On Mon, 29 Oct 2018, Mark Adams via petsc-dev wrote:

On Mon, Oct 29, 2018 at 5:26 PM Balay, Satish <[email protected]> wrote:

...

Can you send corresponding logs?

...

...

...

...

Source: #include "confdefs.h" #include "conffix.h" #ifdef PETSC_HAVE_STDLIB_H #include <stdlib.h> #endif #include <mpi.h>

int main() { MPI_Aint size; int ierr; MPI_Init(0,0); ierr = MPI_Type_extent(MPI_INT64_T, &size); if(ierr || (size == 0)) exit(1); MPI_Finalize(); ; return 0; } Pushing language C Popping language C Executing: cc -o /tmp/petsc-yiGfSd/config.packages.MPI/conftest -O /tmp/petsc-yiGfSd/config.packages.MPI/conftest.o -ldl Testing executable /tmp/petsc-yiGfSd/config.packages.MPI/conftest to see if it can be run Executing: /tmp/petsc-yiGfSd/config.packages.MPI/conftest Executing: /tmp/petsc-yiGfSd/config.packages.MPI/conftest ERROR while running executable: Could not execute "['/tmp/petsc-yiGfSd/config.packages.MPI/conftest']": [Mon Oct 29 17:44:45 2018] [unknown] Fatal error in MPI_Init: Other MPI error, error stack: MPIR_Init_thread(537): MPID_Init(249).......: channel initialization failed MPID_Init(638).......: PMI2 init failed: 1 <<<<<<< So all MPI tests fail on frontend? And you need to use --with-batch? Satish