On Mar 3, 2006, at 3:19 PM, raghav k wrote:
> Hi,
>
> I am remotely logging in onto a Sun Fire x4200 x64 Server via SSH.
> This server has 2 dual core processors. I wish to run some mpi
> programs on these processors. I have LAM installed on the server.
> What IP address do I need to give in my host file so that I am
> able to boot LAM?
You only need to specify the localhost and the number of cpus.
Something like:
$ cat my-hostfile.def
localhost cpu=4
Then lamboot with this file as an argument:
$ lamboot my-hostfile.def
>
> If I give the IP address of the server or "local host" , I get the
> following errors:
>
> n-1<15926> ssi:boot:open: opening
> n-1<15926> ssi:boot:open: opening boot module globus
> n-1<15926> ssi:boot:open: opened boot module globus
> n-1<15926> ssi:boot:open: opening boot module rsh
> n-1<15926> ssi:boot:open: opened boot module rsh
> n-1<15926> ssi:boot:open: opening boot module slurm
> n-1<15926> ssi:boot:open: opened boot module slurm
> n-1<15926> ssi:boot:select: initializing boot module slurm
> n-1<15926> ssi:boot:slurm: not running under SLURM
> n-1<15926> ssi:boot:select: boot module not available: slurm
> n-1<15926> ssi:boot:select: initializing boot module rsh
> n-1<15926> ssi:boot:rsh: module initializing
> n-1<15926> ssi:boot:rsh:agent: rsh
> n-1<15926> ssi:boot:rsh:username: <same>
> n-1<15926> ssi:boot:rsh:verbose: 1000
> n-1<15926> ssi:boot:rsh:algorithm: linear
> n-1<15926> ssi:boot:rsh:no_n: 0
> n-1<15926> ssi:boot:rsh:no_profile: 0
> n-1<15926> ssi:boot:rsh:fast: 0
> n-1<15926> ssi:boot:rsh:ignore_stderr: 0
> n-1<15926> ssi:boot:rsh:priority: 10
> n-1<15926> ssi:boot:select: boot module available: rsh, priority: 10
> n-1<15926> ssi:boot:select: initializing boot module globus
> n-1<15926> ssi:boot:globus: globus-job-run not found, globus boot
> will not run
> n-1<15926> ssi:boot:select: boot module not available: globus
> n-1<15926> ssi:boot:select: finalizing boot module slurm
> n-1<15926> ssi:boot:slurm: finalizing
> n-1<15926> ssi:boot:select: closing boot module slurm
> n-1<15926> ssi:boot:select: finalizing boot module globus
> n-1<15926> ssi:boot:globus: finalizing
> n-1<15926> ssi:boot:select: closing boot module globus
> n-1<15926> ssi:boot:select: selected boot module rsh
>
> LAM 7.1.1/MPI 2 C++/ROMIO - Indiana University
>
> n-1<15926> ssi:boot:base: looking for boot schema in following
> directories:
> n-1<15926> ssi:boot:base: <current directory>
> n-1<15926> ssi:boot:base: $TROLLIUSHOME/etc
> n-1<15926> ssi:boot:base: $LAMHOME/etc
> n-1<15926> ssi:boot:base: /opt/csw/local/etc
> n-1<15926> ssi:boot:base: looking for boot schema file:
> n-1<15926> ssi:boot:base: hostfile
> n-1<15926> ssi:boot:base: found boot schema: hostfile
> n-1<15926> ssi:boot:rsh: found the following hosts:
> n-1<15926> ssi:boot:rsh: n0 localhost (cpu=4)
> n-1<15926> ssi:boot:rsh: resolved hosts:
> n-1<15926> ssi:boot:rsh: n0 localhost --> 127.0.0.1 (origin)
> n-1<15926> ssi:boot:rsh: starting RTE procs
> n-1<15926> ssi:boot:base:linear: starting
> n-1<15926> ssi:boot:base:server: opening server TCP socket
> n-1<15926> ssi:boot:base:server: opened port 39522
> n-1<15926> ssi:boot:base:linear: booting n0 (localhost)
> n-1<15926> ssi:boot:rsh: starting lamd on (localhost)
> n-1<15926> ssi:boot:rsh: starting on n0 (localhost): hboot -t -c
> lam-conf.lamd -d -v -I -H 127.0.0.1 -P 39522 -n 0 -o 0
> n-1<15926> ssi:boot:rsh: launching locally
> n-1<15926> ssi:boot:base:linear: Failed to boot n0 (localhost)
> n-1<15926> ssi:boot:base:server: closing server socket
> n-1<15926> ssi:boot:base:linear: aborted!
> lamboot did NOT complete successfully
I'm wondering if there is a problem with opening of random ports on
this machine back to itself. Is there a firewall running on this
machine? If so you will need to disable it for the localhost.
>
>
> Can I use LAM while working on the server via SSH from Windows??
Most certainly.
-- Josh
>
> Please help!
>
> Thanks in advance
> Raghav
>
>
>
> Yahoo! Mail
> Use Photomail to share photos without annoying attachments.
> _______________________________________________
> This list is archived at http://www.lam-mpi.org/MailArchives/lam/
----
Josh Hursey
jjhursey_at_[hidden]
http://www.lam-mpi.org/
|
