oi! After a lot of looking I think I found the difference between the environments causing the issue. `py-auto-reload = 1` on my development server seems to totally break uwsgi.suspend() when a second request comes in. I have no idea how the two are related, but after removing that from my configuration everything seems to be working correctly again.
Roberto, is there anything I can do to help track down the bug further? On Fri, Mar 20, 2015 at 12:57 PM, Tim Tisdall <[email protected]> wrote: > I'm not really sure how to get any more information on this to help > with debugging, but what's happening is my worker process freezes when > it's just finished running uwsgi.suspend() and another request comes > into uwsgi. I see my logging statement just before the > uwsgi.suspend(), but after another http request comes in there's no > more messages even if there's content on that socket it's suspending > on. The next message I get after some time in the HARAKIRI one. > > What's strange is I'm running similar code on another machine and it > seems to be working fine on there. I did see similar failures, but > after it's been running for a while they seem to have gone away. > > Is there any way to get additional logging for uwsgi to see where it's > locking up? _______________________________________________ uWSGI mailing list [email protected] http://lists.unbit.it/cgi-bin/mailman/listinfo/uwsgi
