You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@flink.apache.org by "Chesnay Schepler (Jira)" <ji...@apache.org> on 2020/04/07 08:05:00 UTC

[jira] [Commented] (FLINK-17012) Expose stage of task initialization

    [ https://issues.apache.org/jira/browse/FLINK-17012?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17077024#comment-17077024 ] 

Chesnay Schepler commented on FLINK-17012:
------------------------------------------

Would this _need_ to be a metric or would a log message suffice?

So far we somewhat stayed away from metrics which don't change, because after you have evaluated it once (after it has normalized) you have no more use for it and it is just overhead.

> Expose stage of task initialization
> -----------------------------------
>
>                 Key: FLINK-17012
>                 URL: https://issues.apache.org/jira/browse/FLINK-17012
>             Project: Flink
>          Issue Type: Improvement
>          Components: Runtime / Metrics
>            Reporter: Wenlong Lyu
>            Priority: Major
>
> Currently a task switches to running before fully initialized, does not take state initialization and operator initialization(#open ) in to account, which may take long time to finish. As a result, there would be a weird phenomenon that all tasks are running but throughput is 0. 
> I think it could be good if we can expose the initialization stage of tasks. What to you think?



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