Re: [myrinet] re: port held open and killing processes
Bob Felderman <[email protected]>
| Newsgroups | gmane.network.myrinet.general |
|---|---|
| Message-ID | <[email protected]> |
=> From the man page for rsh, => => Job control signals stop the local rsh process only; this is => arguably wrong, but currently hard to fix. => => Under linux TERM signals are sent through the rsh but none others. The => problem is that you are not allowed to have signal handling for KILL. => Makes sense. So rsh can't catch the signal and send it to the remote => process and then exit. Well, then we definitely need to be using kill 'TERM' in the mpirun.ch_gm. We're looking at your mpi_abort patch to see if we think it will work and help the problem. We know that MPI_Abort is not well supported.