On Fri, Oct 09, 2015 at 11:55:17AM -0700, Benjamin Mahler <benjamin.mah...@gmail.com> wrote: > First of all, thanks for the great library, the Mesos project has been > using heavily in production for many years. Recently, we've encountered a > rare issue in which SIGPIPE occurs during a call to ev_async_send.
Thanks for you great analysis, and while I haven't exatcly followed it yet, it looks correct. I'll see how it can be fixed (most likely, by not recreeating the pipe in this case atg all). I do hope its rare, as this is essentially an epoll bug workaround, which is extremely costly. One question though: is there a specific reason why you go against the recommendation to ignore SIGPIPE? SIGPIPE is essentially for shell filters only. > This seems to confirm the issue! With eventfd enabled the write will > silently fail rather than trigger SIGPIPE. This also seems problematic, yes? Fortunately no, because events can get lost normally under this condition, so libev already assumes there might have been an event and checks. So ignoring SIGPIPE should work reliably as a workaround. -- The choice of a Deliantra, the free code+content MORPG -----==- _GNU_ http://www.deliantra.net ----==-- _ generation ---==---(_)__ __ ____ __ Marc Lehmann --==---/ / _ \/ // /\ \/ / schm...@schmorp.de -=====/_/_//_/\_,_/ /_/\_\ _______________________________________________ libev mailing list libev@lists.schmorp.de http://lists.schmorp.de/mailman/listinfo/libev