Back to Intelligence

F5 BIG-IP TMM Remote Crash Vulnerabilities: Detection and Remediation Guide for Specific iRule and APM Configurations

SA
Security Arsenal Team
November 18, 2026
11 min read

F5 has disclosed six critical vulnerabilities in its BIG-IP platform that permit unauthenticated remote attackers to crash the Traffic Management Microkernel (TMM), the core data-plane component responsible for processing application traffic. When exploited, these flaws cause a denial-of-service condition that persists until the TMM restarts, directly impacting application availability, load balancing, and SSL termination for affected virtual servers. The vulnerabilities require specific, non-default configurations—either particular iRule commands or specific Access Policy Manager (APM) policy agent combinations—which means not all BIG-IP deployments are equally exposed. However, for organizations running internet-facing BIG-IP LTM or APM virtual servers with complex traffic policies, these flaws represent a material availability risk that demands immediate configuration review and patching.

Technical Analysis

Affected Products and Platforms

The vulnerabilities affect F5 BIG-IP systems running the Traffic Management Microkernel (TMM). The TMM is the high-performance, dedicated microkernel that handles all data-plane traffic processing on BIG-IP hardware and virtual editions. Both physical appliances and virtual editions (VE) are affected when the vulnerable configurations are present. Specific affected modules include BIG-IP Local Traffic Manager (LTM) and BIG-IP Access Policy Manager (APM). The exact affected software versions and hotfix levels are detailed in F5's official security advisory; organizations must consult the F5 Security Advisory page and the F5 Vulnerability Matrix (K000131353 and related articles) to determine exposure for their specific software branch and hotfix level.

Vulnerability Mechanism and Exploitation Requirements

The six vulnerabilities share a common impact: an unauthenticated attacker can send crafted traffic to a vulnerable virtual server, triggering a fatal error or assertion failure within the TMM process. This causes the TMM to crash and restart. During the restart window, all traffic processing for that TMM instance halts. On systems with a single TMM or where all TMMs are affected simultaneously, this results in a complete denial of service for the virtual server and any dependent applications.

Critically, exploitation is not universal. Each vulnerability requires a specific configuration prerequisite on the target virtual server. According to F5's disclosure, the prerequisites fall into two categories:

  • iRule-based prerequisites: The virtual server must have an iRule attached that uses a specific, vulnerable command. The iRule is a Tcl-based scripting engine used to inspect, modify, and route traffic. A vulnerable command, when processing attacker-controlled input, can be driven to an error state that crashes the TMM.
  • APM policy-based prerequisites: The virtual server must be configured with an APM access policy that includes two particular types of agents. The interaction between these agents and attacker-supplied session or request data creates the crash condition.

This configuration dependency is a double-edged sword. It reduces the overall attack surface compared to a default-configuration vulnerability, but it also means exposure is determined by specific, often complex, traffic management policies that may have been in place for years and may not be well-documented. Organizations with extensive iRule libraries or granular APM policies are at higher risk.

Exploitation Status

F5 disclosed these vulnerabilities in a coordinated advisory. At the time of writing, there is no public proof-of-concept exploit code available, and F5 has not confirmed active exploitation in the wild. The vulnerabilities are not currently listed in the CISA Known Exploited Vulnerabilities (KEV) catalog. However, the low barrier to exploitation (unauthenticated network access to a virtual server) combined with the high impact (complete traffic disruption) makes these flaws attractive targets for threat actors seeking to disrupt services. The specific configuration requirements mean that attackers would likely need to perform reconnaissance or have prior knowledge of a target's BIG-IP configuration to successfully exploit these flaws, which slightly raises the bar but does not eliminate risk for high-value targets.

Detection & Response

Detecting exploitation attempts and successful crashes requires monitoring both the BIG-IP data plane and the underlying system for TMM instability. Because these are availability-focused vulnerabilities, the primary detection goal is to identify TMM restarts, core dumps, and anomalous traffic patterns that precede a crash.

Sigma Rules

The following Sigma rules are designed to detect TMM instability and potential exploitation artifacts from BIG-IP system logs, which are commonly forwarded to a SIEM via syslog. These rules assume standard syslog formatting from F5 BIG-IP systems.

YAML
---
title: F5 BIG-IP TMM Process Crash or Unexpected Restart
id: 8b3e4a12-7c9d-4e5f-a1b2-3c4d5e6f7a8b
status: experimental
description: Detects unexpected termination or restart of the Traffic Management Microkernel (TMM) process, which may indicate exploitation of a denial-of-service vulnerability.
references:
  - https://thehackernews.com/2026/11/f5-warns-6-critical-big-ip-flaws-can.html
  - https://my.f5.com/manage/s/article/K000131353
author: Security Arsenal
date: 2026/04/06
tags:
  - attack.impact
  - attack.t1499
logsource:
  product: f5
  service: tmm
detection:
  selection:
    - 'tmm'
    - 'process exited'
    - 'signal'
    - 'core dumped'
    - 'assertion failed'
    - 'panic'
  condition: selection
falsepositives:
  - Planned maintenance, software upgrades, or manual TMM restarts by administrators.
  - Hardware faults or resource exhaustion unrelated to attack.
level: high
---
title: F5 BIG-IP Virtual Server Traffic Processing Halt
id: 9c4f5b23-8d0e-5f6a-b2c3-4d5e6f7a8b9c
status: experimental
description: Detects a sudden halt in traffic processing logs for a specific virtual server, which may indicate a TMM crash affecting that virtual server.
references:
  - https://thehackernews.com/2026/11/f5-warns-6-critical-big-ip-flaws-can.html
author: Security Arsenal
date: 2026/04/06
tags:
  - attack.impact
  - attack.t1499
logsource:
  product: f5
  service: ltm
detection:
  selection:
    - 'virtual server'
    - 'connection reset'
    - 'no available pool members'
    - 'traffic processing stopped'
  condition: selection
falsepositives:
  - Legitimate pool member failure or health monitor flapping.
  - Configuration changes disabling the virtual server.
level: medium
---
title: F5 BIG-IP iRule Execution Error or Tcl Panic
id: 0d5a6c34-9e1f-6a7b-c3d4-5e6f7a8b9c0d
status: experimental
description: Detects critical errors or panics within the iRule Tcl interpreter that may precede or accompany a TMM crash triggered by a malicious iRule command.
references:
  - https://thehackernews.com/2026/11/f5-warns-6-critical-big-ip-flaws-can.html
author: Security Arsenal
date: 2026/04/06
tags:
  - attack.execution
  - attack.t1059
logsource:
  product: f5
  service: tmm
detection:
  selection:
    - 'iRule'
    - 'Tcl error'
    - 'Tcl panic'
    - 'invalid command name'
    - 'stack overflow'
  condition: selection
falsepositives:
  - Bugs or syntax errors in custom iRules during development or deployment.
  - Resource limits being hit by legitimate complex iRules.
level: medium

KQL Hunt Query (Microsoft Sentinel / Defender)

For organizations ingesting F5 BIG-IP logs into Microsoft Sentinel via the F5 ASM/AFM connector or CEF/Syslog, the following KQL query hunts for TMM crash events and correlates them with the virtual servers that were active at the time. This helps identify which specific configurations may have been targeted.

KQL — Microsoft Sentinel / Defender
// Hunt for F5 TMM crashes and correlate with active virtual servers
let TmmCrashEvents = Syslog
    | where Facility == "local0" or Facility == "daemon"
    | where SyslogMessage has_any ("tmm", "process exited", "signal", "core dumped", "assertion failed", "panic")
    | where SyslogMessage has "tmm"
    | project TimeGenerated, HostName, ProcessName, SyslogMessage, SeverityLevel;
TmmCrashEvents
| join kind=leftouter (
    Syslog
    | where SyslogMessage has "virtual server" and SyslogMessage has_any ("connection", "request", "client")
    | summarize LastTrafficTime=max(TimeGenerated) by HostName, VirtualServer=extract(@"virtual server ([^\s]+)", 1, SyslogMessage)
    ) on HostName
| where TimeGenerated between (LastTrafficTime .. LastTrafficTime + 5m)
| project TimeGenerated, HostName, VirtualServer, SyslogMessage, SeverityLevel
| order by TimeGenerated desc

Velociraptor VQL

Velociraptor is not typically deployed directly on F5 BIG-IP appliances, which are hardened, closed systems. However, for organizations running F5 BIG-IP Virtual Edition (VE) on guest VMs within a virtualized environment (e.g., VMware, KVM), Velociraptor can be used to hunt on the underlying hypervisor or management network for related artifacts, such as unexpected process crashes on the VE guest or suspicious network connections targeting the VE management interface. The following VQL artifact is designed to hunt for processes that have crashed or produced core dumps on a Linux-based BIG-IP VE guest, which can be an indicator of TMM exploitation.

VQL — Velociraptor
-- Hunt for crashed processes and core dumps on F5 BIG-IP VE Linux guests
SELECT Pid, Name, Exe, CommandLine, Username, CreateTime, Status
FROM pslist()
WHERE Name =~ 'tmm' OR CommandLine =~ 'tmm'
   OR Status =~ 'zombie' OR Status =~ 'defunct'
UNION
SELECT Pid=NULL, Name=NULL, Exe=NULL, CommandLine=NULL, Username=NULL, CreateTime=NULL, Status=NULL, FilePath=FullPath, Size=Size, ModTime=Mtime
FROM glob(globs='/var/core/*tmm*')

Note: The glob path for core dumps may vary based on the BIG-IP VE's core dump configuration. Adjust the path to match your environment's kernel.core_pattern setting.

Remediation Script

The following Bash script is designed to run on a BIG-IP system (via the advanced shell or TMSH) to audit for potentially vulnerable configurations. It checks for the presence of iRules and APM policies on virtual servers, which is the first step in identifying exposure. This script does not patch; it identifies where further investigation is required.

Bash / Shell
#!/bin/bash
# F5 BIG-IP Vulnerability Configuration Audit Script
# Run this script on the BIG-IP system via advanced shell (bash) with appropriate privileges.
# It audits virtual servers for attached iRules and APM policies.
# Output should be reviewed by a security engineer to determine if specific vulnerable commands or agents are present.

echo "=========================================="
echo "F5 BIG-IP Vulnerable Configuration Audit"
echo "Date: $(date)"
echo "=========================================="
echo ""

echo "[+] Listing all Virtual Servers and their attached iRules..."
tmsh list ltm virtual one-line | while read -r line; do
    vs_name=$(echo "$line" | awk '{print $3}')
    irules=$(echo "$line" | grep -o 'rules { [^}]* }' | sed 's/rules { //;s/ }//')
    if [ -n "$irules" ]; then
        echo "Virtual Server: $vs_name"
        echo "  iRules: $irules"
        echo ""
    fi
done

echo "[+] Listing all Virtual Servers with APM policies..."
tmsh list ltm virtual one-line | while read -r line; do
    vs_name=$(echo "$line" | awk '{print $3}')
    if echo "$line" | grep -q 'apm'; then
        echo "Virtual Server: $vs_name (APM policy attached)"
    fi
done

echo ""
echo "[+] Listing all APM Access Policies..."
tmsh list apm policy access-policy one-line 2>/dev/null | while read -r line; do
    policy_name=$(echo "$line" | awk '{print $4}')
    echo "APM Policy: $policy_name"
done

echo ""
echo "=========================================="
echo "Audit complete. Review the output above."
echo "For each virtual server with an iRule, examine the iRule code for specific commands."
echo "For each APM policy, examine the policy for the presence of two specific agent types."
echo "Refer to the F5 Security Advisory for the exact vulnerable commands and agent combinations."
echo "=========================================="

To run this script:

  1. Save it to a file on the BIG-IP system, e.g., /var/tmp/audit_bigip_config.sh.
  2. Make it executable: chmod +x /var/tmp/audit_bigip_config.sh.
  3. Execute it: /var/tmp/audit_bigip_config.sh.
  4. Review the output. For any virtual server with an iRule, you must then inspect the iRule's code (using tmsh list ltm rule <rule_name>) to check for the specific vulnerable command. For APM policies, you must review the policy's Visual Policy Editor (VPE) or use tmsh list apm policy access-policy <policy_name> to identify the agent types.

Remediation

Immediate Actions

  1. Identify Affected Systems: Use the audit script above and F5's official vulnerability matrix to determine which BIG-IP systems and specific virtual servers are affected. Cross-reference your software version and hotfix level with the F5 advisory.
  2. Apply Patches: F5 has released hotfixes for affected software versions. The specific hotfix version depends on your current software branch. Consult the F5 Security Advisory and the F5 Downloads portal to obtain the correct hotfix for your system. Apply hotfixes during a maintenance window, as a TMM restart is required.
  3. Configuration Review (Workaround): If immediate patching is not possible, the only effective workaround is to remove the vulnerable configuration. For iRule-based vulnerabilities, this means removing or modifying the iRule to eliminate the vulnerable command. For APM-based vulnerabilities, this means modifying the access policy to remove one or both of the specific agent types. This will likely alter application behavior and must be tested thoroughly. This is not a long-term solution; patching is mandatory.
  4. Network-Level Mitigation: As a temporary compensating control, restrict access to vulnerable virtual servers using packet filters or an upstream firewall. Limit connections to trusted source IP ranges where business requirements allow. This reduces the attack surface but does not eliminate the vulnerability for legitimate users who may be attackers.

Long-Term Hardening

  • iRule Governance: Implement a strict review and approval process for all iRules. Treat iRules as code: use version control, static analysis, and peer review. Avoid using deprecated or complex commands where simpler alternatives exist.
  • APM Policy Hygiene: Regularly review and simplify APM access policies. Remove unused agents and consolidate logic where possible. Document the business justification for each agent in a policy.
  • Segmentation: Do not expose BIG-IP management interfaces (self-IPs, management ports) to untrusted networks. Use dedicated management VLANs and strict firewall rules.
  • Monitoring: Ensure that TMM crash events, core dump generation, and virtual server state changes are logged and alerted in your SIEM. Establish a baseline for normal TMM restart behavior to detect anomalies.
  • Vendor Advisories: Subscribe to F5's security mailing list and RSS feeds. Monitor the F5 Vulnerability Matrix for updates on these and future vulnerabilities.

Official References

Related Resources

Security Arsenal Penetration Testing Services AlertMonitor Platform Book a SOC Assessment vulnerability-management Intel Hub

Is your security operations ready?

Get a free SOC assessment or see how AlertMonitor cuts through alert noise with automated triage.

F5 BIG-IP TMM Remote Crash Vulnerabilities: Detection and Remediation Guide for Specific iRule and APM Configurations | Security Arsenal | Security Arsenal