CVE-2026-121812: F5 BIG-IP APM: TMM crash via policy trace DoS
HIGHCVE-2026-121812 is an integer underflow (CWE-191) in F5 BIG-IP APM that lets a remote, unauthenticated attacker crash the Traffic Management Microkernel (TMM) with crafted requests, disrupting traffic while the process restarts. It scores CVSS 7.5, but exposure is narrow: it requires an APM access profile with a per-request policy and the non-default tmm.access.policytrace database key enabled. There is no EPSS score, no public exploit, no Nuclei template, no SSVC decision and no CISA KEV listing, so exploitation likelihood looks low today. This is a generic network appliance rather than an AI package, and it matters only if BIG-IP fronts AI/ML gateways, inference APIs or model-serving endpoints. Check whether tmm.access.policytrace is enabled and disable it if it is not needed for active troubleshooting. Then apply the fixed BIG-IP release listed in F5 advisory K000163944.
What is the risk?
Moderate. Network-reachable with no authentication, low complexity and no user interaction, but it affects availability only (A:H, no confidentiality or integrity impact). The vulnerable configuration needs APM with a per-request policy and the tmm.access.policytrace key, which is disabled by default, so the real exposed population is small. There is no known public exploit, no KEV listing and no EPSS data. Versions past End of Technical Support were not evaluated, so legacy deployments may be affected with no fix available. It is a data plane issue only, with no control plane exposure.
How does the attack unfold?
How severe is it?
What is the attack surface?
What should I do?
1 step1) Identify BIG-IP systems with APM provisioned and virtual servers using an access profile plus a per-request policy. 2) Check the key with 'tmsh list sys db tmm.access.policytrace' and keep it disabled (the default) unless actively troubleshooting. 3) Upgrade to the fixed version listed in F5 advisory K000163944; if the release is past End of Technical Support, plan a migration. 4) Detect by monitoring for unexpected TMM restarts or core files, failover events and HA flapping in /var/log/ltm, and alert on them. 5) Put rate limiting or upstream filtering in front of APM virtual servers where feasible, and prefer HA pairs to limit outage duration.
How is it classified?
Which compliance frameworks are affected?
This CVE is relevant to:
Frequently Asked Questions
What is CVE-2026-121812?
CVE-2026-121812 is an integer underflow (CWE-191) in F5 BIG-IP APM that lets a remote, unauthenticated attacker crash the Traffic Management Microkernel (TMM) with crafted requests, disrupting traffic while the process restarts. It scores CVSS 7.5, but exposure is narrow: it requires an APM access profile with a per-request policy and the non-default tmm.access.policytrace database key enabled. There is no EPSS score, no public exploit, no Nuclei template, no SSVC decision and no CISA KEV listing, so exploitation likelihood looks low today. This is a generic network appliance rather than an AI package, and it matters only if BIG-IP fronts AI/ML gateways, inference APIs or model-serving endpoints. Check whether tmm.access.policytrace is enabled and disable it if it is not needed for active troubleshooting. Then apply the fixed BIG-IP release listed in F5 advisory K000163944.
Is CVE-2026-121812 actively exploited?
No confirmed active exploitation of CVE-2026-121812 has been reported, but organizations should still patch proactively.
How to fix CVE-2026-121812?
1) Identify BIG-IP systems with APM provisioned and virtual servers using an access profile plus a per-request policy. 2) Check the key with 'tmsh list sys db tmm.access.policytrace' and keep it disabled (the default) unless actively troubleshooting. 3) Upgrade to the fixed version listed in F5 advisory K000163944; if the release is past End of Technical Support, plan a migration. 4) Detect by monitoring for unexpected TMM restarts or core files, failover events and HA flapping in /var/log/ltm, and alert on them. 5) Put rate limiting or upstream filtering in front of APM virtual servers where feasible, and prefer HA pairs to limit outage duration.
What systems are affected by CVE-2026-121812?
This vulnerability affects the following AI/ML architecture patterns: model serving behind load balancers, AI API gateways, MLOps portals behind APM access policies.
What is the CVSS score for CVE-2026-121812?
CVE-2026-121812 has a CVSS v3.1 base score of 7.5 (HIGH).
What is the AI security impact?
Affected AI Architectures
MITRE ATLAS Techniques
AML.T0029Denial of AI ServiceAML.T0049Exploit Public-Facing ApplicationCompliance Controls Affected
What are the technical details?
Original Advisory
Description: When BIG-IP APM is provisioned, and a virtual server is configured with an APM access profile and a per-request policy, and the tmm.access.policytrace database key is enabled, undisclosed requests can cause the Traffic Management Microkernel (TMM) to terminate. The key is disabled by default. Impact: Traffic is disrupted while the TMM process restarts. This vulnerability allows a remote unauthenticated attacker to cause a denial-of-service (DoS) on the BIG-IP system . There is no control plane exposure; this is a data plane issue only. Note: Software versions which have reached End of Technical Support (EoTS) are not evaluated.
Exploitation Scenario
An attacker finds an internet-exposed BIG-IP APM virtual server that publishes an internal AI/ML portal or inference API behind an access profile with a per-request policy. The administrator left tmm.access.policytrace enabled after troubleshooting. The attacker sends crafted unauthenticated requests that trigger the integer underflow and crash TMM. By repeating them, the attacker keeps the gateway in a restart loop and takes the AI-backed services behind it offline.
Weaknesses (CWE)
CWE-191 — Integer Underflow (Wrap or Wraparound): The product subtracts one value from another, such that the result is less than the minimum allowable integer value, which produces a value that is not equal to the correct result.
Source: MITRE CWE corpus.
CVSS Vector
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:HReferences
Timeline
Related Vulnerabilities
CVE-2026-3366010.0TensorFlow: type confusion NPD in tensor conversion
Same attack type: DoSCVE-2022-359399.8TensorFlow: ScatterNd OOB write enables RCE/crash
Same attack type: DoSCVE-2022-419009.8TensorFlow: heap OOB RCE in FractionalMaxPool op
Same attack type: DoSCVE-2022-235879.8TensorFlow: integer overflow in Grappler enables RCE
Same attack type: DoSCVE-2023-256689.8TensorFlow: unauthenticated RCE via heap buffer overflow
Same attack type: DoS