divijvaidya commented on PR #12666:
URL: https://github.com/apache/kafka/pull/12666#issuecomment-1253460398

   This is a good idea and definitely an improvement over what already exists. 
But I think that the downside of taking a dependency on something that is 
deprecated as soon as JDK 18 is quite significant. Perhaps, we can solve the 
underlying problem in a different way!
   
   Let me explain my point of view on this.
   
   Some of our tests spawn new threads. The expectation for the happy case path 
is that these threads are terminated by the test gracefully at the end of test 
execution. In case when these threads are not terminated, they may cause 
erroneous/undeterministic behaviour such as calling System.Exit().
   
   Instead of focusing on Exit() (which is not the underlying cause, it's a 
side effect), could we try to formulate a solution which ensures that the test 
would fail if threads created by that test are not terminated at the end of the 
execution? 
   
   I don't have a proposal towards a solution but there are ideas that we could 
explore such as ensuring all threads created by a test have a property in their 
thread local (this could be set in @before and removed in @after).
   
   Thoughts?


-- 
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: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to