This is most probably due to NOX not rejecting send() calls which results in a buffer overflow and the connection I/O becoming out of sync (due to partial write to the send buffer). That could be fixed by refusing a send() when there is no buffer space available. But the more important question is why the send buffer is not drained fast enough. I'll take a look and try to fix it asap but need help to reproduce the bug. Let's continue the discussion on the github page (issue #5).
Amin On Mon, Apr 1, 2013 at 6:24 AM, Rafael Bittencourt <[email protected]> wrote: > I'm using Cbench in throughput mode but after a few seconds of running > Cbench I get the following error messages: > 146439|event-dispatcher|ERR:Event ofp_packet_in processing leaked an > exception: boost::asio::streambuf too long > 146440|event-dispatcher|ERR:Extra information: Throw location unknown > (consider using BOOST_THROW_EXCEPTION) > Dynamic exception > type:boost::exception_detail::clone_impl<boost::exception_detail::error_info_injector<std::length_error> >> > std::exception::what: boost::asio::streambuf too long > > and i got poor results!
