On Tue, Mar 07, 2017 at 10:14:42AM +0100, Matthias Fechner wrote: > Am 2017-03-07 00:32, schrieb Willy Tarreau: > > patch -Rp1 < > > 0001-BUG-MEDIUM-tcp-don-t-poll-for-write-when-connect-suc.patch > > > > I've just tested here on 1.7.3 and it does apply correctly. > > > > With git apply you'll have to pass -R as well. > > Sorry for not being clear the first time. > > so, shortly after I started to 1.7.3 with the patch reverted I got timeouts > again.
Wow we cross-posted. > If I do a netstat -an I see a lot of: > tcp4 0 0 127.0.0.1.443 127.0.0.1.47010 TIME_WAIT > tcp4 0 0 127.0.0.1.443 127.0.0.1.46961 CLOSE_WAIT > tcp4 0 0 127.0.0.1.46961 127.0.0.1.443 FIN_WAIT_2 Is :443 your haproxy or the server after haproxy ? If it's the server behind haproxy it means this one didn't respond to a close. If it's haproxy it means it's stuck doing something else (very bad). > Can I do anything to help you to track the problem to the source? Yes, please test with "nokqueue" in the global section, or start haproxy with "-dk". It will switch to poll() and will tell us if there's a bug in the kqueue poller. Please be aware that your CPU usage will increase a bit. Thanks, Willy
