Page 1 of 1

mpirun Hangup

Posted: Tue Jul 18, 2017 9:05 am
by unnikartha
I get the following error while running a parallel job

--------------------------------------------------------------------------
mpirun noticed that process rank 41 with PID 19713 on node node5 exited on signal 1 (Hangup).
--------------------------------------------------------------------------
21 total processes killed (some possibly by mpirun during cleanup)
--------------------------------------------------------------------------

how can this situation be avoided?

Re: mpirun Hangup

Posted: Wed Aug 02, 2017 3:26 pm
by fmk
try running fewer jobs .. one of the processes died suddenly, could be something in your script