[jira] [Updated] (FLINK-21165) Timestamps/Watermarks chained erroneously with previous operator

["Flink Jira Bot (Jira)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/FLINK-21165?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Flink Jira Bot updated FLINK-21165:
-----------------------------------
    Labels: auto-deprioritized-major stale-minor  (was: auto-deprioritized-major)

I am the [Flink Jira Bot|https://github.com/apache/flink-jira-bot/] and I help the community manage its development. I see this issues has been marked as Minor but is unassigned and neither itself nor its Sub-Tasks have been updated for 180 days. I have gone ahead and marked it "stale-minor". If this ticket is still Minor, please either assign yourself or give an update. Afterwards, please remove the label or in 7 days the issue will be deprioritized.


> Timestamps/Watermarks chained erroneously with previous operator
> ----------------------------------------------------------------
>
>                 Key: FLINK-21165
>                 URL: https://issues.apache.org/jira/browse/FLINK-21165
>             Project: Flink
>          Issue Type: Bug
>          Components: API / DataStream
>    Affects Versions: 1.11.1, 1.12.0
>            Reporter: zlzhang0122
>            Priority: Minor
>              Labels: auto-deprioritized-major, stale-minor
>         Attachments: Lark20210127-203541.png
>
>
> When using assignTimestampAndWatermarks function separately, if it can satisfied the chaining condition, it will chained erroneously with previous operator.
> For example, I have a code like bellow: 
> {code:java}
> // code placeholder
> DataStream<UserAction> ds = dataSource.map(xxx).setParallelism(1);
> ds.assignTimestampAndWatermarks(xxxx);
> DataStream<UserAction> rsStream = ds.keyBy(xxx).timeWindow(xxx).sum(xxx).setParallelism(2);{code}
> I will get the JobGraph bellow.
> !Lark20210127-203541.png!
> If I change my code to this:
> {code:java}
> // code placeholder
> DataStream<UserAction> ds = dataSource.map(xxx).setParallelism(1); 
> DataStream<UserAction> watermarkStream = ds.assignTimestampAndWatermarks(xxxx); 
> DataStream<UserAction> rsStream = watermarkStream.keyBy(xxx).timeWindow(xxx).sum(xxx).setParallelism(2);
> {code}
> I will get the same JobGraph, at the same time the actual execution result is not the same.
> I think the first JobGraph have some problem and should not show like that, maybe the Timestamps/Watermarks operator should show separately and should not chaining with previous operator.



--
This message was sent by Atlassian Jira
(v8.3.4#803005)