Tim Brown wrote:
> On Monday 10 November 2008 04:10:34 Ron wrote:
>
> The licensing will be in relation to the source code, OTOH you need to be in 
> a 
> position where you have a license on the source code that allows you to 
> execute it for the purposes you have outlined.  That means for example that 
> if you're working on nmap scripts for $corp based on Nessus, that $corp has 
> an appropriate license to allow generation of packets from Nessus.  Note also 
> that depending on your locality, various EULA could have further impact on 
> how you are allowed to use the results (such as packet captures) from 
> commercial tools.
I don't have a whole lot of interest in writing scripts specifically for
$corp, I'm writing them for Nmap and releasing them under Nmap's
modified GPL license. That makes it incompatible with most other
licenses, which is why I ask.

I do have licenses for running Nessus and Foundstone, but obviously not
the sourcecode. Id on't know, however, how I'd be allowed to use results
(like packet caps) -- maybe I'll take a look at their licenses.

> Just for kicks, my colleague is the one who wrote the original code  
> (http://labs.portcullis.co.uk/application/ms08-067-check/) on which 
> Metasploits check for MS08-067 is based so I could probably get it ported 
> relatively easily.  OTOH, the check we (OpenVAS) use is based on earlier 
> packet analysis by one of the guys in India of traffic captured by myself 
> using a port of the original POC to Samba's RPC client 
> (http://www.nth-dimension.org.uk/blog.php?id=72).  Traditional reverse 
> engineering style really, one person examines the code, documents (in this 
> case as .pcap's) what happens and then someone else goes off and writes the 
> check.
I think that's more what I was getting at. When you say "examines the
code", do you mean the code for the exploit or for the vulnerability?
Like, if I log metasploit's exploit attempt, and base something on that,
am I then bound to metasploit's license? (I'm aware that Metasploit's
new license should make that moot, but I was using it as an example).

My current check for ms08-067 is based on one that I found in a .py
script floating around, and it's the same one that was incorporated into
Metasploit. Basically, "\AAA...A\..\n", iirc. I don't *think* I violated
any licenses using that, but licensing is hairy business, and vuln
checking seems pretty competitive.

> 
> Essentially, the checks *I* write are either for things I have discovered, 
> things where I can write a check based on an advisory, or things where I can 
> reverse engineer what is going on from packet level dumps.  For obvious 
> reasons, I stay clear of looking at how Nessus 3.x does things.
So do you figure that packet-level dumps are generally ok to use?
Generally, if I can't do it from a packet-level dump, then I probably
won't be able to do it from sourcecode.

On that topic, I don't know what your MS RPC checks are like right now
(that's what I'm working on), but what are your thoughts on basing Nmap
checks on OpenVAS scripts? Because Nmap uses a non-standard license
based on (but technically incompatible with) GPL, whoever owns the
copyright on the OpenVAS scripts would have to grant permission (at
least, that's how I understand it) before they can be included with
Nmap. Who could I talk to about that one? I'd definitely give credit and
all that. The other options is that I could release the scripts
separately from Nmap with a proper GPL, but there's no clean way of
doing that, right now, so I want to avoid it.

> 
> Cheers,
> Tim

Thanks for your time!
Ron

-- 
Ron Bowes
http://www.skullsecurity.org/
_______________________________________________
Openvas-discuss mailing list
[email protected]
http://lists.wald.intevation.org/mailman/listinfo/openvas-discuss

Reply via email to