Skip to navigation

Threat reportVulnerabilityTL-2026-6370

F5 BIG-IP: Six Critical Data-Plane Stack Buffer Overflows (CVE-2026-123637, CVE-2026-121805, CVE-2026-120292, CVE-2026-120285, CVE-2026-116953, CVE-2026-120722) Crash TMM, May Allow RCE

criticalACTIVE

F5 disclosed six critical (CVSS v3.1 9.8 / v4.0 9.3) stack-based buffer overflows in the BIG-IP data plane (TMM), each requiring a specific virtual server configuration. Unauthenticated traffic can crash TMM and F5 says the flaws may also allow remote code execution.

Key facts

IdentifiersTL-2026-6370, CVE-2026-123637, CVE-2026-121805, CVE-2026-120292, CVE-2026-120285, CVE-2026-116953, CVE-2026-120722, CVE-2026-125274, CVE-2026-120718, CVE-2026-125272, CVE-2026-126822, CVE-2026-126336, CVE-2026-121812, CVE-2026-94127
SeverityCRITICAL
CVSS9.8 (CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H)
First disclosed
Exploitation statusLinked CVE exploited in the wild; Listed in CISA KEV: CVE-2026-94127 (added )
EPSSCVE-2026-94127: 1.3% (71.1th percentile)
Affected productsF5 BIG-IP (APM, PEM, DNS and all modules, depending on CVE and virtual server configuration); F5 BIG-IP Next for Kubernetes (BNK); F5 BIG-IP Next CNF (CVE-2026-121805 and CVE-2026-120285)
Weaknesses (CWE)CWE-121
MITRE ATT&CK techniquesT1059 Command and Scripting Interpreter, T1078 Valid Accounts, T1190 Exploit Public-Facing Application, T1203 Exploitation for Client Execution, T1489 Service Stop, T1499.004 Application or System Exploitation, T1550.001 Application Access Token
MotivationUNKNOWN
Target sectorstelecoms, technology, government administration, finance, enterprise
Target regionsGlobal
Detection rules9 (1 shown free)
Sources10 cited sources

Detection coverage

As of 2026-11-18, Threadlinqs Intelligence publishes 9 detection rule(s) for TL-2026-6370 across Splunk SPL, Microsoft KQL and Sigma, covering 9 indicator(s) of compromise. The whole corpus is readable without an account. Every threat comes with one complete detection in Splunk SPL, Microsoft KQL and Sigma, free for everyone; a free account adds daily TL Creds to unlock the rest. Paid plans add a larger daily allowance, correlation and the MCP server. Threadlinqs MCP server · View plans.

Free detection: F5 BIG-IP TMM Crash, SIGABRT or Core File After Data-Plane Traffic (TL-2026-6370 / CWE-121 Stack Overflow)

Detects unexplained TMM (Traffic Management Microkernel) crashes, fatal signals (SIGABRT/SIGSEGV/SIGBUS) or core files written to /var/core on BIG-IP. F5 states the six critical CWE-121 stack buffer overflows (CVE-2026-123637, CVE-2026-121805, CVE-2026-120292, CVE-2026-120285, CVE-2026-116953, CVE-2026-120722) crash TMM when an unauthenticated client sends crafted traffic to a virtual server with a vulnerable configuration, and may allow RCE. Unexplained tmm crashes are the primary observable exploitation-attempt indicator (F5 cited SIGABRT crashes for the related CVE-2026-94127).

  • Languages: Splunk SPL, Microsoft KQL, Sigma
  • ATT&CK: T1190, T1203, T1499.004
  • Data sources: f5_bigip_syslog, splunk_f5_bigip_addon, sentinel_syslog
  • Severity: high
  • Confidence: medium

Splunk SPL

index=network sourcetype=f5:bigip:syslog earliest=-24h tmm ("SIGABRT" OR "SIGSEGV" OR "SIGBUS" OR "SIGILL" OR "/var/core/" OR "tmm core" OR "TMM crash")
| eval signal=case(match(_raw,"SIGABRT"),"SIGABRT",match(_raw,"SIGSEGV"),"SIGSEGV",match(_raw,"SIGBUS"),"SIGBUS",match(_raw,"SIGILL"),"SIGILL",true(),"core_or_crash")
| stats count min(_time) as first_seen max(_time) as last_seen values(signal) as signals by host
| eval first_seen=strftime(first_seen,"%Y-%m-%d %H:%M:%S"), last_seen=strftime(last_seen,"%Y-%m-%d %H:%M:%S")
| table host count signals first_seen last_seen

Microsoft KQL

Syslog
| where TimeGenerated > ago(24h)
| where ProcessName startswith "tmm" or SyslogMessage has "tmm"
| where SyslogMessage has_any ("SIGABRT", "SIGSEGV", "SIGBUS", "SIGILL", "/var/core/", "tmm core", "TMM crash")
| extend Signal = extract(@"(SIG[A-Z]+)", 1, SyslogMessage)
| summarize Events = count(), FirstSeen = min(TimeGenerated), LastSeen = max(TimeGenerated), Signals = make_set(Signal, 5) by Computer
| project Computer, Events, Signals, FirstSeen, LastSeen

Sigma

title: F5 BIG-IP TMM Crash or Fatal Signal (TL-2026-6370)
id: 3b7e1c52-6a0d-4f2e-9d41-8c5a2e7f1a01
status: experimental
description: Detects BIG-IP TMM crashes, fatal signals or core files that may indicate exploitation attempts of the CWE-121 data-plane stack buffer overflows disclosed in F5 K000164058.
references:
    - https://my.f5.com/manage/s/article/K000164058
    - https://thehackernews.com/2026/11/f5-warns-6-critical-big-ip-flaws-can.html
author: Threadlinqs Team
date: 2026-11-18
tags:
    - attack.initial-access
    - attack.t1190
    - attack.execution
    - attack.t1203
    - attack.impact
    - attack.t1499.004
    - cve.2026.123637
    - cve.2026.121805
    - cve.2026.120292
    - cve.2026.120285
    - cve.2026.116953
    - cve.2026.120722
logsource:
    product: linux
    service: syslog
detection:
    kw_process:
        - 'tmm'
    kw_crash:
        - 'SIGABRT'
        - 'SIGSEGV'
        - 'SIGBUS'
        - 'SIGILL'
        - '/var/core/'
        - 'TMM crash'
    condition: kw_process and kw_crash
falsepositives:
    - Known TMM bugs, memory exhaustion or hardware faults causing crashes unrelated to attack traffic
    - Planned troubleshooting that deliberately generates core files
level: high

Known false positives:

  • Non-malicious TMM crashes from unrelated software defects or resource exhaustion
  • Intentional core generation during F5 support troubleshooting

Tuning notes: Adjust index/sourcetype to your BIG-IP syslog ingestion. Correlate hits with the virtual servers carrying PEM::subscriber, MR::message route, PSC::ip_address, GENERICMESSAGE::route or DNS::scrape iRules, and with client source IPs in the preceding minutes. Suppress hosts with an open F5 support case for known crashes.

8 more detection rules (Splunk SPL, Microsoft KQL, Sigma) · unlock with TL Creds. Sign up free to unlock the detection rules or query them with the Threadlinqs MCP server (Purple)
  • F5 BIG-IP Repeated TMM Restarts Within 15 Minutes (Service Disruption, TL-2026-6370) · unlock with TL Creds
  • F5 BIG-IP Shell Command Execution by Non-Admin User (Post-Exploitation, TL-2026-6370) · unlock with TL Creds
  • F5 BIG-IP iControl REST Auth Token Used From Multiple Source IPs (CVE-2026-120718) · unlock with TL Creds
  • F5 BIG-IP iRule Containing Vulnerable Trigger Commands Created or Modified (TL-2026-6370) · unlock with TL Creds
  • F5 BIG-IP [indicator] Database Key Enabled (CVE-2026-121812) · unlock with TL Creds
  • F5 BIG-IP iControl SOAP Request From Non-Internal Source (CVE-2026-125274) · unlock with TL Creds
  • F5 BIG-IP iControl SOAP Management Actions by Non-Admin Account (CVE-2026-125274) · unlock with TL Creds
  • F5 BIG-IP APM OAuth Authentication Failure Burst Against Virtual Server (TL-2026-6370 Related CVE-2026-94127) · unlock with TL Creds

Indicators of compromise

As of 2026-11-18, Threadlinqs Intelligence tracks 9 indicators of compromise for TL-2026-6370.

9 indicators of compromise · unlock with TL Creds. Compare plans

Technical analysis

Fixes are engineering hotfixes only, no workaround is listed, and no exploitation is stated in the F5 advisories.

On 2026-11-18 F5 published a batch of 22 security advisories (10 of them for NGINX), including K000164058 covering six critical stack-based buffer overflows (CWE-121) in the BIG-IP data plane, the Traffic Management Microkernel (TMM). Each flaw is rated CVSS v3.1 9.8 and CVSS v4.0 9.3. An unauthenticated remote attacker who can send traffic to an affected virtual server can crash TMM, interrupting traffic until TMM restarts, and F5 states the flaws may also allow remote code execution. Per reporting, the management interface is not exposed; the attack surface is the data-plane virtual server.

Each flaw is only reachable under a specific configuration. CVE-2026-123637: a BIG-IP APM per-request policy containing both the Client Information agent and Endpoint Security agents. CVE-2026-121805: an iRule using PEM::subscriber on client traffic (BIG-IP PEM, BIG-IP Next for Kubernetes (BNK), and BIG-IP CNF, which was added on the day of publication). CVE-2026-120292: an HTTP profile plus an iRule using MR::message route (all BIG-IP modules, BNK). CVE-2026-120285: an HTTP profile plus an iRule using PSC::ip_address (all BIG-IP modules, BNK, CNF added on the day of publication). CVE-2026-116953: an iRule using GENERICMESSAGE::route on client traffic (all modules, BNK). CVE-2026-120722: an iRule using DNS::scrape on client traffic (BIG-IP DNS, BNK; the advisory lists CNF as not vulnerable but the CVE record updated the same day lists CNF as affected).

Fixes are engineering hotfixes only, with no scheduled release listed: BNK-2.4.2-EHF1 (2.4.0-2.4.2), BNK-2.3.4-EHF1 (2.3.0-2.3.4), BNK-2.2.5-EHF1 (2.2.0-2.2.5), and CNF-2.4.2-EHF1 / CNF-2.3.4-EHF1 / CNF-2.2.5-EHF1 for CVE-2026-121805 and CVE-2026-120285 only. Specific BIG-IP hotfix build numbers were not given in the reporting reviewed. F5 evaluates only versions under technical support, so older releases are not confirmed safe, and per K4918 engineering hotfixes do not undergo the extensive QA of scheduled releases and carry no warranty. Four of the flaws were reported by Sina Kheirkhah of watchTowr; F5 found the other two internally.

The same F5 batch fixed lower-rated BIG-IP issues defenders should triage together: CVE-2026-125274 (iControl SOAP, 8.8; a user with at least the Guest role can send unauthorized management requests, potentially running system commands, creating or deleting files, or disabling services; fixed in 21.1.0.3, 17.5.1.10, 17.1.3.6 and BIG-IQ 8.4.2.3; workaround: restrict iControl SOAP to trusted networks), CVE-2026-120718 (iControl REST, 8.1; tokens may not expire as expected), and TMM-crash DoS flaws CVE-2026-125272 (APM form-based Logon Page, 7.5), CVE-2026-126822 (APM SAML SP, 7.5), CVE-2026-126336 (SAML SP with encrypted assertions, 7.5) and CVE-2026-121812 (APM / SSL Orchestrator with the tmm.access.policytrace database key enabled, 7.5).

No exploitation, public PoC, attribution or network IOCs are reported for the six critical flaws. Context: the separate F5 BIG-IP APM heap overflow CVE-2026-94127 (OAuth authorization server configurations, CVSS 9.8) was disclosed 2026-09-22 as exploited in the wild and added to CISA KEV the same day; F5 cited repeated OAuth authentication failures, suspicious commands and TMM SIGABRT crashes as indicators. That flaw is a distinct heap-based issue, but it shows threat actors are targeting BIG-IP data-plane memory-safety bugs. Defenders should inventory virtual servers and iRules for the trigger configurations, apply the hotfixes from F5 support, and monitor for unexplained TMM restarts and core files.

MITRE ATT&CK techniques by tactic:

Execution

T1059 Command and Scripting Interpreter; T1203 Exploitation for Client Execution

Initial Access

T1078 Valid Accounts; T1190 Exploit Public-Facing Application

Impact

T1489 Service Stop; T1499.004 Application or System Exploitation

lateral-movement

T1550.001 Application Access Token

Remediation

Patches

  • BNK-2.4.2-EHF1 for BNK 2.4.0-2.4.2
  • BNK-2.3.4-EHF1 for BNK 2.3.0-2.3.4
  • BNK-2.2.5-EHF1 for BNK 2.2.0-2.2.5
  • CNF-2.4.2-EHF1, CNF-2.3.4-EHF1, CNF-2.2.5-EHF1 for CVE-2026-121805 and CVE-2026-120285
  • BIG-IP 21.1.0.3, 17.5.1.10, 17.1.3.6 and BIG-IQ 8.4.2.3 for CVE-2026-125274

Immediate actions

  • Inventory virtual servers and iRules for the trigger configurations: PEM::subscriber, MR::message route, PSC::ip_address, GENERICMESSAGE::route, DNS::scrape, and APM per-request policies with Client Information plus Endpoint Security agents
  • Request and apply the F5 engineering hotfixes listed in K000164058 (BNK-2.4.2-EHF1, BNK-2.3.4-EHF1, BNK-2.2.5-EHF1; CNF-2.4.2-EHF1, CNF-2.3.4-EHF1, CNF-2.2.5-EHF1 for CVE-2026-121805 and CVE-2026-120285)
  • Monitor for unexpected TMM restarts, SIGABRT crashes and core files
  • Restrict iControl SOAP to trusted networks (CVE-2026-125274)

Workarounds

  • No workaround is listed for the six critical flaws
  • CVE-2026-125274: restrict iControl SOAP to trusted networks (blocking on self-IPs may break HA)
  • CVE-2026-126822: remove the SAML Auth agent if SAML SP is not needed and restrict client networks

Longer-term hardening

  • Deploy high availability pairs to reduce the impact of TMM crashes
  • Move systems off hotfix builds to scheduled releases once F5 ships them
  • Upgrade end-of-support BIG-IP versions, which F5 does not evaluate

Affected and fixed versions

  • F5 — BIG-IP (APM, PEM, DNS and all modules, depending on CVE and virtual server configuration)
    Vulnerable versions: Supported BIG-IP versions with the trigger configuration; exact ranges not stated in reviewed sources
    Fixed in: Engineering hotfix from F5 support; no scheduled release listed
  • F5 — BIG-IP Next for Kubernetes (BNK)
    Vulnerable versions: 2.2.0-2.2.5; 2.3.0-2.3.4; 2.4.0-2.4.2
    Fixed in: BNK-2.2.5-EHF1; BNK-2.3.4-EHF1; BNK-2.4.2-EHF1
  • F5 — BIG-IP Next CNF (CVE-2026-121805 and CVE-2026-120285)
    Vulnerable versions: 2.2.0-2.2.5; 2.3.0-2.3.4; 2.4.0-2.4.2
    Fixed in: CNF-2.2.5-EHF1; CNF-2.3.4-EHF1; CNF-2.4.2-EHF1

Timeline

  • F5 discloses BIG-IP APM heap overflow CVE-2026-94127 as exploited in the wild; CISA adds it to KEV the same day (separate issue from the six critical flaws)
  • CISA federal remediation deadline for CVE-2026-94127
  • The Hacker News reports the flaws; four were reported by watchTowr's Sina Kheirkhah, two found internally by F5; no exploitation stated
  • Same batch fixes iControl SOAP CVE-2026-125274 (8.8), iControl REST CVE-2026-120718 (8.1) and four APM/SSL Orchestrator TMM-crash flaws (7.5)
  • BIG-IP CNF added as affected for CVE-2026-121805 and CVE-2026-120285; CVE record for CVE-2026-120722 lists CNF as affected despite the advisory
  • F5 publishes 22 advisories including K000164058 for six critical BIG-IP TMM stack buffer overflows (CVSS v3.1 9.8 / v4.0 9.3), fixed via engineering hotfixes only

Sources cited

Drafted by the Threadlinqs research pipeline from 10 sources; published ; last updated —.

Threadlinqs Intelligence — Real-Time Threat Detection Platform

// threat_feed
$ sort --newest
Showing all threats

Live intelligence console

Threat level

Latest Threats