On Thu, 24 Oct 2002, Jeff Squyres wrote:

> On Thu, 24 Oct 2002, sameer kanwar wrote:
>
> > i am running OSCAR 1.2.1 on linux 7.1. i would like to use rsh instead
> > of ssh as is default done by OSCAR,(i know the implications),how does
> > one do it ?
>
> I first have to make the obligatory warning: using rsh is generally a bad
> idea.  It also offers little functionality over ssh -- indeed, ssh offers
> *MUCH* more functionality than rsh (security, automatic X forwarding,
> better job control, etc.).  I would personally *strongly* advises you not
> to use rsh (even if you're on a private cluster).
>
> The next question is: what do you want to use rsh *for*?  Just for MPI?
> Most of the other services in OSCAR don't use rsh/ssh.
>
> In the case of LAM/MPI, all you have to do is set the environment variable
> (preferably in some system-level config file, such as /etc/profile -- on
> the head node *and* every client node) LAMRSH to "rsh".  This environment
> variable is a run-time override for LAM/MPI.
>
> I think there's a corresponding environment variable for MPICH, but I'm
> not sure, offhand.

Advice on using ssh seconded...

To enable rsh-abilities though, you'll also need to modify
/etc/xinet.d/rsh and/or rexec and/or rlogin, cpush these over to the
clients, and cexec a restart of xinetd.

Paul Gray                                         -o)
323 Wright Hall                                   /\\
University of Northern Iowa                      _\_V
  Message void if penguin violated ...  Don't mess with the penguin
  No one ever says "Hey, I can't read that ASCII e-mail ya sent me."



-------------------------------------------------------
This sf.net email is sponsored by: Influence the future 
of Java(TM) technology. Join the Java Community 
Process(SM) (JCP(SM)) program now. 
http://ad.doubleclick.net/clk;4729346;7592162;s?http://www.sun.com/javavote
_______________________________________________
Oscar-users mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/oscar-users

Reply via email to