Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Bug : Strain Rheology in ASPECT 2.5 and deal.II 9.5.0 #5540

Open
naliboff opened this issue Jan 12, 2024 · 1 comment
Open

Bug : Strain Rheology in ASPECT 2.5 and deal.II 9.5.0 #5540

naliboff opened this issue Jan 12, 2024 · 1 comment

Comments

@naliboff
Copy link
Contributor

naliboff commented Jan 12, 2024

A recent forum post reminded me of a bug introduced in #4336 with certain release of deal.II (9.4.2, 9.5.0), but has since been resolved with the following PR (#5335) that switched to using FEPointEvaluation instead of FEFieldFunction when accessing the old solution in the strain rheology module.

Prior to this PR, models using the strain rheology in combination with the free surface would crash soon after startup with the following error:

An error occurred in line <2721> of file </root/bin/tmp/unpack/deal.II-v9.4.2/include/deal.II/dofs/dof_accessor.templates.h> in function
    void dealii::DoFCellAccessor<dim, spacedim, lda>::get_dof_values(const InputVector&, ForwardIterator, ForwardIterator) const [with InputVector = dealii::TrilinosWrappers::MPI::Vector; ForwardIterator = double*; int dimension_ = 2; int space_dimension_ = 2; bool level_dof_access = false]
The violated condition was:
    this->is_artificial() == false
Additional information:
    Can't ask for DoF indices on artificial cells.

Unfortunately, when this error occurs is dependent on resolution and the test for the continental extension cookbook did not catch it as it occurs after 2 time steps in that test (the test is only run for 2 time steps).

Action items:

  1. Update the continental extension cookbook test to run for more time steps
  2. Do a new release of ASPECT 2.5, as the current version will result in model crashes for certain classes of simulations
  3. Before a major release, run the full cookbooks for complicated models (continental extension, etc) even for minor PR changes that affect them (I should have done this for Update strain rheology #4336).
@gassmoeller
Copy link
Member

Are you sure this is a problem that exists in ASPECT 2.5.0? I checked the commit log, and I dont think #4336 made it into v2.5.0. We released 2.5. after your pull request was merged into main, but we created the release branch before the hackathon at which it was merged. The bug should only have been in the development version of ASPECT for about a month, but never in a release.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants