FENICSX error thrown but FESTIM solving

[kaelyndunnell]

From PR #12: simulation solves, plots, and then spits out FENICSx error in the terminal. merged into main

  You need to add matplotlib as a conda dependency. This is what is making the Ci fail. Here, matplotlib is installed alongside HTM as a pip package. This conflicts with dolfinx which is installed with Conda.

(hello @shimwell)

Originally posted by @RemDelaporteMathurin in #12 (comment)

[RemDelaporteMathurin]

I've produced a MWE for this bug.

import festim as F

my_model = F.HydrogenTransportProblem()

my_model.mesh = F.Mesh1D([0, 0.1, 0.2, 0.3, 0.5, 1])

tungsten = F.Material(D_0=1, E_D=0)

w_subdomain = F.VolumeSubdomain1D(id=1, borders=[0, 1], material=tungsten)

my_model.subdomains = [w_subdomain]

my_model.species = [F.Species("H")]

my_model.temperature = 500

my_model.boundary_conditions = []


my_model.settings = F.Settings(atol=1e-10, rtol=1e-10, final_time=1)

my_model.settings.stepsize = F.Stepsize(initial_value=0.1)

my_model.initialise()
my_model.run()


class RandomClass:
    def __init__(self):
        pass

This produces

Solving HydrogenTransportProblem: 100%|█████████████████████████████████████████████████| 1.00/1.00 [00:00<00:00, 663it/s][0]PETSC ERROR: ------------------------------------------------------------------------
[0]PETSC ERROR: Caught signal number 11 SEGV: Segmentation Violation, probably memory access out of range
[0]PETSC ERROR: Try option -start_in_debugger or -on_error_attach_debugger
[0]PETSC ERROR: or see https://petsc.org/release/faq/#valgrind and https://petsc.org/release/faq/
[0]PETSC ERROR: configure using --with-debugging=yes, recompile, link, and run 
[0]PETSC ERROR: to get more information on the crash.
[0]PETSC ERROR: Run with -malloc_debug to check if memory corruption is causing the crash.
--------------------------------------------------------------------------
MPI_ABORT was invoked on rank 0 in communicator MPI_COMM_WORLD
  Proc: [[13633,0],0]
  Errorcode: 59

NOTE: invoking MPI_ABORT causes Open MPI to kill all MPI processes.
You may or may not see output from other processes, depending on
exactly when Open MPI kills them.
--------------------------------------------------------------------------

Environment file:

name: hisp-env
channels:
  - conda-forge
  - defaults
dependencies:
  - fenics-dolfinx=0.8.0
  - matplotlib
  - scipy
  - pip
  - tqdm
  - pip:
      - git+https://github.com/remdelaportemathurin/FESTIM@ad3b10b1a0c736ea3cd1e77835ef70de82208a10
    # - h_transport_materials~=0.16

(both with dolfinx 0.8 and 0.9)

1. Not defining RandomClass at the end fixes the issue. 😕
2. adding my_model.show_progress_bar = False fixes the issue
   This removes the tqdm progress bar

[RemDelaporteMathurin]

Note: the bug also occurs with festim@fenicsx

[RemDelaporteMathurin]

If we don't define the RandomClass in the same file but import it from another file, the error doesn't occur:

import festim as F

my_model = F.HydrogenTransportProblem()

my_model.mesh = F.Mesh1D([0, 0.1, 0.2, 0.3, 0.5, 1])

tungsten = F.Material(D_0=1, E_D=0)

w_subdomain = F.VolumeSubdomain1D(id=1, borders=[0, 1], material=tungsten)

my_model.subdomains = [w_subdomain]

my_model.species = [F.Species("H")]

my_model.temperature = 500

my_model.boundary_conditions = []


my_model.settings = F.Settings(atol=1e-10, rtol=1e-10, final_time=1)

my_model.settings.stepsize = F.Stepsize(initial_value=0.1)

my_model.initialise()
my_model.run()

from random_class import RandomClass

[RemDelaporteMathurin]

@kaelyndunnell adding

my_model.progress_bar.close()

After my_model.run() fixes the issue.

I'll open an issue on the FESTIM repo and push a fix to do this inside my_model.run

[RemDelaporteMathurin]

This PR should fix it after it's merged festim-dev/FESTIM#894