He-Pin commented on issue #1727:
URL: https://github.com/apache/pekko/issues/1727#issuecomment-4576972818

   Root-cause analysis: `DispatcherActorSpec` "respect the throughput setting" 
uses a `thread-pool-executor` with `fixed-pool-size = 1` and `throughput = 101` 
(DispatcherActorSpec.scala:32-37). On a genuine single-threaded pool the test 
is deterministic — `slowOne` processes its full 100-message batch (throughput 
101) before `fastOne` ever runs `sabotage`, so `works` stays true and the latch 
reaches 0.
   
   This report was filed against the virtual-thread PR (#1724): under the 
virtualized dispatcher the single-carrier / throughput-batching guarantee no 
longer holds, so `sabotage` can interleave and flip `works` mid-batch. That is 
the same class of issue as the virtualized-scheduler work in #2925 (dispatcher 
shutdown ordering) and #2949 (JDK21+/virtualized nightly stabilisation), not a 
test-timeout bug.
   
   A test tweak cannot fix this without destroying what the test verifies 
(atomic throughput batching). The fix belongs in the virtualized dispatcher 
honoring the throughput contract. Recommend tracking under the 
virtualized-dispatcher effort rather than a test-only change.


-- 
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]


---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to