Re: vnc reverse question

[neil-on-metasploit at restricted.dyndns.org (Neil)]

mmiller at hick.org writes:
> No, the target server does not have to be running VNC, nor does it
> matter if the host is already running VNC.  Reverse refers to how the 
> second stage payload (the thing that loads the VNC DLL into memory) will be 
> obtained, whether by connecting back to the attacker on a given port,
> or by having the attacker connect to the target on a given port.  After
> the VNC DLL has been read in and loaded, the VNC DLL will re-use the
> connection that was created between msf and the target machine for 
> the actual VNC protocol.  As such, no ports need to be bound on the target
> machine for VNC to accept incoming connections because the VNC session
> is simply tunneled through the existing connection. 
>
> Remember that the VNC "server" that is injected into the target process
> is a stripped down version of VNC -- it requires no installation and
> does nothing intrusive to the actual machine itself (such as modifying
> the registry).  It was modified specifically for use with the library
> injection system in metasploit. 

Good thing you mentioned about "no installation and does nothing intrusive 
to the actual machine". I am actually a little concerned with metasploit's 
exploits. Are there exploits bundled with metasploit that actually modifies 
something in the target system? I would like to know because I don't want 
our production server having a strain after I test it. 

So before I do it in PROD, I would like to know first if that's the case. 

Thanks buddy. 

Neil