Metasploit mailing list archives

Re: kind of off topic but not much -- Windows XP connects to mail server before logging in MBR exploit?


From: Joshua Smith <lazydj98 () gmail com>
Date: Tue, 2 Aug 2011 22:59:47 -0400

servifythis will also do that

On Tue, Aug 2, 2011 at 9:42 PM, Average SecurityGuy <
averagesecurityguy () gmail com> wrote:

You could also set up the meterpreter executable as a service. You would
need srvany.exe from Microsoft. It might be difficult to find but it is
available. Srvany will take any executable and turn into a service.


On Tue, Aug 2, 2011 at 5:32 PM, Robin Wood <robin () digininja org> wrote:

On 2 August 2011 22:27, Chip <jeffschips () gmail com> wrote:
On 8/2/2011 4:49 PM, Robin Wood wrote:

On 2 August 2011 17:53, Chip<jeffschips () gmail com>  wrote:

I know this is not entirely along the lines of metasploit but knowing
that
almost all who subscribe to this list are network gurus I thought I
would
post my question here -- it may, in fact, have something to do with an
exploit affecting the MBR of a host machine.

I have a windows xp machine which connects to a mail server with
encrypted
traffic before anyone logs in, right after start up and when the
windows
xp
login splash screen comes up.   I know this is the case because I
attached
an inline tap to the network to watch traffic and see this activity
every
time the machine starts up.

Reading a lot recently about MBR exploits I'm wondering how I can
track
down
what culprit is doing this?  Since it occurs ONLY prior to login, I
cannot
look at tasklist or netstat in a shell and see anything.  I'm hoping
someone
on this list could advise either in the list or off the list.

I do have some captured packets of the activity, but it is encrypted
TLS
traffic.

Create a meterpreter binary and run it on the machine with psexec
you've then got full access to the machine and can watch what happens
when you log in at the keyboard.

Have you checked who owns the mail server the traffic is going to? Is
it a legit one or some dodgy backstreet one? If it goes via domain
name you could try doing a DNS MITM to send it to a machine you own,
it might not validate that the certificate is the correct one so you
would be able to capture the traffic and decode it. You can do the
same if it does IP you just need to setup routing and IPs correctly.

Robin

Very, very interesting option and since it does do domain name lookup --
I
can see that in the network tap -- I find that might be the way to go!
 My
question is if I used the meterpreter binary psexec method, will that
load a
session prior to login, because this malware seems to drop silent once
login
occurs.  And if I get a meterpreter session, are you then suggesting I
run
tasklist or netstat then to tease out the culprit?

Thanks.

It depends on what it counts as a login, it may only be watching for
one from the GUI, if so psexec may not be noticed. If it is then just
install an old ftp server from exploit-db and exploit that to get your
Meterpreter shell, there is no login then.

And yes, watch with tasklist and other tools. I'm no good with
forensics, I'm sure there are others on this and other lists who will
be able to tell you what to look for.

Robin
_______________________________________________
https://mail.metasploit.com/mailman/listinfo/framework




--
Stephen Haywood
Security Consultant

Twitter: @averagesecguy
Blog: averagesecurityguy.info <http://averagesecurityguy.info>


_______________________________________________
https://mail.metasploit.com/mailman/listinfo/framework




-- 
- Josh
_______________________________________________
https://mail.metasploit.com/mailman/listinfo/framework

Current thread: