RE: Possilbe New Arp DoS - dosprmwin.exe

[Harlan Carvey <keydet89 () yahoo com>]

David,

Thanks for the response...

> What did you do to be able to accomplish this?  I
> think it would be
> educational to the list if you could elaborate just
> a little on how you
> went about doing this.
>
> > > We used Ethereal to detect the traffic. We found
> > > dosprmwin by comparing 
> > > the process list on machines that were
> > > broadcasting and then used process 
> > > of elimination.

Hhhmmm...not the most conclusive approach, but hey, if
it worked...  ;-)
 
> I'm still not entirely clear on your line of
> reasoning here...what am I
> missing?  I get the first two sentences, but if you
> were able to tie the
> activity to a particular KB article as you did,
> wouldn't that then tell
> how you the program was propagating?  
>
> > > That just told us which exploit it was using to
> > > install itself. We still
> > > aren't sure if all of these users visited
> > > an infected web site or if just 
> > > one user did and then the worm
> > > propagated using the arp broadcasts.

Okay, now I'm *really* confused!  I'm not the smartest
guy in the world about these things, but the
vulnerability you mentioned is KB840374,
"Vulnerability in Help and Support Center Could Allow
Remote Code Execution".  Reading the article, this has
to do with URL validation, not ARP.

Also, can you elaborate on how a worm would spread via
ARP?  

> > We were able to solve the problem by running
> > Windows Updates and then 
> > stopping the dosprmwin.exe process and removing
> > the file from 
> > \windows\system32. The file was marked as hidden,
> > read-only, and system.

You've got to admit, though...from your above
explanation, it's not clear that the updates have
anything to do with the activity stopping.  After all,
by your admission, you installed the updates, and then
killed the process.  Killing the process alone would
be enough to stop the ARP broadcasts, if they were in
fact a result of something to do with the executable
you found.

> Do you have a copy of this file available for
> someone to look at?
>
> > > I do not. 

I'd like to ask that you also provide some captures of
the activity you saw, as well.  I think having others
look at the file, as well as the ARP broadcasts (to
see what was being requested) would be very
beneficial.

> > Running fport or netstat ?a while the process was
> > still running would 
> > show a large number of listening TCP ports. After
> > the process was 
> > killed, these listening ports disappeared. 
>
> Does this mean that the output of fport explicitly
> tied the open ports to this executable file?  
>
> > > Yes. 

Okay, this is interesting.  A couple of things...first
of all...what were the ports?  Were they sequential? 
Where'd they start?  What was the range?

Second, it would be interesting to see how these open
TCP ports related to the ARP broadcasts you were
seeing.

Thanks,

Harlan

---------------------------------------------------------------------------
Ethical Hacking at the InfoSec Institute. Mention this ad and get $545 off 
any course! All of our class sizes are guaranteed to be 10 students or less 
to facilitate one-on-one interaction with one of our expert instructors. 
Attend a course taught by an expert instructor with years of in-the-field 
pen testing experience in our state of the art hacking lab. Master the skills 
of an Ethical Hacker to better assess the security of your organization. 
Visit us at: 
http://www.infosecinstitute.com/courses/ethical_hacking_training.html
----------------------------------------------------------------------------