[gpaw-users] problem getting gpaw test to run

[Sarlo, Jeffrey S]

We found the source of the problem.  We had to rebuild a new version of openmpi with the configure option --disable-dlopen and then rebuilt gpaw.  After that it seems to work OK.

Thanks.
Jeff

-----Original Message-----
From: gpaw-users-bounces at listserv.fysik.dtu.dk [mailto:gpaw-users-bounces at listserv.fysik.dtu.dk] On Behalf Of Sarlo, Jeffrey S via gpaw-users
Sent: Monday, February 05, 2018 6:32 PM
To: gpaw-users at listserv.fysik.dtu.dk
Subject: [gpaw-users] problem getting gpaw test to run

I have built gpaw 1.3.0 from source on RHEL 6.  I am using openmpi 1.10.2 and gnu compilers.  I have python 2.7.  The build seems to go OK without errors.  When I try to run gpaw test, I get MPI errors.  I am not sure what I need to do different for gpaw to be able to run.

Any suggestions would be appreciated.

Thanks.
Jeff


$ gpaw test
[jeff.hpc.uh.edu:15559] mca: base: component_find: unable to open /share/apps/openmpi-1.10.2/lib/openmpi/mca_shmem_mmap: /share/apps/openmpi-1.10.2/lib/openmpi/mca_shmem_mmap.so: undefined symbol: opal_show_help (ignored) [jeff.hpc.uh.edu:15559] mca: base: component_find: unable to open /share/apps/openmpi-1.10.2/lib/openmpi/mca_shmem_sysv: /share/apps/openmpi-1.10.2/lib/openmpi/mca_shmem_sysv.so: undefined symbol: opal_show_help (ignored) [jeff.hpc.uh.edu:15559] mca: base: component_find: unable to open /share/apps/openmpi-1.10.2/lib/openmpi/mca_shmem_posix: /share/apps/openmpi-1.10.2/lib/openmpi/mca_shmem_posix.so: undefined symbol: opal_show_help (ignored)
--------------------------------------------------------------------------
It looks like opal_init failed for some reason; your parallel process is likely to abort.  There are many reasons that a parallel process can fail during opal_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):

  opal_shmem_base_select failed
  --> Returned value -1 instead of OPAL_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):

  opal_init failed



_______________________________________________
gpaw-users mailing list
gpaw-users at listserv.fysik.dtu.dk
https://listserv.fysik.dtu.dk/mailman/listinfo/gpaw-users