[ 
https://issues.apache.org/jira/browse/FLINK-29500?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Dawid Wysakowicz closed FLINK-29500.
------------------------------------
    Fix Version/s: 1.17.0
                   1.15.3
                   1.16.1
         Assignee: Dawid Wysakowicz
       Resolution: Fixed

Fixed in:
* master
** cadfab59a9a5f6eaaaae7f95e2a9fb3ad7b1b1d9
* 1.16
** 6c6f152e9f3ac10e8d6993e1d9ab8076053e2a44
* 1.15
** 74bc6d2f776baad35b2f7f5be41cdfe87759192f

> InitializeOnMaster uses wrong parallelism with AdaptiveScheduler
> ----------------------------------------------------------------
>
>                 Key: FLINK-29500
>                 URL: https://issues.apache.org/jira/browse/FLINK-29500
>             Project: Flink
>          Issue Type: Bug
>          Components: API / Core, Runtime / Coordination
>    Affects Versions: 1.16.0, 1.15.2, 1.14.6
>            Reporter: Dawid Wysakowicz
>            Assignee: Dawid Wysakowicz
>            Priority: Major
>              Labels: pull-request-available
>             Fix For: 1.17.0, 1.15.3, 1.16.1
>
>
> {{InputOutputFormatVertex}} uses {{JobVertex#getParallelism}} to invoke 
> {{InitializeOnMaster#initializeGlobal}}. However, this parallelism might not 
> be the actual one which will be used to execute the node in combination with 
> Adaptive Scheduler. In case of Adaptive Scheduler the execution parallelism 
> is provided via {{VertexParallelismStore}}.



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

Reply via email to