Penetration testing methods

WHAT IS A PENETRATION TEST?

A penetration test is a subclass of ethical hacking; it comprises a set of methods and procedures that aim at testing/protecting an organization’s security. The penetration tests prove helpful in finding vulnerabilities in an organization and check whether an attacker will be able to exploit them to gain unauthorized access to an asset.

VULNERABILITY ASSESSMENTS VERSUS PENETRATION TEST

Oftentimes, a vulnerability assessment is confused with a penetration test; however, these terms have completely different meanings. In a vulnerability assessment, our goal is to figure out all the vulnerabilities in an asset and document them accordingly.
In a penetration test, however, we need to simulate as an attacker to see if we are actually able to exploit a vulnerability and document the vulnerabilities that were exploited and the ones that turned out to be false-positive.
Preengagement
Before you start doing a penetration test, there is whole lot of things you need to discuss with
clients. This is the phase where both the customer and a representative from your company would
sit down and discuss about the legal requirements and the “rules of engagement.”4  Ethical Hacking and Penetration Testing Guide.

RULES OF ENGAGEMENT

Every penetration test you do would comprise of a rules of engagement, which basically defines how a penetration test would be laid out, what methodology would be used, the start and end dates, the milestones, the goals of the penetration test, the liabilities and responsibilities, etc. All of them have to be mutually agreed upon by both the customer and the representative before the penetration test is started. Following are important requirements that are present in almost every ROE:
 A proper “permission to hack” and a “nondisclosure” agreement should be signed by both the parties.
 The scope of the engagement and what part of the organization must be tested.
 The project duration including both the start and the end date.
 The methodology to be used for conducting a penetration test.
 The goals of a penetration test.
 The allowed and disallowed techniques, whether denial-of-service testing should be per-
formed or not.
 The liabilities and responsibilities, which are decided ahead of time. As a penetration tester you might break into something that should not be accessible, causing a denial of service; also, you might access sensitive information such as credit cards. Therefore, the liabilities should be defined prior to the engagement.

Milestones
Before starting a penetration test, it’s good practice to set up milestones so that your project is delivered as per the dates given in the rules of engagement.
You can use either a GANTT chart or a website like Basecamp that helps you set up milestones to keep track of your progress. The following is a chart that defines the milestones followed by the date they should be accomplished.

Penetration Testing Methodologies
In every penetration test, methodology and the reporting are the most important steps. Let’s first talk about the methodology. There are several different types of penetration testing methodologies that address how a penetration test should be performed. Some of them are discussed in brief next.

OSSTMM

An open-source security testing methodology manual (OSSTMM) basically includes almost all the steps involved in a penetration test. The methodology employed for penetration test is concise yet it’s a cumbersome process which makes it difficult to implement it in our everyday life. Penetration tests, despite being tedious, demands a great deal of money out of company’s budgets for their completion which often are not met by a large number of organizations.
NIST
NIST, on the other hand, is more comprehensive than OSSTMM, and it’s something that you would be able to apply on a daily basis and in short engagements. The screenshot indicates the four steps of the methodology, namely, planning, discovery, attack, and reporting.
The testing starts with the planning phase, where how the engagement is going to be performed is decided upon. This is followed by the discovery phase, which is divided into two parts—the first part includes information gathering, network scanning, service identification, and OS detection, and the second part involves vulnerability assessment.
After the discovery phase comes the attack phase, which is the heart of every penetration test. If you are able to compromise a target and a new host is discovered, in case the system is dual-homed or is connected with multiple interfaces, you would go back to step 2, that is, discovery, and repeat it until no targets are left. The indicating arrows in the block phase and the attack phase to the reporting phase indicate that you plan something and you report it—you attack a target and report the results.
The organization also has a more detailed version of the chart discussed earlier, which actually explains more about the attack phase. It consists of things such as “gaining access,” “escalating privileges,” “system browsing,” and “install additional tools.”

OWASP
As you might have noticed, both the methodologies focused more on performing a network penetration test rather than something specifically built for testing web applications. The OWASP testing methodology is what we follow for all “application penetration tests” we do here at the RHA InfoSEC. The OWASP testing guide basically contains almost everything that you would test a web application for. The methodology is comprehensive and is designed by some of the best web application security researchers.

CATEGORIES OF PENETRATION TEST

When the scope of the penetration test is defined, the category/type of the penetration test engagement is also defined along with it. The entire penetration test can be Black Box, White Box, or Gray Box depending upon what the organization wants to test and how it wants the security paradigm to be tested.
Black Box
A black box penetration test is where little or no information is provided about the specifiedtarget.In the case of a network penetration test this means that the target’s DMZ, target operating system, server version, etc., will not be provided; the only thing that will be provided is the IP ranges that you would test. In the case of a web application penetration test, the source code of the web application will not be provided. This is a very common scenario that you will encounter when performing an external penetration test.
White Box
A white box penetration test is where almost all the information about the target is provided. In the case of a network penetration test, information on the application running, the corresponding versions, operating system, etc., are provided. In the case of a web application penetration test,the application’s source code is provided, enabling us to perform the static/dynamic “source code analysis.” This scenario is very common in internal/onsite penetration tests, since organizations are concerned about leakage of information.
Gray Box
In a gray box test, some information is provided and some hidden. In the case of a network penetration test, the organization provides the names of the application running behind an IP; however, it doesn’t disclose the exact version of the services running. In the case of a web application penetration test, some extra information, such as test accounts, back end server, and databases, is provided.

TYPES OF PENETRATION TESTS

There are several types of penetration tests; however, the following are the ones most commonly performed:-
Network Penetration Test
In a network penetration test, you would be testing a network environment for potential security vulnerabilities and threats. This test is divided into two categories: external and internal penetration tests.An external penetration test would involve testing the public IP addresses, whereas in an internal test, you can become part of an internal network and test that network. You may be provided VPN access to the network or would have to physically go to the work environment for the penetration test depending upon the engagement rules that were defined prior to conducting the test.
Web Application Penetration Test
Web application penetration test is very common nowadays, since your application hosts critical data such as credit card numbers, usernames, and passwords; therefore this type of penetration test has become more common than the network penetration test.
Mobile Application Penetration Test
The mobile application penetration test is the newest type of penetration test that has become common since almost every organization uses Android- and iOS-based mobile applications to provide services to its customers. Therefore, organizations want to make sure that their mobile applications are secure enough for users to rely on when providing personal information when using such applications.
Social Engineering Penetration Test
A social engineering penetration test can be part of a network penetration test. In a social engineering penetration test the organization may ask you to attack its users. This is where you use speared phishing attacks and browser exploits to trick a user into doing things they did not intend to do.
Physical Penetration Test
A physical penetration test is what you would rarely be doing in your career as a penetration tester. In a physical penetration test, you would be asked to walk into the organization’s building physically and test physical security controls such as locks and RFID mechanisms