Comments on: Multiple Threads Testing Idioms In Java http://www.codelord.net/2008/05/06/multiple-threads-testing-idioms-in-java/ A place a coder rants at... Sat, 21 Aug 2010 12:13:46 +0000 hourly 1 http://wordpress.org/?v=3.0.1 By: Heinz Kabutz http://www.codelord.net/2008/05/06/multiple-threads-testing-idioms-in-java/comment-page-1/#comment-5 Heinz Kabutz Sat, 10 May 2008 02:04:23 +0000 http://codedump.wordpress.com/?p=4#comment-5 Nice. We had similar issues with some of the exercises of our Java Specialist Master Course. There is a bug (in my opinion) in the Condition.await() method, in that it does not mirror the behavior of the Object.wait() method accurately. If you stop the thread whilst it is doing await(), the thread does not pick up the lock again. Thus, when you unlock in finally, you end up with IllegalMonitorStateExceptions. The point is, we saw the same thing go wrong. It appears to me that JUnit does not wait for the other threads to complete even, because we don't always see the output from all of these threads. Anyway, it might also be the IntelliJ plugin that causes this, I have not had the time to investigate this further. Nice. We had similar issues with some of the exercises of our Java Specialist Master Course. There is a bug (in my opinion) in the Condition.await() method, in that it does not mirror the behavior of the Object.wait() method accurately. If you stop the thread whilst it is doing await(), the thread does not pick up the lock again. Thus, when you unlock in finally, you end up with IllegalMonitorStateExceptions.

The point is, we saw the same thing go wrong. It appears to me that JUnit does not wait for the other threads to complete even, because we don’t always see the output from all of these threads.

Anyway, it might also be the IntelliJ plugin that causes this, I have not had the time to investigate this further.

]]>