I was finally able to obtain this problem on "reasonable" mesh sizes. To me, it seems that some ranks are remain stuck in the while loop at line SU2_CFD/src/solvers/CSolver.cpp:3260 even though the master rank says that the interpolation has finished. I was able to reproduce this error with 8-12-16-18 cores but no error was encountered with 4-6-10-14-20 cores. It does not change if a combination of threads and cores is used as long as the number of cores is among the list of those reproducing the error.

Digging a little bit in the code, it seems to be stuck in the while loop as said before. In particular, it seems that for some points it is not able to pass the two checks at lines 3275 and 3276 of the same file. I will try digging more.

At the following link you can find the files to reproduce the error.

https://polimi365-my.sharepoint.com/:f:/g/personal/10507725_polimi_it/EjMatPcoa8hPsvVyADfOZL0BQ7YMu7K3WvZaSCaoEeXy1Q?e=x3nRZy

Desktop (please complete the following information):

• OS: CentOS 7
• C++ compiler and version: gcc 10.2.0
• MPI implementation and version: mpirun (Open MPI) 4.1.6
• SU2 Version: v8.0.1