Secure Bulletin Navigating the cyber sea with knowledge
Home > Articolo > F5 Issues Emergency Fixes for Six Critical BIG-IP Bugs That Can Take Websites Offline Without a Login
F5 Issues Emergency Fixes for Six Critical BIG-IP Bugs That Can Take Websites Offline Without a Login
Read Time:3 Minute, 53 Second

F5 has published emergency, out-of-band fixes for six critical vulnerabilities in BIG-IP, the traffic-management appliance that sits in front of websites, applications, and remote-access portals at a large share of enterprises and service providers worldwide. Left unpatched, the flaws let an attacker who can simply reach the device — no credentials required — knock its services offline, and in some cases potentially execute code.

BIG-IP’s position in the network makes this class of bug disproportionately disruptive. Because the appliance is deliberately placed between the public internet and the systems it protects, a single crashed instance can take down everything routed through it at once — websites, APIs, and VPN or remote-access portals alike — even though none of the backend infrastructure itself has been touched.

Six Bugs, One Shared Weak Point

BIG-IP’s Traffic Management Microkernel, known as TMM, is the component that actually processes connections flowing through the appliance. When TMM crashes, everything depending on that device goes dark with it — not just the website it’s protecting, but any remote-access portal or application routed through the same box, even though the backend servers behind it may be running perfectly fine.

Because these are data-plane vulnerabilities — meaning they live in the part of BIG-IP that handles live traffic rather than the administrative console — a common mitigation doesn’t apply here. Keeping the management interface off the public internet, standard advice for securing BIG-IP, does nothing to protect against these particular bugs if the affected service itself is still reachable.

iRules Scripts Are the Common Thread

Five of the six vulnerabilities — CVE-2026-120285, CVE-2026-120292, CVE-2026-116953, CVE-2026-120722, and CVE-2026-121805 — involve iRules, the scripting feature administrators use to customize how BIG-IP processes traffic. Some of these also extend to BIG-IP Next for Kubernetes and BIG-IP Cloud-Native Network Functions deployments.

The sixth, CVE-2026-123637, is narrower but still critical: it affects BIG-IP Access Policy Manager specifically when a virtual server runs a per-request policy combining the Client Information and Endpoint Security agents.

A separate flaw, CVE-2026-126822, affecting single sign-on, can also trigger an unauthenticated outage, but F5 rated it high rather than critical and it isn’t counted among the six.

No Confirmed Exploitation — Unlike September’s Zero-Day

F5’s November 18 advisory states that no exploitation had been observed at the time of disclosure. That puts this round of patches in a different category than September’s BIG-IP OAuth zero-day, which F5 confirmed attackers were already actively exploiting before a fix existed. Still, the combination of unauthenticated access, no published workaround, and the sheer reach of BIG-IP deployments means defenders shouldn’t read “not yet exploited” as “low urgency.”

Patch Guidance Depends on the Exact Build

F5’s hotfix guidance spans TMOS 17.1.x, 17.5.x, and 21.1.x, along with separate builds for BIG-IP Next for Kubernetes and Cloud-Native Network Functions:

  • 17.1.0–17.1.3: apply the engineering hotfix built for the 17.1.3 base release.
  • 17.5.0–17.5.1: the correct hotfix differs depending on which specific engineering build is already installed — F5’s guide lists separate packages for different 17.5.1 sub-builds.
  • 21.1.0: a dedicated engineering hotfix is available for this release.
  • BIG-IP Next for Kubernetes and CNF (2.2.x–2.4.x): version-specific emergency hotfixes (EHF) are available for each affected release line.

Teams running customized builds outside these standard release lines should contact F5 Support directly for a replacement package, since the self-service hotfix matrix assumes a known starting point.

Recommended Response Steps

  • Start with internet-facing virtual servers and any BIG-IP instance carrying traffic for business-critical services.
  • Record the exact installed build before selecting a hotfix — the wrong package for a given engineering build won’t apply cleanly.
  • Test the update outside production where feasible before a wider rollout.
  • After deploying, confirm that end users can actually reach the websites and portals served by the device — a successful administrator login to the management interface says nothing about whether live traffic is flowing again.

Organizations running BIG-IP in highly available pairs should also confirm that both the active and standby units receive the hotfix, and that failover testing doesn’t inadvertently reintroduce an unpatched node into production. Given how frequently BIG-IP sits in front of customer-facing services, treating this patch cycle with the same urgency as a confirmed active exploit — rather than waiting for one to materialize — is the more defensible posture for any organization that can’t tolerate an unplanned, unauthenticated outage.

Share: Twitter  |  Facebook  |  LinkedIn
Join the discussion

This is a blog in the Fediverse: you can find this article everywhere with @blog@securebulletin.com and every comment/answer will appear here.

If you want to comment on F5 Issues Emergency Fixes for Six Critical BIG-IP Bugs That Can Take Websites Offline Without a Login, use the discussion on Forum.

>> forum community

Comments

Leave a Reply