> On Fri, Apr 27, 2018 at 2:46 PM, INADA Naoki <songofaca...@gmail.com> > wrote: > >> On Fri, Apr 27, 2018 at 8:42 PM Руслан Закиров <r...@sports.ru> wrote: >> >> > On Fri, Apr 27, 2018 at 10:36 AM, INADA Naoki <songofaca...@gmail.com> >> wrote: >> >> >> https://uwsgi-docs.readthedocs.io/en/latest/Options.html#listen >> >> https://serverfault.com/questions/271380/how-can-i-increase-the-value-of- >> somaxconn >> >> >> > Guys, >> >> > I know how to change size of the listen queue. >> >> > Why do you think I should change it? Do you suggest me to increase it >> or >> decrease? >> >> >> Because the log say "uWSGI listen queue of socket "0.0.0.0:5002" (fd: >> 10) >> full !!! (101/100)". >> >> You used ab -c 200. 200 connections are created at once. 100 listen >> queue >> size is not enough. >> >> I'm curious why you didn't think you should change listen queue size.... >> > > Because it's not right thing to do in production. > > However, I agree that 100 is less than 200 and queue will be overflown if > ab > generates requests faster than the app processes them. Sure ab fires > things > quite fast. > > I oversimplified my production case. In the presented case it's 1.5ms > between accept() > calls. In production requests I see 600ms gaps between accept() calls. > This > is what > brought me here to the list. > > Probably I have to dig more into production straces... > > What do you think? Ideas? > > -
Hi, if i understand correctly you are testing a production code-base so isolating the problem would be very hard. Can you try with a test with a simple hello world ? (just to ensure coroae is working right) -- Roberto De Ioris http://unbit.com _______________________________________________ uWSGI mailing list uWSGI@lists.unbit.it http://lists.unbit.it/cgi-bin/mailman/listinfo/uwsgi