On May 30, 2013, at 8:55 AM, Victor Vysotskiy <victor.vysotskiy_at_[hidden]> wrote:
> Hi Ralph,
>> -mca orte_abort_non_zero_exit 0
> Thank you for the hint. That it is exactly what I need! BTW, does it help if one of the working node occasionally dies during the MPMD run?
I'm afraid not - failure of a node is a terminating condition. There has been work done on running thru such conditions IF the application isn't using MPI, but I don't think that work has been fully ported to the 1.7 or trunk yet. Hopefully not too far in the future.
> With best regards,
> users mailing list