Implementing Web Application Logging With Modsecurity

Implementing Web Application Logging with ModSecurity

Overview

ModSecurity is an open-source WAF engine that works with Apache, Nginx, and IIS. The OWASP Core Rule Set (CRS) provides generic attack detection rules covering SQL injection, XSS, RCE, LFI, and other OWASP Top 10 attacks. ModSecurity logs full request/response data in audit logs for forensic analysis and generates alerts that feed into SIEM platforms.

When to Use

• When deploying or configuring implementing web application logging with modsecurity capabilities in your environment
• When establishing security controls aligned to compliance requirements
• When building or improving security architecture for this domain
• When conducting security assessments that require this implementation

Common Misconfigurations & Verification

• Engine left in DetectionOnly: the most common silent failure. CRS matches, alerts fire, but nothing is ever blocked. Grep the active config for SecRuleEngine — if it's DetectionOnly (or the post-tuning switch to On never happened), the WAF is a logger, not a control.
• SecRequestBodyAccess Off or module not enabled per-vhost: POST/JSON bodies are never inspected, so SQLi/XSS in request bodies sail through even at high paranoia. Confirm SecRequestBodyAccess On and that the directive applies to the served vhost, not just the global context.
• Anomaly threshold too high / blocking rule (949110) excluded: CRS is scoring-based; if tx.inbound_anomaly_score_threshold is raised or SecRuleRemoveById 949110 (the blocking evaluator) was used to "fix false positives", individual rules match but the request is never denied.
• Over-broad SecRuleRemoveById/SecRuleRemoveByTag: disabling whole families (e.g. 942xxx) to silence FPs blinds SQLi detection entirely. Prefer targeted SecRuleUpdateTargetById exclusions for specific params.
• Paranoia Level 1 only: PL1 misses many evasions; obfuscated payloads pass. Note PL is a coverage/FP tradeoff, not a "set and forget" default.
• SecAuditEngine off or SecAuditLogParts missing request/response body parts (no IJ/E): alerts exist but the forensic body is gone, so triage is impossible; and logs that never reach the SIEM (local file only) mean no alerting.
• VERIFY it actually blocks: send a known-malicious request and require a 403, not a 200: curl -i "https://target/?q=union+select+1,2,3--" and curl -i "https://target/?x=<script>alert(1)</script>". Then confirm the matching audit entry exists (e.g. rule id 942100, anomaly score increment) in the audit log and that it was forwarded to the SIEM. A 200 with only a log line = detection only, control not enforced. Also test a request-body payload (curl -i -d "q=' OR 1=1-- -") to prove body inspection works, not just query strings.

Prerequisites

• Web server (Apache 2.4+ or Nginx) with ModSecurity v3 module
• OWASP CRS v4.x installed
• Log aggregation infrastructure (ELK, Splunk, or Wazuh)

Steps

1. Install ModSecurity and configure SecRuleEngine in DetectionOnly mode
2. Deploy OWASP CRS v4 and set paranoia level (PL1-PL4)
3. Configure SecAuditEngine for relevant-only logging
4. Tune false positives with SecRuleRemoveById and rule exclusions
5. Switch to blocking mode (SecRuleEngine On) after tuning period
6. Forward audit logs to SIEM for correlation and alerting

Expected Output

ModSecurity: Warning. Pattern match "(?:union\s+select)" [file "/etc/modsecurity/crs/rules/REQUEST-942-APPLICATION-ATTACK-SQLI.conf"] [line "45"] [id "942100"] [msg "SQL Injection Attack Detected via libinjection"] [severity "CRITICAL"]