[
https://issues.apache.org/jira/browse/FLINK-17012?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17091552#comment-17091552
]
Stephan Ewen commented on FLINK-17012:
--------------------------------------
Sounds fair. I am not super opinionated on the solution, I mainly want to avoid
is moving further away from eager initialization to a "lazy initialization by
design".
As it sounds, the two options we have would be:
1. Keep initialization as it is for now, but give the AbstractInvokable
control of when it switched to running
2. Add a factory for the AbstractInvokable and the StreamTask switches to
RUNNING just before it calls {{invoke()}} on the AbstractInvokable.
Both options look like they need some reasonable amount of refactoring.
Which one would you prefer?
> 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, Runtime / Task
> 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)
