Hi,
I ran the steps 6-8 in examples/3d/hex8 and everything works fine, no error messages show up. However, when I try to view the stepXX-fault_info.xmf file in Paraview, it seems that the results for final_slip_creep and final_slip_one, as well as slip_time_creep and slip_time_one, are inverted. I did not change the original .cfg files, so I wonder what could cause this?
Thanks!
You are correct! Thanks for reporting this.
I am seeing that the field label final_slip_creep is really final_slip_one and vice versa. The Xdmf is correctly pointing to the corresponding field in the HDF5 file, so the error is occurring in getting the info fields within PyLith. The likely issue is that we are either using a buffer with stale data or we have a mismatch between the earthquake rupture data and the name of the earthquake rupture. My guess is that it is the latter. I don’t see an obvious mistake in taking a quick look at the code.
We are getting close to finishing the v3.0 release, so we may decide to make note of this bug but not fix it. The v3.0 release is almost a complete rewrite of PyLith and the way we store this type of information and write it out is completely different, so this same bug doesn’t exist in v3.0.
See Fault info fields for ruptures swapped in HDF5 output in the GitHub PyLith repo issues to track any further discussion or information (including potential resolution) about this bug.