tar is much friendlier not, but ...

[Jeffrey Pyne]

On Friday, October 25, 2002 3:48 PM, George Gambill wrote:

> When I enter the command "rdesktop xxxxxxxx" where xxxxxxxx 
> is the name of
> the Windows 2000 Terminal Server, I get:
> 
> -bash: rdesktop: command not found

This is telling you that your shell (in this case bash) did not find a
command called "rdesktop."  
 
> I get the same resultes without xxxxxxxx which makes sense

Yep.
 
> FWIW I can ping the Terminal Server by name (xxxxxxxx) and it 
> does respond
> well.

Okay.
 
> I read the -rwxr-xr-x on rdesktop  (in green) as being executable.

Cool.
 
> Where am I going wrong?

Typically, when you install a program from source code, you will execute 3
commands, in the following order:

./configure
make
make install

I happen to have the rdesktop source code on my machine, and I see that they
kindly omitted any INSTALL or README files which would have explained this
to you.  But I do see that they have a configure script, so the above
sequence must be the route I took when I installed rdesktop on my machine
several months ago (it's very kewl, by the way).  The configure script will
examine your machine's environment and create a custom Makefile that takes
into account the vagaries of your environment.  The "make" command compiles
the source.  The "make install" command installs the program (usually into
/usr/local/bin).  I'm leaving out a lot of details, and this sequence isn't
always The Way, but 99 times out of 100, CMMI is how you install a program
from source.

Since it looks like the "make" succeeded (because you have an executable
"rdesktop"), go ahead and run a "make install".  This should install
rdesktop.  I looked on my machine and it installed into /usr/local/bin.
Type "which rdesktop" to verify that your shell can find the program.  If it
does, try running it again.  If it doesn't, we'll have to explain $PATH and
aliases.  :)

> Thanking you all in advance.
> 
> George

Hope that helps,

~Jeff