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
sporadic python MPI unit test failure in test_sub_bbtree_box: incompatible constructor arguments #2070
Comments
The error looks to be due to the fact that the input array has gotten the wrong dtype (float64) insstead of int32. In [1]: import numpy as np
In [2]: a = np.unique([])
In [3]: a.dtype
Out[3]: dtype('float64')
In [4]: a = np.unique([1,2,3])
In [5]: a.dtype
Out[5]: dtype('int64') On main, the unique has been removed:
|
Thanks @jorgensd . I'll try patching it without unique and see if that clears it up. |
the |
After rebuilding dolfinx (0.3.0) against PETSc 3.16 (and scalapack 2.23, scotch 7.0.1), the
test_sub_bbtree_box
python unit test from python/test/unit/geometry/test_bounding_box_tree.py fails inconsistently under mpi (openmpi)See also debian test logs at https://ci.debian.net/data/autopkgtest/unstable/amd64/f/fenics-dolfinx/20037551/log.gz or https://ci.debian.net/packages/f/fenics-dolfinx/
The failure is sporadic in the sense that if I run only test_sub_bbtree_box on 2 processes,
then sometimes it passes, sometimes it still fails. What would make it behave inconsistently like that?
Has anyone else experienced this kind of failure or know what's causing it? Is it likely to have been fixed since 0.3.0?
The text was updated successfully, but these errors were encountered: