There are so many microsoft newsgroups I have no idea if this is the
right place to ask but I'll try anyhow...  8-)

I have a testing environment that is set up in such a way that after one
logs in the application is automatically started and a window named
'test' is displayed that waits for the user to hit the ENTER key and
then begins executing.  This system also has rcmd service running on it
for doing various remote mgmt tasks such as rebooting it.  What I'd like
to do is use rcmd to remotely run a local program that sends the ENTER
to the system and so remotely start the tests running.

Being a perl user - though I don't think this is a perl problem - I
found a tool that allows me to what I want.  It looks for the window
named 'test', selects it and send an ENTER to it.  Works great when
testing it locally on that test machine but when I run it remotely via
rcmd the call that sets the foregoround window fails.

My initial thought is that the reason is because the script isn't
associated with any windows and therefore may not have the appropriate
context to find any other windows.  To test this I used rcmd to simply
create an 'at' job that runs the script via "cmd /c scriptname" to give
it a windows context but that fails too.  Anybody ever see this sort of
behavior before?  Might it simply be a question of adding some more
switches to 'cmd'?

A second thing I've noticed with rcmd is that if I execute 'net use' on
a remote mahine it tells me all my network drives are unavailable but I
know that are in fact associated with valid paths.  I'm guessing this
might be some sort of permissions thing, but haven't a clue how to track
this one down.

Can anyone help or at least point me in the right direction?

-mark