mpi.py fails to run: invalid MPI destination

Bug #73474 reported by Alexandre Otto Strube
14
This bug affects 2 people
Affects Status Importance Assigned to Milestone
python-scientific (Ubuntu)
Expired
Medium
Unassigned

Bug Description

Binary package hint: python-scientific

python 2.4.11-1
mpichpython 2.4.11-1

updated feisty.

$ mpichpython mpi.py
Traceback (most recent call last):
  File "mpi.py", line 11, in ?
    communicator.send(data, 1, 0)
Scientific.MPI.core.MPIError: invalid MPI destination

Revision history for this message
Barry deFreese (bddebian) wrote :

I can confirm this, unfortunately I'm struggling with tracking down the problem. Thanks.

Changed in python-scientific:
status: Unconfirmed → Confirmed
Revision history for this message
Barry deFreese (bddebian) wrote :

OK, obviously this has changed significantly in Gutsy. I don't even have an mpichpython anymore. Can anyone re-confirm this? Thanks!

Changed in python-scientific:
status: Confirmed → Incomplete
Revision history for this message
Rolf Leggewie (r0lf) wrote :

Closing this bug report as no further information has been provided.
Please feel free to reopen this bug with further information if you
still experience this problem.

Thank you for your help in making ubuntu the best distro out there.

Changed in python-scientific:
assignee: nobody → r0lf
status: Incomplete → Invalid
Revision history for this message
sam tygier (samtygier) wrote :

still seeing this in hardy.
i installed python-mpi and lampython
then ran
lamboot
then
mpipython /usr/share/doc/python-mpi/examples/mpi.py

Changed in python-scientific:
status: Invalid → New
Rolf Leggewie (r0lf)
Changed in python-scientific:
assignee: r0lf → nobody
Daniel T Chen (crimsun)
Changed in python-scientific:
importance: Undecided → Medium
status: New → Confirmed
Revision history for this message
James Gray (jamespgray) wrote :

I am getting the same message, but I am using python-mpi package directly by executing:
mpiexec -n 5 python /usr/share/doc/python-mpi/examples/mpi.py

Tracked this back a little further. The traceback that I saw was from File "/usr/lib/python2.6/dist-packages/Scientific/MPI/core.py", line 303, in send. This is within the DummyCommunicator class. It indicates that the try condition above has failed. I added "import traceback" and "traceback.print_exc()" to core.py at line 284.

After running again another traceback appears just before the other:

Traceback (most recent call last):
  File "/usr/lib/python2.6/dist-packages/Scientific/MPI/core.py", line 279, in <module>
    from Scientific_mpi import *
ImportError: No module named Scientific_mpi

The python-mpi module does not contain any such module nor is there one available on my system. Perhaps instead of Scientific_mpi it should be Scientific.MPI (seperated with a period).

Traceback (most recent call last):
  File "/usr/lib/python2.6/dist-packages/Scientific/MPI/core.py", line 280, in <module>
    from Scientific.MPI import _C_API, _registerErrorObject
ImportError: cannot import name _C_API

Not sure where to go from here. Could this be the result of a missing dependency?

$ cat /etc/lsb-release
DISTRIB_ID=Ubuntu
DISTRIB_RELEASE=9.04
DISTRIB_CODENAME=jaunty
DISTRIB_DESCRIPTION="Ubuntu 9.04"

uname -m
x86_64

Revision history for this message
sam tygier (samtygier) wrote :

None of the packages mentioned in this report seem to exist in xenial/16.04, is this still an issue?

Changed in python-scientific (Ubuntu):
status: Confirmed → Incomplete
Revision history for this message
Launchpad Janitor (janitor) wrote :

[Expired for python-scientific (Ubuntu) because there has been no activity for 60 days.]

Changed in python-scientific (Ubuntu):
status: Incomplete → Expired
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.