Understanding Common Web Application Attacks and the OWASP Top 10

Nowadays, application development is moving more and more onto the Web. The Web hosts entire productivity suites such as Google Docs, calculators, email, storage, maps, weather and news — everything we need in our daily lives. Our mobile phones are useless without the Internet since nearly all mobile applications connect to the cloud, storing our pictures, usernames and passwords and private information. Even our home devices are now connecting to the Web, with Internet of Things platforms such as Wink that allow users to dim their house lights right from their mobile phone.

Application Attacks

The application layer is the hardest to defend. The vulnerabilities encountered here often rely on complex user input scenarios that are hard to define with an intrusion detection signature. This layer is also the most accessible and the most exposed to the outside world. For the application to function, it must be accessible over Port 80 (HTTP) or Port 443 (HTTPS).

In the diagram below, the Web application is completely exposed to the outside world in spite of network defenses such as firewalls and intrusion prevention systems:

In 2014, SQL injections, a type of application attack, were responsible for 8.1 percent of all data breaches. That makes it the third most used type of attack, behind malware and distributed denial‑of‑service attacks. You will also find on the list other common application attacks such as security misconfiguration, using components with known vulnerabilities and cross‑site scripting. Attackers were able to manipulate application input and obtain confidential data without being detected by network defense systems.

Most vulnerabilities found in the proprietary code of Web applications are unknown to security defense systems; these are called zero‑day vulnerabilities. This is because these vulnerabilities are specific to each application and have never been known before. A skilled attacker can easily find these vulnerabilities and exploit the issue without being detected.

The best defense against these attacks is to develop secure applications. Developers must be aware of how application attacks work and build software defenses right into their applications.

Educating and informing developers about application vulnerabilities is the goal of the Open Web Application Security Project (OWASP). The organization has put together a list of the 10 most common application attacks. This list is renewed every three years, with the latest refresh in 2013.

10. Unvalidated Redirects and Forwards

This category of vulnerabilities is used in phishing attacks in which the victim is tricked into navigating to a malicious site. Attackers can manipulate the URLs of a trusted site to redirect to an unwanted location.

9. Using Components With Known Vulnerabilities

This category is about using unpatched third‑party components. Attackers can easily exploit old third‑party components because their vulnerabilities have been publicized, and tools and proof of concepts often allow cyber‑criminals to take advantage of these flaws with ease.

8. Cross‑Site Request Forgery

This type of attack is used in conjunction with social engineering. It allows attackers to trick users into performing actions without their knowledge.

7. Missing Function Level Access Control

This category covers situations in which higher‑privilege functionality is hidden from a lower‑privilege or unauthenticated user rather than being enforced through access controls.

6. Sensitive Data Exposure

This category deals with a lack of data encryption in transport and at rest. If your Web applications do not properly protect sensitive data, such as credit cards or authentication credentials, attackers can steal or modify the data to conduct credit‑card fraud, identity theft or other crimes.

5. Security Misconfiguration

Moving into the top five, we’re exploring another extremely dangerous category of flaws that deals with the incorrect misconfiguration of the server or of the application itself.

4. Insecure Direct Object References

This type of insecure direct object reference allows attackers to obtain data from the server by manipulating file names, effectively performing path‑traversal attacks.

3. Cross‑Site Scripting

Cross‑site scripting is a type of vulnerability that lets attackers insert JavaScript in the pages of a trusted site, enabling them to alter site content, steal credentials, or perform other malicious actions.

2. Broken Authentication and Session Management

Several programming flaws allow attackers to bypass the authentication methods that are used by an application, leading to unauthorized access.

1. Injection

Injections let attackers modify a back‑end statement or command through unsanitized user input. Examples include SQL injection attacks that can expose entire user tables, including passwords.

The IBM X‑Force Ethical Hacking Team has created a video series that demonstrates each OWASP Top 10 attack, explains prevention techniques, and shows how to use automated tools to test for vulnerabilities. The series, originally for internal use, is now publicly available.