Hi, which version of mpi4py are you using? We had problems with mpi4py 3.0.0 above a certain number of MPI jobs, which varied between machines. Using mpi4py version 2.0.0 magically solved the MPI problems. Best regards, Philipp Am 10.09.19 um 17:04 schrieb Nathan:
I also don’t see the error Mpi4py is generating, just the traceback. The error message might be helpful.
On Tue, Sep 10, 2019 at 8:28 AM Prateek Gupta <prateekgidolia@gmail.com <mailto:prateekgidolia@gmail.com>> wrote:
Hi Sam,
I tried likely the same code just now with my ENZO dataset with the normal run and also parallel with MPI, both work fine for me. Which version of yt you are using. And did you tried the same code without a parallel run?
On Tue, Sep 10, 2019 at 4:20 PM Matthew Turk <matthewturk@gmail.com <mailto:matthewturk@gmail.com>> wrote:
Hi Sam,
Before digging any deeper, are you in fact running in parallel with MPI?
On Tue, Sep 10, 2019 at 8:11 AM Sam Patrick via yt-users <yt-users@python.org <mailto:yt-users@python.org>> wrote: > > Hi all, > > I've been encountering an error recently whenever I try to save a slice > plot, such as: > > data_sets = yt.load("DD*/output_[0-9][0-9][0-9][0-9]") > for ds in data_sets: > filename = > "density_t={:.3f}_ho.png".format(ds.current_time.in_units('Myr').value) > val, loc = ds.find_max('density') > print("{:.3f}".format(val)) > plt = yt.SlicePlot(ds, L, field, center= loc, width= (size, 'pc')) > plt.save(filename) > > YT seems to think issue is with the call to 'find_max' but a little > experimentation shows this is not the case. I'm unable to work out any > of what comes back to me asides that: > > File "plot.py", line 67, in <module> > val, loc = ds.find_max('density') > File "/opt/apps/pkgs/yt/3.5.1/src/yt/data_objects/static_output.py", > line 826, in find_max > source.quantities.max_location(field) > File > "/opt/apps/pkgs/yt/3.5.1/src/yt/data_objects/derived_quantities.py", > line 635, in __call__ > rv = super(MaxLocation, self).__call__(field, sample_fields) > File > "/opt/apps/pkgs/yt/3.5.1/src/yt/data_objects/derived_quantities.py", > line 591, in __call__ > rv = super(SampleAtMaxFieldValues, self).__call__(field, sample_fields) > File > "/opt/apps/pkgs/yt/3.5.1/src/yt/data_objects/derived_quantities.py", > line 68, in __call__ > for sto, ds in parallel_objects(chunks, -1, storage = storage): > File > "/opt/apps/pkgs/yt/3.5.1/src/yt/utilities/parallel_tools/parallel_analysis_interface.py", > line 520, in parallel_objects > to_share, datatype = 'dict', op = 'join') > File > "/opt/apps/pkgs/yt/3.5.1/src/yt/utilities/parallel_tools/parallel_analysis_interface.py", > line 284, in passage > return func(self, *args, **kwargs) > File > "/opt/apps/pkgs/yt/3.5.1/src/yt/utilities/parallel_tools/parallel_analysis_interface.py", > line 723, in par_combine_object > data.update(self.comm.recv(source=i, tag=0)) > File "mpi4py/MPI/Comm.pyx", line 1173, in mpi4py.MPI.Comm.recv > File "mpi4py/MPI/msgpickle.pxi", line 302, in mpi4py.MPI.PyMPI_recv > File "mpi4py/MPI/msgpickle.pxi", line 261, in mpi4py.MPI.PyMPI_recv_match > > Many thanks, > Sam Patrick > _______________________________________________ > yt-users mailing list -- yt-users@python.org <mailto:yt-users@python.org> > To unsubscribe send an email to yt-users-leave@python.org <mailto:yt-users-leave@python.org> _______________________________________________ yt-users mailing list -- yt-users@python.org <mailto:yt-users@python.org> To unsubscribe send an email to yt-users-leave@python.org <mailto:yt-users-leave@python.org>
_______________________________________________ yt-users mailing list -- yt-users@python.org <mailto:yt-users@python.org> To unsubscribe send an email to yt-users-leave@python.org <mailto:yt-users-leave@python.org>
_______________________________________________ yt-users mailing list -- yt-users@python.org To unsubscribe send an email to yt-users-leave@python.org