Newsgroups: comp.parallel.mpi
From: "Joel R. Pullen" <jrpullen@ocean.st.usm.edu>
Subject: Re: LAM 6.1 on Linux
Organization: University of Southern Mississippi
Date: Mon, 15 Dec 1997 20:58:16 -0600
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Message-ID: <3495EE48.FF84A117@ocean.st.usm.edu>

Andrea Belloni wrote:

> I have compiled and installed LAM 6.1 on my pentium based PC with Linux Red
> Hat
> 4.1 (kernel 2.0.27) with no problem, but i am not able to boot LAM.
> An interactive session follows:
>
>     [andrea@galileo andrea]$ lamboot -v -d
>     LAM 6.1 - Ohio Supercomputer Center
>     lamboot: boot schema file: /usr/local/lib/lam61/boot/bhost.def
>     lamboot: found 1 host node(s)
>     lamboot: origin node is 0
>     hboot n0 (127.0.0.1)...
>     tkill ...
>     boot ...
>     [1]  10178 lamd -H 127.0.0.1 -P 1151 -n 0 -o 0
>     lamboot: hboot failed on n0 (127.0.0.1)
>     lamboot: unknown host
>     wipe ...
>     tkill n0 (127.0.0.1)...
>     [andrea@galileo andrea]$
>     [andrea@galileo andrea]$ lamd -H 127.0.0.1 -P 1157 -n 0 -o 0
>     dli_inet (sfh_sock_open_clt_inet_stm): Connection refused
>     [andrea@galileo andrea]$
>
> I would like to have a solution for this problem. If you need more
> informations
> let me know.
> Thanks in advance
> Andrea Belloni
> belloni@box1.tin.it

  Have you tried to debug the problem with   recon -v bootfile.dat??? They
have -d options to show more details... I haven't found LAM 6.1 to my
liking... MPICH is much simpler to use and I would recommend try it instead...
But also I think you need to supply a hostname for your LAM nodes in the
/etc/hosts file.
Also make sure you have a .rhosts file with the proper information to be able
to rsh to it :)

Hope this helps ya.

-Joel R. Pullen

email: jrpullen@ocean.st.usm.edu