running MPI, and login in locally into slave nodes

german kogan gkogan at students.uiuc.edu
Fri Aug 3 14:58:20 PDT 2001


On Fri, 3 Aug 2001, Sean Dilda wrote:

> On Fri, 03 Aug 2001, german kogan wrote:
>
> >
> >
> > Hi.
> >
> > I am really new at this. I just loaded Scyld  on my front end Machine and
> > got one slave node up using NodeBoot.img over network from the master
> > node. Eventually I will build a cluster. So I have a few question.
> >
> >
> > 1) Was the Software that was on the master node installed onto the
> > slave node?
>
> No, it wasn't.  Our software is designed so that we don't need a full
> install on the slave nodes.  All our jobs start on the master node and
> move, using BProc, to the slave nodes.
>
> >
> > 2) Is it possible to login locally into a slave node? Because there is no
> > prompt for loging in on the slave node and for some reason I could not tell
> > into it from the master node.
>
> That is correct.  You don't log into slave nodes, you just send jobs to
> them.  If you need to run a command on a slave node, try:
> bpsh <node number> <command>

But what if I need to lets say change a specific file on a particular
slave node? Also I ran a command bpsh 0 ls -la, I have one slave node up,
and I got back the exact same list of files as in my home directory on the
master node. Is this supposed to be like this? Are the slave node exact
copies of the master node? Also my ping is working but I can't telnet or
ssh to the slave node.




> >
> > 3) I was trying to run a simple MPI program. And when I tried to run it
> > using the command /usr/mpi-beowulf/bin/mpirun -np 2 a.out I got the
> > following error message "p0_2813: p4_error: net_create_slave: host not
> > bproc node: -3 p4_error: latest msg from perror: Success. However, it did
> > work when I used 1 process instead of 2. Any ideas what
> > does this mean and what am I doing wrong.
>
> Hrm.. do you have any slave nodes up?  You can check that with 'bpstat'.
> Also, are you trying to use a pgp4 file?
>

Ye I got one slave node up, number 0. It comes up when I do bstat. Whats a
pgp4 file? I was trying to run a simple C based MPI program.

Thanks






More information about the Beowulf mailing list