Prompt
Fortinet FortiWeb is blocking too much traffic
Latest observation
Oct 29, 2026
OpenAI APIWeb search: off
If FortiWeb is blocking legitimate traffic, the goal is usually to identify which protection is firing, confirm whether it’s a false positive, and then tune the policy safely rather than just disabling protections.
Quick things to check
-
Find the block reason
- Review FortiWeb logs for the blocked requests.
- Note the policy, signature/attack category, source IP, URL, and rule name.
-
Determine if it’s a false positive
- Compare blocked requests with normal application behavior.
- Check if the app sends unusual query strings, JSON bodies, special characters, large headers, or file uploads.
-
Tune the relevant protection Common causes:
- OWASP signature/attack signatures
- Parameter validation
- Bot mitigation
- CSRF protection
- IP reputation / geo blocking
- Rate limiting
- File upload controls
-
Use exceptions instead of global disable
- Exempt specific URLs, parameters, clients, or source IPs.
- Narrow exceptions to the smallest possible scope.
-
Test in monitor/log-only mode if possible
- For the affected policy, temporarily switch certain checks to monitoring so you can observe impact before enforcing changes.
-
Check learning/auto-learning settings
- If FortiWeb is using learned profiles, retrain or update the baseline after app changes.
Practical troubleshooting flow
- Pick one blocked request from the logs.
- Identify the exact rule that matched.
- Reproduce the request in a test environment.
- Decide whether to:
- relax the rule,
- add an exception,
- adjust a threshold,
- or keep the block if it’s real malicious traffic.
Safe tuning suggestions
- Whitelist by URL path + parameter, not entire site, when possible.
- Raise thresholds gradually for rate-based protections.
- For APIs, verify JSON schema/expected methods/headers.
- For uploads, ensure allowed file types and size limits match the application.
If you want, I can help you narrow it down
Send me any of these and I’ll help interpret them:
- the FortiWeb log entry
- the attack/signature name
- the URL or parameter being blocked
- whether this is a web app, API, or upload form