https://bz.apache.org/bugzilla/show_bug.cgi?id=64202
--- Comment #16 from Remy Maucherat <r...@apache.org> --- Thanks ! So annoying ... I was 100% certain the problem was some thread safety problem, so since I couldn't reproduce I got nowhere. I'll revert the change to use a latch since it is likely useless. No problem for write, it already loops since it must write the whole buffer. I probably won't make a change to keep track of the overall timeout at this time, the timeout is the IO timeout and something did happen. So this would be a deviation from the 9 behavior but it's not necessarily a bad thing. Up for discussion I guess ... If the timeout becomes the overall timeout then the write must be modified as well. -- 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
