SQL Injection (CWE-89)

SQL Injection (CWE-89)

CWE-89 describes SQL Injection as follows:

“The software constructs all or part of an SQL command using externally-influenced input from an upstream component, but it does not neutralize or incorrectly neutralizes special elements that could modify the intended SQL command when it is sent to a downstream component.”

Any SQL-Injection attack basically consists of insertion (or “injection”) of malicious code within the SQL command executed by the app.

Effects of such malicious code injections can be unpredictable, depending on attacker’s intelligence and SQL-interpreter’s characteristics, but most common are:

• Read/modify sensitive data
• Execute administrative operations
• Execute commands on underlying OS

The most basic SQL-injection attack is based on exploiting a dynamically constructed SQL query based on input data.

Let’s suppose an app that display user’s data based on user’s name as typed from the application user in a web form (for example).

Dynamically constructed SQL in app code might be something as

"select * from users where name = '" + userName + "'";

userName is user-supplied data that is directly inserted into the query and it will be sent to SQL-engine to be executed.

Let’s imagine the result when the attacker supplies next text

• ‘ or '1'='1

In this case, sql-engine will return all users’ data because 1=1 will always be TRUE.

This attack would the attacker be able to access users’ data (involving a privacy breach), but consequences can be more serious combining query chaining with administrative commands such as

• Smith’;drop table users; truncate audit_log;--

In this case, the attacker would be able to delete the users table or truncate system tables- Everything depends on the concrete case, but “the door is open” and, as you can imagine, imagination is the limit.  How does the attacker know the app database tables? Depending on the error messages the application produces when a SQL-injection attack happens, a smart attacker might be “inferring” database structure information from the error page. It’s a matter of patience to discover useful information.

You could be thinking on a common app error mgmt approach consisting on providing a generic error page, not displaying any exploitable information about app internals.

Even in this case, the app is still vulnerable to sql-injection. In these cases, attackers use a technique known as “Blind SQL Injection”. This hacking technique is based on asking the database questions and determines the answer based on the applications response. This attack is often used when the web application is configured to show generic error messages, but has not mitigated the code that is vulnerable to SQL injection.

Some variants of SQL-Injection apply to specific frameworks or conditions:

• CWE-564: SQL Injection: Hibernate
• CWE-566: Authorization Bypass Through User-Controlled SQL Primary Key

SQL Injection (CWE-89) coverage by Kiuwan

In Kiuwan, you can search rules covering SQL-Injection (CWE-89)  filtering by Vulnerability Type (“Injection”) and/or by CWE tag (“CWE:89”).  

Kiuwan incorporates next rules for SQL-Injection (CWE-89) for the following languages. Please, visit the documentation page for every rule to obtain detailed information on functionality, coverage, parameterization, remediation, example codes, etc.