https://bz.apache.org/bugzilla/show_bug.cgi?id=65454
--- Comment #3 from Aaron Ogburn <aogb...@redhat.com> --- Thanks, Mark. I also wasn't coming up with a proper fix in the current implementation. The best thing I saw here was to adjust https://github.com/apache/tomcat/blob/9.0.x/java/org/apache/tomcat/util/threads/TaskQueue.java#L87 to the following to ensure at least an extra idle thread if enqueuing: if (parent.getSubmittedCount()<(parent.getPoolSize())) { But that'd only reduce chance and wouldn't be foolproof as we could still potentially end up with both idle threads timing out in the same problem timing. Otherwise, it was looking like re-implementing was required to fix :( -- You are receiving this mail because: You are the assignee for the bug. --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org For additional commands, e-mail: dev-h...@tomcat.apache.org
