[jetty-dev] [jira] (JETTY-1527) QueuedThreadPool Anomaly
"Jan Bartel (JIRA)" <[email protected]> Mon, 2 Jul 2012 13:45:21 -0500 (CDT)
| Newsgroups | gmane.comp.java.jetty.general |
|---|---|
| Message-ID | <28219179.19585.1341254721138.JavaMail.j2ee-jira@codehaus01.managed.contegix.com> |
Jan Bartel commented on JETTY-1527 QueuedThreadPool Anomaly Hi David, Thanks for providing the test case. Looking at it, the QueuedThreadPool no doubt has a few funnies in it - for example getting the jobs size and the number of idle threads aren't synchronized in any way. Probably moving those 2 lines down to after the call to insert a new job in the queue changes the timing a bit and your test happens to work. I can also get it to work if I insert a little pause after each call to pool.execute() in the test. We're a bit reluctant to make changes to the QueuedThreadPool class at this point, as its a highly concurrent class and changes could introduce all sorts of subtle effects, and additionally, for jetty-9, we're probably ditching it in favour of a disruptor style mechanism. thanks Jan This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators . For more information on JIRA, see: http://www.atlassian.com/software/jira --------------------------------------------------------------------- To unsubscribe from this list, please visit: http://xircles.codehaus.org/manage_email