Friday, September 11, 2026

Cloudflare Zero-Day Flaw Allows Attackers to Bypass Security and Access Any Host

A critical zero-day vulnerability in Cloudflare’s Web Application Firewall (WAF) allowed attackers to bypass security controls and directly access protected origin servers.

Security researchers from FearsOff discovered on October 9, 2025, that requests targeting a specific certificate-validation path could completely circumvent customer-configured WAF rules designed to block unauthorized traffic.

The Hidden Backdoor in Certificate Validation

The vulnerability exploited a path used by the Automatic Certificate Management Environment (ACME) protocol.

ACME automates SSL/TLS certificate issuance by having certificate authorities verify domain ownership through a specific URL pattern: /.well-known/acme-challenge/{token}.

This path is supposed to be a narrow, controlled gateway used only for certificate validation robots to verify domain ownership.

Cloudflare’s edge network contained a logic flaw in how it processed requests to this ACME challenge path.

When any request targeted /.well-known/acme-challenge/*, the system would turn off WAF security features to prevent interference with legitimate certificate validation.

However, the code failed to verify whether the token in the request actually matched an active challenge for that specific hostname.

This meant attackers could send arbitrary requests to the ACME path and bypass WAF protections entirely, reaching the origin server directly.

Error message
Error message (Source: fearsoff)

FearsOff researchers created demonstration environments to prove the severity of this bypass.

They configured WAF rules to block all traffic to test domains, including cf-php.fearsoff.org, cf-spring.fearsoff.org, and cf-nextjs.fearsoff.org.

Routine requests to these domains correctly returned Cloudflare block pages. However, identical requests targeting the ACME challenge path received responses directly from origin servers, bypassing all security controls.

The impact extended beyond simple access. On Spring Boot applications, researchers used servlet path traversal techniques (e.g., ..;/) to access sensitive actuator endpoints, such as /actuator/env, which exposed environment variables, database credentials, and API tokens.

On Next.js applications, server-side rendering logic exposes operational details that are never intended for public access. PHP applications with local file inclusion vulnerabilities became exploitable, allowing attackers to read system files like /etc/hosts.

Account-Level Rules Completely Ignored

To confirm this wasn’t limited to framework error pages, researchers configured account-level WAF rules to block requests containing a specific test header (X-middleware-subrequest). At typical paths, these rules correctly blocked flagged requests.

The identical request targeting the ACME challenge path was allowed through without evaluation.

This meant that entire classes of header-based attacks, including SQL injection via header concatenation, server-side request forgery via X-Forwarded-Host, cache poisoning, and HTTP method override tricks, could reach vulnerable origins.

Cloudflare deployed a permanent fix on October 27, 2025, modifying the ACME challenge logic only to turn off WAF features when the request token matches a valid, active challenge for that specific hostname.

The vulnerability timeline shows responsible disclosure: the report was submitted via HackerOne on October 9; vendor validation began on October 13; HackerOne triaged it on October 14; and the final fix was deployed on October 27.

Cloudflare confirmed that no customer action is required and stated they found no evidence of malicious exploitation.

The collaboration involved the Crypto.com Security Team for independent verification. It worked directly with Cloudflare CEO Matthew Prince to expedite the patch.

The bug highlights how maintenance paths intended for automation can become dangerous attack vectors when security controls apply inconsistently across different code paths.

Follow us on Google NewsLinkedIn, and X to Get Instant Updates ancd Set GBH as a Preferred Source in Google.

Divya
Divya
Divya is a Senior Journalist at GBhackers covering Cyber Attacks, Threats, Breaches, Vulnerabilities and other happenings in the cyber world.

Hot this week

How To Access Dark Web Anonymously and know its Secretive and Mysterious Activities

What is Deep Web The deep web, invisible web, or...

How to Build and Run a Security Operations Center (SOC Guide) – 2023

Today’s Cyber security operations center (CSOC) should have everything...

Russian Hackers Bypass EDR to Deliver a Weaponized TeamViewer Component

TeamViewer's popularity and remote access capabilities make it an...

Web Server Penetration Testing Checklist – 2026

Web server pentesting is performed under three significant categories: identity,...

ATM Penetration Testing – Advanced Testing Methods to Find The Vulnerabilities

ATM Penetration testing, Hackers have found different approaches to...

Researchers Uncover 10,000+ Malware Loaders Behind YouTube and SEO Poisoning Campaign

A long-running pay-per-install (PPI) operation that used YouTube gaming...

VLC Media Player Flaws Let Attackers Corrupt Memory and Leak Sensitive Data

Two security vulnerabilities in VLC media player versions 3.0.0...

CISA Adds Exploited MikroTik RouterOS Flaws to Security Alert

The U.S. Cybersecurity and Infrastructure Security Agency (CISA) has...

cPanel Urges Users to Patch ConfigServer Firewall Remote Code Execution Flaw

A recently disclosed vulnerability in ConfigServer Security & Firewall...

Hackers Weaponize AI Safety Guardrails to Hide Malware From LLM-Powered Security Scanners

Threat actors are adapting malware not only for conventional...

Hackers Exploit JFrog Artifactory Flaws to Bypass Authentication and Gain Admin Access

Threat actors are actively exploiting three vulnerabilities in JFrog...

Related Articles

Recent News