SQL Injection

web HIGH

SQL injection exploits the way applications construct database queries by mixing user input directly into SQL statements.

How It Works

SQL injection exploits the way applications construct database queries by mixing user input directly into SQL statements. When developers concatenate untrusted data into queries without proper sanitization, attackers can inject SQL syntax that changes the query's logic. For example, entering ' OR '1'='1 into a login form might transform SELECT * FROM users WHERE username='input' into a query that always returns true, bypassing authentication.

Attackers follow a methodical process: first probing input fields with special characters like quotes or semicolons to trigger database errors, then identifying whether the application is vulnerable. Once confirmed, they escalate by injecting commands to extract data (UNION-based attacks to merge results from other tables), manipulate records, or probe the database structure. Blind SQL injection variants work without visible error messages—boolean-based attacks infer data by observing application behavior changes, while time-based attacks use database sleep functions to confirm successful injection through response delays.

Advanced scenarios include second-order injection, where malicious input is stored in the database and later executed in a different context, and out-of-band attacks that exfiltrate data through DNS queries or HTTP requests when direct data retrieval isn't possible. Some database systems enable attackers to execute operating system commands through built-in functions like MySQL's LOAD_FILE or SQL Server's xp_cmdshell, escalating from database compromise to full server control.

Impact

Complete data breach — extraction of entire database contents including credentials, personal information, and proprietary data
Authentication bypass — logging in as any user without knowing passwords
Data manipulation — unauthorized modification or deletion of critical records
Privilege escalation — granting administrative rights to attacker-controlled accounts
Remote code execution — leveraging database features to run operating system commands and compromise the underlying server
Lateral movement — using compromised database credentials to access other connected systems

Real-World Examples

FreePBX's CVE-2025-66039 demonstrated a complete attack chain where SQL injection across 11 parameters in four different endpoints allowed attackers to write malicious entries into the cron_jobs table. When the system's scheduler executed these entries, the injected SQL transformed into operating system commands, granting full server control. The vulnerability required no authentication, making it immediately exploitable.

E-commerce platforms have suffered massive breaches through shopping cart SQL injection, where attackers inserted skimming code into stored procedures that executed during checkout, harvesting credit card data from thousands of transactions. Healthcare systems have been compromised through patient portal vulnerabilities, exposing millions of medical records when attackers injected UNION queries to merge data from supposedly isolated tables.

Mitigation

Parameterized queries (prepared statements) — separates SQL logic from data, making injection syntactically impossible
Object-Relational Mapping (ORM) frameworks — abstracts database interactions with built-in protections when used correctly
Strict input validation — whitelist acceptable characters and formats, reject suspicious patterns
Least privilege database accounts — applications should use credentials with minimal necessary permissions
Web Application Firewall (WAF) — detects and blocks common injection patterns as a secondary defense layer
Database activity monitoring — alerts on unusual query patterns or privilege escalation attempts

Recent CVEs (16111)

CVE-2023-31619

EPSS 1% CVSS 7.5

HIGH POC This Week

An issue in the sch_name_to_object component of openlink virtuoso-opensource v7.2.9 allows attackers to cause a Denial of Service (DoS) via crafted SQL statements. Rated high severity (CVSS 7.5), this vulnerability is remotely exploitable, no authentication required, low attack complexity. Public exploit code available and no vendor patch available.