was that on restore it could happen that late elements were not
> dropped, because the watermarks needed to be re-established after restore
> first.
> [~aljoscha] Do you know whether there is a special reason for explicitly not
> checkpointing watermarks?
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
