Experiencing an issue with ScheduledExecutorService alongside VT

[Alan Bateman]

On 28/07/2024 12:41, Yuval Lombard wrote:
> Hi Team,
>
> For the sake of accuracy and integrity, I have noticed the 
> previous thread dump was taken from a piece of code that does not 
> match the provided example I attached in the first couple of mails I 
> have sent.
> I have attached a new thread dump file that was taken from the exact 
> version of the code I was testing.
>
> Will appreciate it if you could take a look again just to make sure 
> you see the issue ( related to preemption when cancelling a timer 
> after Object.wait(millis) ) you found in the previous dump I attached.

Looks to me the same thing.

In this dump, #302878 and #302879 are scheduling a timeout as part of a 
timed-Socket.connect. ForkJoinPool-1-worker-2 and 
ForkJoinPool-1-worker-3 are blocked trying to schedule a timeout. The 
interesting one is #302780 as it seems to have been preempted while 
blocking as it removes a cancelled timeout.

This issue is fixed but the EA download is a bit behind and we need to 
get it refreshed. Also I think you said something about the version of 
Spring needing to updated to work with an EA build (as the EA builds are 
tracking main line for JDK 24).

-Alan