Import dolfin fail, with mpi error

Legacy DOLFIN
1

Hi,
I tried to import dolfin in Ubuntu 20.04. However, “import dolfin” incurs the following error:

--------------------------------------------------------------------------
Sorry!  You were supposed to get help about:
    opal_init:startup:internal-failure
But I couldn't open the help file:
    /home/conda/feedstock_root/build_artifacts/openmpi-mpi_1679555231648/_h_env_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold/share/openmpi/help-opal-runtime.txt: No such file or directory.  Sorry!
--------------------------------------------------------------------------
--------------------------------------------------------------------------
Sorry!  You were supposed to get help about:
    orte_init:startup:internal-failure
But I couldn't open the help file:
    /home/conda/feedstock_root/build_artifacts/openmpi-mpi_1679555231648/_h_env_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold/share/openmpi/help-orte-runtime: No such file or directory.  Sorry!
--------------------------------------------------------------------------
--------------------------------------------------------------------------
Sorry!  You were supposed to get help about:
    mpi_init:startup:internal-failure
But I couldn't open the help file:
    /home/conda/feedstock_root/build_artifacts/openmpi-mpi_1679555231648/_h_env_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold/share/openmpi/help-mpi-runtime.txt: No such file or directory.  Sorry!
--------------------------------------------------------------------------
*** An error occurred in MPI_Init_thread
*** on a NULL communicator
*** MPI_ERRORS_ARE_FATAL (processes in this communicator will now abort,
***    and potentially your MPI job)
[Turing:2064478] Local abort before MPI_INIT completed completed successfully, but am not able to aggregate error messages, and not able to guarantee that all other processes were killed!

The packages I have are:

# Name                    Version                   Build  Channel
fenics                    2019.1.0        py38h375c83e_37    conda-forge
fenics-dijitso            2019.1.0        py38h578d9bd_37    conda-forge
fenics-dolfin             2019.1.0        py38hc6db33b_37    conda-forge
fenics-ffc                2019.1.0        py38h578d9bd_37    conda-forge
fenics-fiat               2019.1.0        py38h578d9bd_37    conda-forge
fenics-libdolfin          2019.1.0            h789a71a_37    conda-forge
fenics-ufl                2019.1.0        py38h578d9bd_37    conda-forg

It seems the failure reason is that MPI is not working. How can I fix it? Can I disable the mpi?

2

Never post error messages or code as an image.
Just copy and paste the error in a code block, starting and ending with “```”.

cc @minrk for the actual conda error.

3

Please do not open several new posts (in particular Import dolfin shows "Could not find DOLFIN pkg-config file" ) at once.
I’ll just assume that you have sorted this one out, and close it.