zhuzhurk commented on issue #10048: [FLINK-14462][coordination] Remove 
JobGraph#allowQueuedScheduling flag because it is always true
URL: https://github.com/apache/flink/pull/10048#issuecomment-548226349
 
 
   > @zhuzhurk I think it is a test scope problem that we now waiting for a 
long time before execution graph fail global. I'm trying to set allocation 
timeout and rpc timeout to 0. However, due to asynchronous natural it is 
unstable. Do you have an idea how we port such fail immediately tests to 
current wait "forever" case?
   
   If it is to test the SchedulerImp/SlotPoolImp itself, I have no good common 
idea but to expect the allocation to be not completed within the given timeout. 
Or maybe to fix them case by case.
   If it is to test other things that depend on SlotProvider/SlotOwner, you can 
use `SimpleSlotProvider` which throws error if the slot request cannot be 
fulfilled 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.
 
For queries about this service, please contact Infrastructure at:
[email protected]


With regards,
Apache Git Services

Reply via email to