Hi Patrick, On Thu, Aug 09, 2018 at 01:01:27AM -0400, Patrick Hemmer wrote: > There's an issue on current master (287527a) where haproxy is losing > track of its connections, and they're getting stuck in CLOSE_WAIT. And > it's counting these connections towards limits (such as maxconn). > Eventually maxconn is reached and everything starts queuing, and timing out.
Without being certain about it, it looks like it could be related to the few corner cases we've already met with the start of rework of the layering, that Olivier fixed yesterday. Your config is running here fine with 100ms response time and I haven't seen any single sQ in the logs. However I couldn't see it fail with commit 287527a either, so it doesn't mean much. I'm aware there's still one issue with this that I met this morning and which causes some connections not to terminate, so if you don't have much time to re-run a test, let's wait for our confirmation that it's fixed first. Thanks, Willy
