Am 30.08.2012 um 15:50 schrieb Ian Barber: > On Thu, Aug 30, 2012 at 5:16 AM, Michael Haberler <[email protected]> wrote: > >> test_shutdown_stress running... >> /bin/bash: line 5: 23543 Segmentation fault ${dir}$tst >> FAIL: test_shutdown_stress >> >> if I change test_shutdown_stress to start the program under gdb like so >> >> #exec "$progdir/$program" ${1+"$@"} >> gdb "$progdir/$program" ${1+"$@"} >> >> it seems to run ok: lots of >> [Thread 0xa6de7b70 (LWP 26506) exited] >> [New Thread 0xa65e6b70 (LWP 26507)] >> ... >> >> but then "Program exited normally." >> >> looks like a bit of a Heisenbug - or a race condition - to me > > Yeah, the shutdown stress test tends to find those :) I wasn't getting > that, but I'll see if I can replicate - I think Brian was saying he > was seeing the same thing on the list the other day.
Ian - in that case it's good news for me because I was just curious whether the platform has any impact, but rt-preemept seems to be bug-compatible. thanks Michael > > Ian > _______________________________________________ > zeromq-dev mailing list > [email protected] > http://lists.zeromq.org/mailman/listinfo/zeromq-dev _______________________________________________ zeromq-dev mailing list [email protected] http://lists.zeromq.org/mailman/listinfo/zeromq-dev
