Open MPI logo

Open MPI User's Mailing List Archives

  |   Home   |   Support   |   FAQ   |   all Open MPI User's mailing list

Subject: [OMPI users] orterun, orted, and chroot
From: Adam C Powell IV (hazelsct_at_[hidden])
Date: 2008-07-31 16:18:44


Greetings,

I can't get OpenMPI programs to run in a chroot environment on Debian.
If I run the program, it dies as follows:

# ./ex0
[workhorse:23752] [0,0,0] ORTE_ERROR_LOG: Error in file runtime/orte_init_stage1.c at line 312
--------------------------------------------------------------------------
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_pls_base_select failed
  --> Returned value -1 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: orte_init_stage1 failed
  --> Returned "Error" (-1) instead of "Success" (0)
--------------------------------------------------------------------------
*** An error occurred in MPI_Init
*** before MPI was initialized
*** MPI_ERRORS_ARE_FATAL (goodbye)
[workhorse:23752] Abort before MPI_INIT completed successfully; not able to guarantee that all other processes were killed!

If I run it with mpirun, it does slightly differently:

$ mpirun -np 1 ./ex0
[workhorse:23457] [0,0,0] ORTE_ERROR_LOG: Error in file runtime/orte_init_stage1.c at line 312
--------------------------------------------------------------------------
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_pls_base_select failed
  --> Returned value -1 instead of ORTE_SUCCESS

--------------------------------------------------------------------------
[workhorse:23457] [0,0,0] ORTE_ERROR_LOG: Error in file runtime/orte_system_init.c at line 42
[workhorse:23457] [0,0,0] ORTE_ERROR_LOG: Error in file runtime/orte_init.c at line 52
--------------------------------------------------------------------------
Open RTE was unable to initialize properly. The error occured while
attempting to orte_init(). Returned value -1 instead of ORTE_SUCCESS.
--------------------------------------------------------------------------

The above output is identical whether I use -v, --verbose, or -mca btl
self.

Also, I can't start orted in the chroot: it detaches without leaving a
running process behind. In a normal non-chroot environment it works
fine.

Is it possible to run OpenMPI programs in a chroot environment? I
understand that it might not be possible to use a cluster with chroots
on each node, but how about running a single instance of an MPI program?
Or on a multi-CPU (multi-core etc.) shared memory machine?

Thanks,
-Adam

-- 
GPG fingerprint: D54D 1AEE B11C CE9B A02B  C5DD 526F 01E8 564E E4B6
Engineering consulting with open source tools
http://www.opennovation.com/