DFTB+ is getting segmentation fault during mpirun in docker container

[pritam-acog]

Describe the bug
When executing dftb+ inside a docker container using docker run command we are getting Segmentation fault

To Reproduce

1. Execute the dftb+ in mpirun using docker run. For example:

docker run -it --rm --name dftbplus -v $PWD/input:/input -v $PWD/scripts:/scripts -v $PWD/output:/output dftbplus:mpi bash -c "mpirun -np 4 dftb+ >dftb_out.log"

2. Check the dftb_out.log it will show bellow error at the end.

===================================================================================
=   BAD TERMINATION OF ONE OF YOUR APPLICATION PROCESSES
=   PID 10 RUNNING AT 0e928b6d4c7f
=   EXIT CODE: 9
=   CLEANING UP REMAINING PROCESSES
=   YOU CAN IGNORE THE BELOW CLEANUP MESSAGES
===================================================================================
YOUR APPLICATION TERMINATED WITH THE EXIT STRING: Segmentation fault (signal 11)
This typically refers to a problem with your application.
Please see the FAQ page for debugging suggestions

But

1. If Execute dftb+ manually it is not giving any Segmentation fault. Example:

docker run -it --rm --name dftbplus -v $PWD/input:/input -v $PWD/scripts:/scripts -v $PWD/output:/output dftbplus:mpi bash

And inside the container execute

mpirun -np 4 dftb+ >dftb_out.log

• This will not give any Segmentation fault

Note:

All the inputs are same for both manual and automatic execution in docker container.

Expected behaviour
No Segmentation fault should be there while automatically executing inside docker container using Docker run command.

Additional context
In the docker Image, bellow packages are installed.

1. apt update && apt-get update && apt-get install build-essential git vim -y
2. conda install mamba -c conda-forge -n base
3. mamba install -c conda-forge 'dftbplus=23.1=mpi_mpich_*' -y
4. conda install -c conda-forge dftbplus-tools -y

And OMP_NUM_THREADS is set as 1 inside the docker container.