[Pkg-openmpi-maintainers] Bug#839387: libopenmpi2: Programs crash if run under fakeroot

Michael Banck mbanck at debian.org
Sat Oct 1 10:34:02 UTC 2016


Package: libopenmpi2
Severity: important
Version: 2.0.1-5

Hi,

if I run a program linked to OpenMPI's libmpi under fakeroot, it
crashes:

|(unstable-amd64-sbuild)mba at hartree:/$ OMPI_MCA_orte_rsh_agent=/bin/false
|pw.x
|
|     Program PWSCF v.5.4.0 starts on  1Oct2016 at 10:29:47
|
|     This program is part of the open-source Quantum ESPRESSO suite
|     for quantum simulation of materials; please cite
|         "P. Giannozzi et al., J. Phys.:Condens. Matter 21 395502
|(2009);
|          URL http://www.quantum-espresso.org",
|     in publications or presentations arising from this work. More
|details at
|     http://www.quantum-espresso.org/quote
|
|     Parallel version (MPI), running on     1 processors
|Open_input_file: error opening input_tmp.in
|
| %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|     Error in routine read_input (2):
|     opening input file
| %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
|     stopping ...
|At line 97 of file error_handler.f90 (unit = 99)
|Fortran runtime error: Cannot open file 'CRASH': Permission denied
|
|Error termination. Backtrace:
|#0  0x7f4082814d4a
|#1  0x7f4082815825
|#2  0x7f4082815f79
|#3  0x7f40828d5e4a
|#4  0x7f40828d608d
|#5  0x657c0a
|#6  0x6adf9a
|#7  0x407cfd
|#8  0x7f4081b042b0
|#9  0x407d59
|#10  0xffffffffffffffff
|(unstable-amd64-sbuild)mba at hartree:/$ OMPI_MCA_orte_rsh_agent=/bin/false
|fakeroot pw.x
|[hartree:10002] PMIX ERROR: INVALID-CREDENTIAL in file
|src/server/pmix_server_listener.c at line 540
|[hartree:10001] PMIX ERROR: UNREACHABLE in file src/client/pmix_client.c
|at line 983
|[hartree:10001] PMIX ERROR: UNREACHABLE in file src/client/pmix_client.c
|at line 199
|--------------------------------------------------------------------------
|It looks like orte_init failed for some reason; your parallel process is
|likely to abort.  There are many reasons that a parallel process can
|fail during orte_init; some of which are due to configuration or
|environment problems.  This failure appears to be an internal failure;
|here's some additional information (which may only be relevant to an
|Open MPI developer):
|
|  init pmix failed
|  --> Returned value Unreachable (-12) instead of ORTE_SUCCESS
|--------------------------------------------------------------------------
|--------------------------------------------------------------------------
|It looks like orte_init failed for some reason; your parallel process is
|likely to abort.  There are many reasons that a parallel process can
|fail during orte_init; some of which are due to configuration or
|environment problems.  This failure appears to be an internal failure;
|here's some additional information (which may only be relevant to an
|Open MPI developer):
|
|  orte_ess_init failed
|  --> Returned value Unreachable (-12) instead of ORTE_SUCCESS
|--------------------------------------------------------------------------
|--------------------------------------------------------------------------
|It looks like MPI_INIT failed for some reason; your parallel process is
|likely to abort.  There are many reasons that a parallel process can
|fail during MPI_INIT; some of which are due to configuration or
|environment
|problems.  This failure appears to be an internal failure; here's some
|additional information (which may only be relevant to an Open MPI
|developer):
|
|  ompi_mpi_init: ompi_rte_init failed
|  --> Returned "Unreachable" (-12) instead of "Success" (0)
|--------------------------------------------------------------------------
|*** An error occurred in MPI_Init
|*** on a NULL communicator
|*** MPI_ERRORS_ARE_FATAL (processes in this communicator will now abort,
|***    and potentially your MPI job)
|[hartree:10001] Local abort before MPI_INIT completed completed
|successfully, but am not able to aggregate error messages, and not able
|to guarantee that all other processes were killed!
|(unstable-amd64-sbuild)mba at hartree:/$ libfakeroot, when sending message:
|Invalid argument
|libfakeroot, when sending message: Invalid argument
|libfakeroot, when sending message: Invalid argument
|semop(1): encountered an error: Invalid argument

pw.x (from quantum-espresso) is linked to the fortran MPI library as
well, not sure if that matters:

|(unstable-amd64-sbuild)mba at hartree:/$ ldd /usr/bin/pw.x | grep mpi
|        libscalapack-openmpi.so.1 => /usr//lib/libscalapack-openmpi.so.1 (0x00007fded925f000)
|        libblacs-openmpi.so.1 => /usr//lib/libblacs-openmpi.so.1 (0x00007fded902a000)
|        libblacsF77init-openmpi.so.1 => /usr//lib/libblacsF77init-openmpi.so.1 (0x00007fded8e28000)
|        libmpi_usempif08.so.20 => /usr//lib/libmpi_usempif08.so.20 (0x00007fded7d85000)
|        libmpi_usempi_ignore_tkr.so.20 => /usr//lib/libmpi_usempi_ignore_tkr.so.20 (0x00007fded7b7f000)
|        libmpi_mpifh.so.20 => /usr//lib/libmpi_mpifh.so.20 (0x00007fded7928000)
|        libmpi.so.20 => /usr//lib/libmpi.so.20 (0x00007fded7636000)
|        libblacsCinit-openmpi.so.1 => /usr//lib/libblacsCinit-openmpi.so.1 (0x00007fded6073000)




Michael




More information about the Pkg-openmpi-maintainers mailing list