https://bz.apache.org/bugzilla/show_bug.cgi?id=63916
--- Comment #14 from Mark Thomas <ma...@apache.org> --- I think I have a fix. I've updated the test to keep writing until the non-blocking write returns 0 bytes written and only then add the connection to the Poller. Performance is back to normal for me even with the (relatively) small buffer. Can you test this on your platforms as well? Meanwhile, I'm going to be looking at implementing a similar change in the Tomcat code. -- 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