Re: MPI problem?

[Andreas Härtel]

Hi Steffen,

Calling pypresso with just the print statement does not produce the 
warning. If I import espressomd, the warning occurs.

Calling "mpiexec pypresso script" with just the printing statement does 
print Hello several times; importing espressomd produces an error:
(...)
Primary job  terminated normally, but 1 process returned
a non-zero exit code.. Per user-direction, the job has been aborted.
-------------------------------------------------------
--------------------------------------------------------------------------
mpiexec detected that one or more processes exited with non-zero status, 
thus causing the job to be terminated. The first process to do so was:
(...)


If I run the following using "mpiexec pypresso":
from mpi4py import MPI
comm = MPI.COMM_WORLD
rank = comm.Get_rank()
if rank != 0:
 exit(0)
print("Hallo")
I find only one "Hello".



Best regards,
Andreas



On 27.02.20 10:16, Steffen Hirschmann wrote:
> Hi Andreas,
>
> does this warning also occur if you don't "import espressomd" and just
> mpiexec pypresso with that print statement or an empty script?
>
> Greetings,
> Steffen
>
> On 14:38 Wed 26 Feb     , Andreas Härtel wrote:
> > I have to add a note:
> > adding particles is extremely slow: 13000 particles can take hours.
> > Maybe - hopefully - this is related ...
> >
> > Thanks and best regards,
> > Andreas
> >
> >
> >
> > On 26.02.20 10:32, Steffen Hirschmann wrote:
> > > Hi Andreas,
> > >
> > > On 14:40 Tue 25 Feb     , Andreas Härtel wrote:
> > > > [...]
> > > > An MPI process has executed an operation involving a call to the
> > > > "fork()" system call to create a child process.
> > > > [...]
> > > do you manually start a new process from the Python simulation script?
> > > E.g. using os.system(), the modules subprocess or multiprocessing?
> > > It would be beneficial if you could post the simulation script to the
> > > mailing list.
> > >
> > > Greetings,
> > > Steffen
> >
> > --
> > Dr. Andreas Härtel
> > Albert-Ludwigs-Universität Freiburg (University of Freiburg)
> > SOΦT - Statistical Physics of Soft Matter and Complex Systems
> > Physikalisches Institut (Institute of Physics)
> > Hermann-Herder-Straße 3, Hochhaus
> > Room 407
> > 79104 Freiburg
> > Germany
> > Phone: +49 761 203-97783
> > Email: address@hidden
> > Web page: http://andreashaertel.anno1982.de/

--
Dr. Andreas Härtel
Albert-Ludwigs-Universität Freiburg (University of Freiburg)
SOΦT - Statistical Physics of Soft Matter and Complex Systems
Physikalisches Institut (Institute of Physics)
Hermann-Herder-Straße 3, Hochhaus
Room 407
79104 Freiburg
Germany
Phone: +49 761 203-97783
Email: address@hidden
Web page: http://andreashaertel.anno1982.de/