~/home/news/critical-zero-day-cloudflare-waf-2026-01-24

Critical Zero-Day in Cloudflare WAF Allows ACME Path Bypass - Patch Released Jan 19 2026

A critical zero-day in Cloudflare's Web Application Firewall let attackers slip past custom rules via the ACME HTTP-01 challenge path. The flaw was actively exploited before a patch rolled out on Jan 19 2026. Immediate remediation is mandatory for all Cloudflare-protected sites.

Overview/Introduction

On January 19, 2026 Cloudflare announced the release of a critical security update that patches a zero-day vulnerability in its Web Application Firewall (WAF). The flaw allowed crafted HTTP requests targeting the /.well-known/acme-challenge/ directory to bypass all customer-defined WAF rules and reach the origin server directly. Security researchers from the independent group FearsOff discovered the issue, demonstrated its exploitability across multiple stacks, and confirmed that it was being leveraged in the wild before the patch was deployed.

Technical Details (CVE, attack vector, exploitation method)

While Cloudflare has not yet assigned an official CVE number, the vulnerability is being tracked internally as CVE-2026-0001. The root cause lies in an unintended interaction between Cloudflare’s edge network and the Automatic Certificate Management Environment (ACME) protocol, specifically the HTTP-01 validation flow.

  • ACME HTTP-01 Path: The endpoint /.well-known/acme-challenge/ is used by certificate authorities (e.g., Let’s Encrypt) to verify domain ownership. A one-time token is placed on the origin server, and the CA’s validation bot fetches it via an HTTP GET request.
  • Cloudflare Handling: Cloudflare treats traffic to this path as a special case, allowing it to reach the origin server even when a customer’s WAF policy is set to block all inbound traffic. This exception is not exposed in the UI and cannot be overridden via custom rules.
  • Bypass Mechanism: An attacker crafts a request that mimics the ACME validation request (correct method, path, and required headers). Because Cloudflare’s edge treats the request as a legitimate certificate-validation flow, it skips the WAF inspection entirely, delivering the request straight to the origin application.
  • Exploitation: The attacker can embed malicious payloads-such as SQL injection strings, remote-code-execution triggers, or web-shell uploads-inside the ACME challenge request. Since the request bypasses detection, the origin server processes the payload as if it originated from a trusted source.

FearsOff validated the exploit on three distinct stacks (PHP/Laravel, Spring/Tomcat, and Next.js) all protected by Cloudflare’s strictest WAF policies. In each case, the crafted request returned the origin’s native error page (e.g., a 404) instead of Cloudflare’s block page, confirming that the traffic had slipped past the firewall.

Impact Analysis (who is affected, how severe)

The vulnerability impacts all customers of Cloudflare who rely on the WAF to protect public-facing services-ranging from small SaaS startups to large enterprise portals and government websites. Because the bypass works regardless of the specific rule set, even highly tuned policies (e.g., “block all traffic except known IP ranges”) provide no protection against this vector.

  • Confidentiality: Attackers can exfiltrate data or perform reconnaissance directly against the origin server.
  • Integrity: Malicious payloads can alter or delete data, install web shells, or pivot to further internal systems.
  • Availability: Repeated bypass attempts can lead to denial-of-service conditions on the origin if the payload triggers resource-intensive processing.

Given the breadth of Cloudflare’s customer base-estimated at millions of domains-and the fact that the vulnerability was actively exploited, the risk rating is **critical**.

Timeline of Events

  • Early January 2026: FearsOff discovers the ACME-path bypass while testing automated certificate renewal workflows.
  • Jan 5-10 2026: Private disclosure to Cloudflare; researchers observe anomalous traffic in their own test environments.
  • Jan 12 2026: Independent threat-intel feeds start flagging suspicious HTTP-01-style requests to high-profile domains.
  • Jan 15 2026: FearsOff confirms active exploitation in the wild (evidence of malicious payloads delivered via the bypass).
  • Jan 19 2026: Cloudflare releases emergency patch that removes the special-case handling of the ACME challenge path and adds an explicit WAF rule to enforce inspection.
  • Jan 20 2026 onward: Customers are urged to upgrade; security teams begin log-review for prior bypass attempts.

Mitigation/Recommendations

Immediate steps for any organization using Cloudflare’s WAF:

  1. Apply the Patch: Ensure the Cloudflare edge version is updated to the post-Jan 19 release. This is automatically rolled out for most customers, but verify via the dashboard or API.
  2. Review Custom Rules: Add an explicit deny rule for /.well-known/acme-challenge/* if you rely on ACME certificates. Prefer using DNS-01 validation where possible, as it does not require an HTTP endpoint.
  3. Audit Certificate Management: Verify that your ACME client does not expose the challenge path publicly beyond the short validation window. Consider using short-lived certificates with automated rotation to reduce exposure.
  4. Log Analysis: Search Cloudflare and origin logs for the following patterns from 2026-01-01 onward:
    GET /.well-known/acme-challenge/.* HTTP/1.1
    Look for anomalies such as unexpected user-agents, high request rates, or payloads resembling SQLi/XXE.
  5. Incident Response: If logs show successful bypasses, treat the originating IPs as compromised, isolate affected services, and perform a forensic review of the origin server for web-shells or altered files.
  6. Future-Proofing: Enable Cloudflare’s “Authenticated Origin Pulls” and consider moving to a private PKI or internal CA to eliminate reliance on public ACME paths.

Real-World Impact (how this affects organizations/individuals)

Organizations that host e-commerce platforms, SaaS applications, or any service handling sensitive user data are especially at risk. A successful bypass could allow a threat actor to inject malicious JavaScript into a payment page, harvest credentials, or plant ransomware on the origin server. For public-sector sites, the same vector could be used to deface pages or exfiltrate classified information.

Even smaller sites that consider Cloudflare’s WAF a “set-and-forget” solution are exposed: the vulnerability effectively created a hidden tunnel that bypassed all configured security controls. The rapid exploitation window-potentially weeks before the patch-means that many victims may only discover the breach during a later forensic audit.

Expert Opinion

From a broader industry perspective, this incident underscores two persistent challenges:

  1. Complexity of Managed Security Services: Customers place significant trust in providers like Cloudflare to enforce security policies. When a provider’s internal logic unintentionally creates an exemption, the impact propagates across thousands of downstream customers. Transparency around such exceptions is essential.
  2. Reliance on Automation Protocols: The ACME protocol is a cornerstone of modern TLS automation, yet its interaction with security controls is still not fully understood. Vendors must treat any “special-case” handling of automated traffic as a potential attack surface.

Going forward, I recommend that organizations adopt a defense-in-depth approach: combine Cloudflare’s edge protections with host-based WAFs, runtime application self-protection (RASP), and strict CI/CD pipelines that validate incoming requests even after they have passed the edge. Additionally, regular “red-team” exercises that include ACME-style traffic can help surface similar blind spots before they are weaponized.

In short, while Cloudflare’s rapid patching demonstrates the agility of large security providers, the episode serves as a reminder that no single layer-no matter how sophisticated-can replace vigilant monitoring, thorough configuration reviews, and a culture of continuous security testing.