Original comment by Bitbucket user (Bitbucket: wangh0a, GitHub: wangh0a).
Hi Aurelien,
Do you have a solution for it?
Right now, I’m replacing MPI_Abort(…) with kill(getppid(), SIGTERM) to kill the 'orted' process. However, it might not clean up all the mpi processes, right?
Original comment by Bitbucket user (Bitbucket: wangh0a, GitHub: wangh0a).
Hi Aurelien,
Thank you for the prompt reply. Do you think there would be cases where dependent processes not being killed (result to defunct processes) even if the parent daemon is killed?
Original report by Aurelien Bouteiller (Bitbucket: abouteiller, GitHub: abouteiller).
After a fault, MPI_Abort does not kill 'orted' daemons and the
mpirun
remains stuck