You are viewing a plain text version of this content. The canonical link for it is here.

Posted to issues@flink.apache.org by "Flavio Pompermaier (JIRA)" <ji...@apache.org> on 2019/02/05 11:29:00 UTC

[jira] [Commented] (FLINK-11525) Add option to manually set job ID

    [ https://issues.apache.org/jira/browse/FLINK-11525?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16760707#comment-16760707 ] 

Flavio Pompermaier commented on FLINK-11525:
--------------------------------------------

Do you think it could be possible to provide a way to fix also https://issues.apache.org/jira/browse/FLINK-10879?

It will be super useful have a way to define a macro-job that spawns other jobs (but having a unique id to monitor the job progress)?

> Add option to manually set job ID 
> ----------------------------------
>
>                 Key: FLINK-11525
>                 URL: https://issues.apache.org/jira/browse/FLINK-11525
>             Project: Flink
>          Issue Type: New Feature
>          Components: Client, REST
>    Affects Versions: 1.7.0
>            Reporter: Ufuk Celebi
>            Priority: Minor
>
> When submitting Flink jobs programmatically it is desirable to have the option to manually set the job ID in order to have idempotent job submissions. This simplifies failure handling on the user side as duplicate submissions will be rejected by Flink. In general allowing to manually set the job ID can be beneficial for third party tooling.
> The default behavior should not be altered. The following job submission entry points should be extended to allow to specify this option:
> 1. REST API
> 2. StandaloneJobClusterEntrypoint
> 3. CLI
> Note that for 2. FLINK-10921 already suggested to allow to configure the job ID manually.
> If there is agreement to have this feature, I'll go ahead and create sub tasks for the mentioned entry points (and provide the implementation for 1. and 2.).



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)