You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@flink.apache.org by GitBox <gi...@apache.org> on 2022/07/11 09:37:48 UTC

[GitHub] [flink] xintongsong commented on pull request #19840: [FLINK-24713][Runtime/Coordination] Support the initial delay for SlotManager to wait fo…

xintongsong commented on PR #19840:
URL: https://github.com/apache/flink/pull/19840#issuecomment-1180179352

   @Aitozi I think the previous PR needs to be improved, but the general idea of having a `recoveryFuture` aligns with my suggestions.
   
   This future should be completed when:
   - Whenever the recovered workers become empty
   - The timeout is reached
   - For standalone resource manager, at the very beginning
   
   And all the `processResourceRequirements` (or `declareRequiredResources`) should be wrapped with `whenComplete` on this future.
   
   The implementation should be simplified with awareness of:
   1. If a completable future is completed multiple times, only the first time matters.
   2. Calling `whenComplete` on a future that is already completed will trigger the execution immediately.


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: issues-unsubscribe@flink.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org