On Fri, 23 Jan 2004, 70uf33q Hu5541n wrote:
> I can ssh and rsh into the other node without problems.
-----
n0<3474> ssi:boot:rsh: attempting to execute "rsh node2 -n echo $SHELL"
n0<3474> ssi:boot:rsh: remote shell /bin/bash
n0<3474> ssi:boot:rsh: attempting to execute "rsh node2 -n hboot -t -c lam-conf.lamd -d -s -I "-H 192.168.0.1 -P 32989 -n 1 -o 0""
ERROR: LAM/MPI unexpectedly received the following on stderr:
hboot: cannot find process schema lam-conf.lamd: No such file or directory
-----
Clearly, this part is your problem.
Are you sure that LAM/MPI is installed properly on the 192.168.0.1
machine? What is happening (as you surmised) is that LAM's hboot command
is not finding the file lam-conf.lamd. The usual suspects here are:
1. You have a broken installation of LAM, and the lam-conf.lamd file is
missing (it should be in /etc/lam/lam-conf.lamd if you installed LAM via
RPM, or $prefix/etc/lam-conf.lamd if you installed it via source).
2. You have multiple installations of LAM on your 0.1 machine, and you're
using the hboot from the "wrong" installation. In this case, it's
probably a whole lot easier just to ensure that there's only one LAM
installation on the machine, and ensure that your $PATH is set (via your
"dot" files -- see the LAM FAQ) to point to the Right one.
Hope this helps.
--
{+} Jeff Squyres
{+} jsquyres_at_[hidden]
{+} http://www.lam-mpi.org/
|
