SecPod

Learn Search

Search across all Learn content

← Back to Security Research
New MadeYouReset Exploit Bypasses HTTP/2 DoS Protections

New MadeYouReset Exploit Bypasses HTTP/2 DoS Protections

A novel attack technique named MadeYouReset has been discovered, targeting multiple implementations of the HTTP/2 protocol. This flaw, sitting at a comfortable 7.5 on the CVSS scale, allows attackers to bypass existing mitigations and launch significant denial-of-service (DoS) attacks. The vulnerabi...

Aug 17, 2025By Meghana Raatni4 min read

A novel attack technique named MadeYouReset has been discovered, targeting multiple implementations of the HTTP/2 protocol. This flaw, sitting at a comfortable 7.5 on the CVSS scale, allows attackers to bypass existing mitigations and launch significant denial-of-service (DoS) attacks. The vulnerability is especially concerning because it circumvents the standard server-imposed limit of 100 concurrent HTTP/2 requests per TCP connection, which is designed to prevent DoS attacks.

Vulnerability Details

The CVE-2025-8671 identifier has been assigned to this generic vulnerability, which impacts several products, including Apache Tomcat (CVE-2025-48989), F5 BIG-IP (CVE-2025-54500), and Netty (CVE-2025-55163). The MadeYouReset vulnerability allows attackers to exploit malformed HTTP/2 control frames and send thousands of requests, overwhelming the server and causing denial-of-service conditions for legitimate users. In some cases, this can cause significant increases in CPU usage and even lead to out-of-memory crashes.

This “MadeYouReset” attack is an evolution of the “Rapid Reset” attack from 2023 but circumvents the mitigations that were deployed against the earlier threat by manipulating servers into stream resets via crafted malformed frames.

Root Cause

The root cause of MadeYouReset lies in a mismatch between the HTTP/2 specifications and the internal architectures of many real-world web servers. This discrepancy leads to resource exhaustion, which attackers can exploit to trigger a DoS attack. The vulnerability leverages the fact that the RST_STREAM frame is used both for client-initiated cancellation and to signal stream errors.

Proof of Concept (PoC)

To trigger the RST_STREAM frames on the server, attackers can use the following primitives:

block
1. WINDOW_UPDATE frame with an increment of 0
2. PRIORITY frame whose length is not 5 (the only valid length for it)
3. PRIORITY frame that makes a stream dependent on itself
4. WINDOW_UPDATE frame with an increment that makes the window exceed 2^31 - 1 (the largest window size allowed)
5. HEADERS frame sent after the client has closed the stream (via the END_STREAM flag)
6. DATA frame sent after the client has closed the stream (via the END_STREAM flag)

By sending carefully crafted frames that trigger protocol violations, the server is prompted to reset the stream by issuing an RST_STREAM.

Affected Products

The vulnerability affects multiple products:

F5 BIG-IP (CVE-2025-54500):

  • BIG-IP 17.x (versions 17.5.0-17.5.1 and 17.1.0-17.1.2)
  • BIG-IP 16.x (versions 16.1.0-16.1.6)
  • BIG-IP 15.x (versions 15.1.0-15.1.10)
  • BIG-IP Next products (version 20.3.0 and various SPK, CNF, and Kubernetes implementations)

Apache Tomcat (CVE-2025-48989):

  • Tomcat 11.0.0-M1 through 11.0.9
  • Tomcat 10.1.0-M1 through 10.1.43
  • Tomcat 9.0.0.M1 through 9.0.107
  • Older, EOL versions might also be affected

Netty (CVE-2025-55163):

  • Netty prior to 4.1.124.Final
  • Netty prior to 4.2.4.Final

Other affected projects include h2o and swift-nio-http2.

Tactics, Techniques, and Procedures (TTPs)

Attackers exploit malformed HTTP/2 control frames to bypass the maximum concurrent streams limit. By crafting specific invalid control frames or violating protocol sequencing, attackers can force servers to generate RST_STREAM frames while backend systems continue processing requests.

  • TA0040 – Impact: Exploit the vulnerability to cause a denial-of-service condition.

The specific technique and sub-technique employed are:

  • T1499 – Endpoint Denial of Service
  • T1499.003 – Application Exhaustion Flood

Impact & Exploit Potential

Successful exploitation of the MadeYouReset vulnerability can lead to severe consequences, including denial-of-service (DoS) conditions for legitimate users and, in certain scenarios, complete out-of-memory crashes on the targeted server. This vulnerability is actively being exploited in the wild, making it a critical concern for organizations using the affected products.

Mitigation & Recommendations

The flaw has been patched in Netty 4.1.124.Final and 4.2.4.Final, as well as Apache Tomcat 11.0.10, 10.1.44 and 9.0.108.

F5 has released engineering hotfixes for the 17.x and 16.x branches:

  • Hotfix-BIGIP-17.5.1.0.80.7-ENG.iso
  • Hotfix-BIGIP-17.1.2.2.0.259.12-ENG.iso
  • Hotfix-BIGIP-16.1.6.0.27.3-ENG.iso

For systems where patching is not immediately feasible, F5 recommends disabling HTTP/2 and reverting to HTTP. Other mitigation strategies include implementing BIG-IP ASM/Advanced WAF DoS protection profiles with TPS and stress-based attributes, including Behavioral DoS Detection and Mitigation capabilities. System administrators should also monitor HTTP/2 profile statistics for signs of attack, such as high numbers of RST_STREAM frames sent and WINDOW_UPDATE frames received.

Other generic recommendations include:

  • Ensuring that your HTTP/2 implementations are up-to-date with the latest security patches.
  • Implementing rate limiting to restrict the number of requests from a single client.
  • Monitoring server resources and setting up alerts for unusual activity.
  • Using a web application firewall (WAF) to filter out malicious requests.

Protecting against subtle, specification-compliant attacks like MadeYouReset is crucial, given that HTTP/2 is a fundamental component of modern web infrastructure.

Instantly Fix Risks with Saner Patch Management

Saner patch management is a continuous, automated, and integrated software that instantly fixes risks exploited in the wild. The software supports major operating systems like Windows, Linux, and macOS, as well as 550+ third-party applications.

It also allows you to set up a safe testing area to test patches before deploying them in a primary production environment. Saner patch management additionally supports a patch rollback feature in case of patch failure or a system malfunction.

Experience the fastest and most accurate patching software here.

Featured Posts

Open Gorbag Targets Ukrainian Organizations with BeautyStealer and WinRAR Exploitation
malware researchGorbag Targets Ukrainian Organizations with BeautyStealer and WinRAR Exploitation

CVE Research

Gorbag Targets Ukrainian Organizations with BeautyStealer and WinRAR Exploitation

Gorbag uses spearphishing and weaponized WinRAR archives to deploy BeautyStealer against Ukrainian organizations. The malware steals browser credentials, VPN configuration files, and targeted documents before exfiltrating the data and deleting itself.

Nov 5, 2026

Open Synology NAS Under Siege: C3RB3R Ransomware Strikes Unpatched Devices
vulnerability researchSynology NAS Under Siege: C3RB3R Ransomware Strikes Unpatched Devices

CVE Research

Synology NAS Under Siege: C3RB3R Ransomware Strikes Unpatched Devices

Synology has warned customers about an ongoing wave of ransomware attacks targeting network-attached storage (NAS) devices running outdated versions of DiskStation Manager (DSM). The attacks have been associated with C3RB3R ransomware, a threat capable of encrypting stored files and demanding payment for their recovery.

Nov 4, 2026

Open NetamLoader Malware Targets Asian Government Windows 10 Systems Through Advanced .NET Injection
malware researchNetamLoader Malware Targets Asian Government Windows 10 Systems Through Advanced .NET Injection

CVE Research

NetamLoader Malware Targets Asian Government Windows 10 Systems Through Advanced .NET Injection

NetamLoader is a multi-stage .NET malware loader targeting customized Windows 10 systems used by Asian government organizations. Discover its AppDomainManager injection techniques, XAML-based execution, AMSI bypass, encrypted command-and-control communications, indicators of compromise, and mitigation strategies.

Nov 2, 2026

Open Agentic AI in the Wild: How AI-Driven Attacks Are Changing Vulnerability Management
Agentic AI in the Wild: How AI-Driven Attacks Are Changing Vulnerability Management

CVE Research

Agentic AI in the Wild: How AI-Driven Attacks Are Changing Vulnerability Management

Oct 30, 2026