On Fri, Apr 20, 2018 at 09:01:24PM +0200, Willy Tarreau wrote: > On Fri, Apr 20, 2018 at 08:59:32PM +0300, Slawa Olhovchenkov wrote: > > Ok, I am got respons from thread 3 and thread 7. > > And never the other ones ? That's kind of strange!
I am stop after got distinct threads w/ different CPU load. 3 3 7 7 2 7 6 3 3 6 6 7 1 1 3 7 1 3 3 5 6 3 6 3 > > Can I mean this threads mapped to CPU 3 and 7? > > No, this is totally unrelated, unless of course you're using cpu-map to > map thread 1 to cpu 1, thread 2 to cpu 2 etc... I am use cpu-map auto:1/1-8 0-7 CPU 2 busy 0.09 CPU 6 busy 0.32 > > CPU 3 busy at 0.42, CPU 7 busy at 0.17, show activity is about same: > > > > thread_id: 3 > > date_now: 1524245903.896442 > > loops: 119914128 184812851 173923683 168285397 165260657 155576226 > > 167823891 142903166 > > wake_cache: 94411535 148233540 144512131 124600418 127222467 121950507 > > 127136520 112878590 > > wake_tasks: 58100 101310 22429 290349 154221 99918 207275 77221 > > wake_applets: 0 0 0 0 0 0 0 0 > > wake_signal: 0 0 0 0 0 0 0 0 > > poll_exp: 94469635 148334850 144534560 124890767 127376688 122050425 > > 127343795 112955811 > > poll_drop: 1238561 1910768 636228 4940166 2745656 1892360 3545183 1547972 > > poll_dead: 0 0 0 0 0 0 0 0 > > poll_skip: 0 0 0 0 0 0 0 0 > > fd_skip: 299688784 399135394 429530804 248537214 327557232 340203944 > > 300964717 331714706 > > fd_lock: 27928376 42948659 46096172 22460864 31946577 34536787 28610880 > > 33200161 > > fd_del: 6494887 10577819 3337682 37986875 16346727 10378443 23322889 8187644 > > conn_dead: 0 0 0 0 0 0 0 0 > > stream: 18389406 28258935 9395022 73963729 40378554 27746726 52452679 > > 22637467 > > empty_rq: 102497936 158696266 164717166 111903546 129615612 130001228 > > 123814511 121700524 > > long_rq: 0 0 0 0 0 0 0 0 > > > > thread_id: 7 > > date_now: 1524245909.919941 > > loops: 119974591 184907272 174012444 168371911 165345936 155659245 > > 167910558 142976547 > > wake_cache: 94459348 148309188 144585673 124664378 127288213 122015537 > > 127202211 112936359 > > wake_tasks: 58123 101357 22438 290505 154304 99961 207379 77257 > > wake_applets: 0 0 0 0 0 0 0 1 > > wake_signal: 0 0 0 0 0 0 0 0 > > poll_exp: 94517471 148410545 144608111 124954883 127442517 122115498 > > 127409590 113013617 > > poll_drop: 1239145 1911817 636564 4942753 2747081 1893376 3546921 1548730 > > poll_dead: 0 0 0 0 0 0 0 0 > > poll_skip: 0 0 0 0 0 0 0 0 > > fd_skip: 299841823 399338153 429745993 248661716 327726077 340384467 > > 301118348 331883440 > > fd_lock: 27943327 42970960 46120113 22472636 31963565 34555176 28626056 > > 33217139 > > fd_del: 6497739 10583157 3339476 38006597 16354809 10383821 23334628 8191852 > > conn_dead: 0 0 0 0 0 0 0 0 > > stream: 18397654 28273212 9399987 74002315 40398851 27761908 52479235 > > 22649415 > > empty_rq: 102550508 158777515 164801058 111960753 129682899 130070277 > > 123878819 121762731 > > long_rq: 0 0 0 0 0 0 0 0 > > So between these two the traffic is perfectly balanced. I don't see why you > wouldn't have other threads :-/ > > Well, let's try something. In your global section you have a "stats socket > ..." > with a path. Please replicate this line to have 8 more of them, all with a > different path (eg: /var/run/haproxy-stats becomes haproxy-stats1, -stats2 > etc) > and with a different thread binding using "process" like this : > > stats socket /var/run/haproxy-stats1 level admin process 1/1 > stats socket /var/run/haproxy-stats2 level admin process 1/2 > stats socket /var/run/haproxy-stats3 level admin process 1/3 > ... > stats socket /var/run/haproxy-stats8 level admin process 1/8 > > Then you can use the same command as above to dump exactly one thread's > activity by connecting to each of these sockets. We definitely need to > understand what the other threads are doing. Ok, I am need run haproxy some time for data collected.
