[zeromq-dev] weird bug

Lucas Hope lucas.r.hope at gmail.com
Mon Mar 5 00:24:34 CET 2012

As far as I know, this is actually by design, not a bug. Multithreaded apps
just interrupt the send and recv when threads are waiting on them. This
used to be handled internally by zmq, but now is not.

For my own apps, I wrote a simple handler to re-call the recv if EINTR is
signalled. There's also a way to have this done automagically, but I am not
an authority on that.

This bit me exactly the same when I upgraded recently from 2.1.7 to 2.1.11.



On Mon, Mar 5, 2012 at 10:04 AM, Andrew Hume <andrew at research.att.com>wrote:

> i am hoping this is a well-known bug with a known work-around.
> i have been (more or less) happily running with 2.1.7, but for various
> reasons,
> i wanted to try a more recent version. so i am running with 2.1.10 and am
> getting core dumps aplenty. with code that looks like
> ret = zmq_recv(sock, &msg, 0);
> assert(ret == 0);
> the assert fails with an error of EINTR.
> what is going on here?
> the programs are always multithreaded, but i follow the rules on contexts
> etc.
> andrew
> ------------------
> Andrew Hume  (best -> Telework) +1 623-551-2845
> andrew at research.att.com  (Work) +1 973-236-2014
> AT&T Labs - Research; member of USENIX and LOPSA
> _______________________________________________
> zeromq-dev mailing list
> zeromq-dev at lists.zeromq.org
> http://lists.zeromq.org/mailman/listinfo/zeromq-dev

Dr Lucas Hope - lucas.r.hope at skype
Machine Learning and Software Engineering Consultant
Melbourne, Australia
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.zeromq.org/pipermail/zeromq-dev/attachments/20120305/6bc88157/attachment.htm>

More information about the zeromq-dev mailing list