I tried to run pyfr on a cluster with the openmp backend and got the following error output:
An MPI process has executed an operation involving a call to the
“fork()” system call to create a child process. Open MPI is currently
operating in a condition that could result in memory corruption or
other system errors; your MPI job may hang, crash, or produce silent
data corruption. The use of fork() (or system() or other calls that
create child processes) is strongly discouraged.
Any idea what might have caused this problem? Thank you!
On 31/01/2018 12:27, Yifan Bai wrote> I tried to ran pyfr on a cluster
with the openmp backend and got t>
following error output:
An MPI process has executed an operation involving a call to the
"fork()" system call to create a child process. Open MPI is currently
operating in a condition that could result in memory corruption or
other system errors; your MPI job may hang, crash, or produce silent
data corruption. The use of fork() (or system() or other calls that
create child processes) is strongly discouraged.
Any idea what might have caused this problem? Thank you!
PyFR goes to great lengths to ensure that it does not fork() once MPI
has been initialised. This is accomplished by using the pytools module
which includes a preforker. If you are getting this warning it is
likely that something is initialising MPI before PyFR gets a chance to
run its prefork code.
Can you give any details regarding the versions of mpi4py, h5py
(including MPI support information), and OpenMPI?