Upcoming pen test - ntpasswd solutions?
First of all, let me say at the beginning of this thread, that I am not being lazy by simply coming to this board and asking for the answer! I am certainly doing in-depth research on the subject myself, but the answers seem to be quite sparse.
Okay, I have an upcoming vulnerability assesment / penetration test that I need to conduct for a small network. Unfortunately, a portion of the network consists of Windows (2003 mainly) computers. (I say unfortunately, because Windows is not my area of expertise)
The pen test will consist of three phases. Local, remote and a wireless segment.
Now, I already know from some preliminary information, that it will be possible to gain admin access to the Windows 2003 server locally through the use of ntpasswd.
As part of the post-assesment report, I am going to need to present the client with solutions to secure his network. This includes giving him a solution to protect his network from tools like ntpasswd.
I have been researching this matter, and the only solid answer I can seem to find is to use an encrypted filesystem with a master password. Some of the other solutions I have heard are:
1) Use a BIOS password - (not a secure solution, because the password can be reset from the MB)
2) Deny CDROM and Floppy drive access on boot from within BIOS - (again, not a secure solution, because the BIOS password can be reset, giving an attacker the ability to give himself access to the CDROM and/or floppy drive on boot)
These are the only suggestions I have heard so far. Is an encrypted filesystem + master password truly the only reliable solution to protect a Windows server from the use of ntpasswd?