I don't think that could be the problem. I can ssh between machines,
have a coulple of common directories shared with NFS, etc. And OpenMPI
runs (or starts, anyway) under ssh, doesn't it?
On Fri, 23 Jul 2010 14:17:48 -0700, Ralph Castain <rhc_at_[hidden]> wrote:
> Check for a firewall blocking tcp communications - that's the most
> common issue.
> On Jul 23, 2010, at 3:05 PM, James wrote:
>> I am trying to get OpenMPI running on my home network. This has two
>> machines, t61 and quad, both running SuSE 11. I'm using the "hello_c"
>> program from the examples as a test. It will run fine on each machine,
>> using whatever number or processes I specify. However, when I try to
>> run on multiple machines, it hangs.
>> If I start from t61 with the command "mpiexec -host t61,quad -np 2
>> then I see that command when I do a ps -ax on t61. On quad I see
>> "orted --daemonize (long parameter string)". Both of them seem to be
>> silently waiting on some event, but I've no idea what.
>> Both machines are running OpenMPI 1.4.2 (compiled from same tar file),
>> installed in /opt/openmpi. The executables are in the same user/path
>> on each machine (/home/me/src/openmpi/examples), and path,
>> LD_LIBRARY_PATH, and so on all seem the same.
>> Any suggestions?
>> PS: Also, may I suggest putting something in the FAQ pointing out
>> that the environment vars need to be set in .tcshrc, not .login?
>> It would have saved me several hours.
>> users mailing list
> users mailing list