JupiteX Get the app
Defence & Security21 Nov 2026 · about 7 min

F5 releases emergency patches for critical BIG-IP RCE flaws

The brief

F5 warned customers to patch five critical BIG-IP vulnerabilities: CVE-2026-121805, CVE-2026-120292, CVE-2026-120285, CVE-2026-116953, and CVE-2026-123637. The stack-based buffer overflow flaws affect BIG-IP Access Policy Manager (APM) and the iRules programmable scripting feature. They matter because attackers may disrupt services or execute code on vulnerable systems. An unauthenticated attacker can exploit the five flaws to terminate the Traffic Management Microkernel, or TMM, and run arbitrary code remotely when a system has a vulnerable configuration. A separate, high-severity APM SAML issue, CVE-2026-126822, comes from an integer underflow. F5 says remote unauthenticated attackers can use it for denial-of-service attacks, and it may allow remote code execution. F5 said it has no mitigation measures for these flaws and released out-of-band engineering hotfixes on Wednesday. It has not flagged any of the issues as exploited in attacks. Still, the company advised admins to upgrade to the latest version as soon as possible, given the history of attackers exploiting F5 flaws.

01

Which BIG-IP vulnerabilities did F5 patch, and what kinds of attacks can they enable?

F5 warned customers to patch five critical BIG-IP vulnerabilities: CVE-2026-121805, CVE-2026-120292, CVE-2026-120285, CVE-2026-116953, and CVE-2026-123637. The stack-based buffer overflow flaws affect BIG-IP Access Policy Manager (APM) and the iRules programmable scripting feature. They matter because attackers may disrupt services or execute code on vulnerable systems.

An unauthenticated attacker can exploit the five flaws to terminate the Traffic Management Microkernel, or TMM, and run arbitrary code remotely when a system has a vulnerable configuration. A separate, high-severity APM SAML issue, CVE-2026-126822, comes from an integer underflow. F5 says remote unauthenticated attackers can use it for denial-of-service attacks, and it may allow remote code execution.

F5 said it has no mitigation measures for these flaws and released out-of-band engineering hotfixes on Wednesday. It has not flagged any of the issues as exploited in attacks. Still, the company advised admins to upgrade to the latest version as soon as possible, given the history of attackers exploiting F5 flaws.

02

What are BIG-IP APM and iRules, and what do they do?

BIG-IP APM stands for Access Policy Manager. F5 describes it as a centralized access management proxy solution. In broad terms, it supports managing access through a central point. The article does not detail its individual access policies or how customers configure them.

BIG-IP iRules is a programmable scripting feature. It lets users create scripts for BIG-IP, though the source does not explain what particular scripts do. The reported buffer overflow flaws affect APM and iRules, so the risk is tied to these named parts of the BIG-IP platform rather than every product feature described in the article.

This matters because attackers do not need an account to exploit the five critical flaws on systems with vulnerable configurations. F5 says they can terminate the Traffic Management Microkernel, disrupting traffic, and execute code remotely. F5 also disclosed a separate APM SAML vulnerability that can be abused for denial-of-service attacks. Customers should apply the hotfixes F5 released for vulnerable devices.

03

What could an unauthenticated attacker do to a vulnerable BIG-IP system?

The five critical BIG-IP flaws can give an unauthenticated threat actor a way to affect a vulnerable system without first signing in. F5 says attackers can exploit them to terminate the Traffic Management Microkernel, or TMM, and execute arbitrary code remotely. The flaws affect APM and iRules, and exploitability depends on having a vulnerable configuration.

A concrete consequence is service disruption: when TMM terminates and restarts, traffic is disrupted. F5 also warned that the separate APM SAML vulnerability, CVE-2026-126822, can let a remote unauthenticated attacker cause denial of service and may allow remote code execution. That issue stems from an integer underflow, while the five other flaws are stack-based buffer overflows.

F5 released out-of-band engineering hotfixes and said it has no mitigation measures for these flaws. It has not flagged them as exploited in attacks, but it advises customers to upgrade to the latest version quickly. The article does not say that every BIG-IP installation is vulnerable; it refers to systems with vulnerable configurations.

04

How many BIG-IP instances and APM-fingerprinted IP addresses are exposed online, and does that mean they are all vulnerable?

Internet threat watchdog Shadowserver currently tracks more than 177,000 BIG-IP instances and over 14,600 IP addresses with BIG-IP APM fingerprints exposed online. These figures show the scale of online visibility, but they are not a count of confirmed vulnerable devices. An exposed fingerprint does not, by itself, show whether a system has the affected configuration.

The article specifically notes that there is no information on how many of the tracked instances or addresses are honeypots or have already been patched. A honeypot is a decoy system used to observe activity; that general meaning does not establish how Shadowserver classifies any particular address. The source provides no breakdown of vulnerable, patched, or decoy systems.

The figures therefore indicate a potentially broad set of systems for administrators to check, not a confirmed victim count. F5 urged customers to install its out-of-band hotfixes and upgrade vulnerable systems as soon as possible. The company has not flagged the newly disclosed flaws as exploited in attacks, but it points to a history of F5 vulnerabilities being abused.

05

Why does F5 urge customers to patch quickly even though it has not reported attacks using these flaws?

F5 has not flagged any of the newly disclosed vulnerabilities as exploited in attacks. Even so, it urged administrators to upgrade vulnerable systems to the latest version as soon as possible. F5 said it has no mitigation measures for these flaws, making its out-of-band hotfixes important for affected devices.

The warning reflects a broader history described in the article. Attackers have used F5 security flaws to breach corporate networks, map internal servers, hijack devices, deploy data-wiping malware, and steal sensitive files. Since November 2021, the US Cybersecurity and Infrastructure Security Agency has flagged seven F5 vulnerabilities as actively exploited; ransomware gangs also used four of them.

There are recent examples, too: F5 patched a critical BIG-IP APM flaw, CVE-2026-94127, in September, and it is being actively exploited in zero-day remote code execution attacks. The article also reports that state-backed hackers breached F5 systems in August 2025 to steal source code and information about undisclosed BIG-IP vulnerabilities. This history explains the urgency, not proof of attacks using the current flaws.

06

What is a stack-based buffer overflow, and how can it let an attacker run code remotely?

A stack-based buffer overflow occurs when a program puts more data into a stack-allocated buffer than it can hold. The stack is a region of memory programs use while functions run, including for temporary data. This definition comes from general computing knowledge; the article identifies the BIG-IP flaws as stack-based buffer overflows but does not describe their code-level details.

In some cases, data written past a buffer can overwrite nearby stack contents, including information used to control what the program does next. If an attacker can influence that information, they may redirect execution to attacker-chosen code. Whether that works depends on the program and its protections; an overflow does not automatically mean remote code execution is possible.

For these BIG-IP flaws, F5 says unauthenticated threat actors can terminate TMM and execute arbitrary code remotely on systems with vulnerable configurations. The article does not describe the exact exploit steps or the specific memory contents attackers overwrite. F5 released hotfixes and advised customers to upgrade vulnerable systems quickly.

07

What is the difference between a system’s data plane and control plane, and why does that distinction matter here?

In general networking terms, the data plane handles the traffic a device processes or forwards. The control plane is involved in managing the device and directing its operation. These broad definitions are general background; the article does not provide a detailed BIG-IP architecture guide.

The distinction appears in F5’s description of the separate APM SAML vulnerability, CVE-2026-126822. F5 said the issue is a data plane problem only, with no control plane exposure. It also said traffic is disrupted while TMM restarts, and that a remote unauthenticated attacker can cause denial of service and may perform remote code execution.

That distinction helps describe where F5 says the vulnerability’s exposure lies: in the data plane, not the control plane. It does not mean the issue is harmless, since F5 warned of service disruption and possible code execution. The article’s statement is about the SAML vulnerability; it does not establish that every one of the other five flaws has the same plane classification.

This brief was written by AI from the original reporting and checked by other models. Names, figures and quotes come from the source; read it for full context.

Read more in the JupiteX app

Pulse is free. New stories every 4 hours, each one broken into the questions that explain it.

Or read more news on the web