[
https://issues.apache.org/jira/browse/FLINK-17012?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17091611#comment-17091611
]
Piotr Nowojski edited comment on FLINK-17012 at 4/24/20, 2:17 PM:
------------------------------------------------------------------
Solution 1. is not a proper solution, but a quick fix/dirty hack for me, so I
would strongly prefer the option 2., as it would also clean up other things in
our code and in the long run we will need to go in this direction anyway. But I
could accept 1. if it's really simple and not complicating existing code too
much, but it will be ultimately a wasted effort :( (at least as I see it).
was (Author: pnowojski):
Solution 1. is not a proper solution, but a quick fix/dirty hack for me, so I
would strongly prefer the option 2., as it would also clean up other things in
our code and in the long run we will need to go in this direction anyway. But I
could accept 1. if it's really simple and not complicating existing code too
much, but it will be ultimately a wasted effort.
> 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)
