Maciek Lewiński
2016-05-02 21:36:11 UTC
Hi, I'm having problem with Open MPI version 1.10.2.
I've installed two virtual machines on VirtualBox, both are the same images
of Ubuntu 12.04 64bit.
Both have the same accounts, both have everything configured almost exactly
the same.
I have configured OMPI only with the --prefix to specify my location of
install folder which is /home/$USER/.openmpi.
Users on both machines are identical.
On both machines running mpirun on one of the example programs work
perfectly, on both machines I've added bin and lib to corresponding PATHs,
env command ran on master and through ssh on slave1 gives these results:
***@osboxes:~/cloud$ env | grep PATH
LD_LIBRARY_PATH=:/usr/local/lib:/usr/local/lib:/home/osboxes/.openmpi/lib
PATH=/usr/lib/lightdm/lightdm:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/bin:/usr/local/bin:/home/osboxes/.openmpi/bin
On both hosts I have installed ssh which works, I've enabled passwordless
connection, which again, works,
I have created an NFS folder in which I hold the hello_c program from
examples and hosts file that looks like this:
192.168.0.191 master
192.168.0.190 slave1
The same two lines are placed in /etc/hosts for ssh to work.
On both machines, running:
***@osboxes:~/cloud$ mpirun -np 1 ./hello_c
Hello, world, I am 0 of 1, (Open MPI v1.10.2, package: Open MPI
***@osboxes Distribution, ident: 1.10.2, repo rev:
v1.10.1-145-g799148f, Jan 21, 2016, 126)
Works, even running this command on slave1 through ssh from master works as
expected.
Yet when I try to execute the following command I get the error:
***@osboxes:~/cloud$ mpirun -np 2 --hostfile hosts ./hello_c
bash: orted: command not found
--------------------------------------------------------------------------
ORTE was unable to reliably start one or more daemons.
This usually is caused by:
* not finding the required libraries and/or binaries on
one or more nodes. Please check your PATH and LD_LIBRARY_PATH
settings, or configure OMPI with --enable-orterun-prefix-by-default
* lack of authority to execute on one or more specified nodes.
Please verify your allocation and authorities.
* the inability to write startup files into /tmp
(--tmpdir/orte_tmpdir_base).
Please check with your sys admin to determine the correct location to use.
* compilation of the orted with dynamic libraries when static are required
(e.g., on Cray). Please check your configure cmd line and consider using
one of the contrib/platform definitions for your system type.
* an inability to create a connection back to mpirun due to a
lack of common network interfaces and/or no route found between
them. Please check network connectivity (including firewalls
and network routing requirements).
--------------------------------------------------------------------------
The same happens when I use -host master,slave1 instead of --hostfile
hosts. I'm slowly running out if ideas, I've tried anything I could find on
the internet or OMPI FAQ, nothing seems to work. What am I doing wrong?
I've installed two virtual machines on VirtualBox, both are the same images
of Ubuntu 12.04 64bit.
Both have the same accounts, both have everything configured almost exactly
the same.
I have configured OMPI only with the --prefix to specify my location of
install folder which is /home/$USER/.openmpi.
Users on both machines are identical.
On both machines running mpirun on one of the example programs work
perfectly, on both machines I've added bin and lib to corresponding PATHs,
env command ran on master and through ssh on slave1 gives these results:
***@osboxes:~/cloud$ env | grep PATH
LD_LIBRARY_PATH=:/usr/local/lib:/usr/local/lib:/home/osboxes/.openmpi/lib
PATH=/usr/lib/lightdm/lightdm:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/bin:/usr/local/bin:/home/osboxes/.openmpi/bin
On both hosts I have installed ssh which works, I've enabled passwordless
connection, which again, works,
I have created an NFS folder in which I hold the hello_c program from
examples and hosts file that looks like this:
192.168.0.191 master
192.168.0.190 slave1
The same two lines are placed in /etc/hosts for ssh to work.
On both machines, running:
***@osboxes:~/cloud$ mpirun -np 1 ./hello_c
Hello, world, I am 0 of 1, (Open MPI v1.10.2, package: Open MPI
***@osboxes Distribution, ident: 1.10.2, repo rev:
v1.10.1-145-g799148f, Jan 21, 2016, 126)
Works, even running this command on slave1 through ssh from master works as
expected.
Yet when I try to execute the following command I get the error:
***@osboxes:~/cloud$ mpirun -np 2 --hostfile hosts ./hello_c
bash: orted: command not found
--------------------------------------------------------------------------
ORTE was unable to reliably start one or more daemons.
This usually is caused by:
* not finding the required libraries and/or binaries on
one or more nodes. Please check your PATH and LD_LIBRARY_PATH
settings, or configure OMPI with --enable-orterun-prefix-by-default
* lack of authority to execute on one or more specified nodes.
Please verify your allocation and authorities.
* the inability to write startup files into /tmp
(--tmpdir/orte_tmpdir_base).
Please check with your sys admin to determine the correct location to use.
* compilation of the orted with dynamic libraries when static are required
(e.g., on Cray). Please check your configure cmd line and consider using
one of the contrib/platform definitions for your system type.
* an inability to create a connection back to mpirun due to a
lack of common network interfaces and/or no route found between
them. Please check network connectivity (including firewalls
and network routing requirements).
--------------------------------------------------------------------------
The same happens when I use -host master,slave1 instead of --hostfile
hosts. I'm slowly running out if ideas, I've tried anything I could find on
the internet or OMPI FAQ, nothing seems to work. What am I doing wrong?