T1686.002 — Network Device Firewall
Tactic: Defense Impairment · Platforms: Network Devices
Adversaries may disable network device-based firewall mechanisms entirely or add, delete, or modify particular rules in order to bypass controls limiting network usage. Adversaries may obtain access to devices such as routers, switches, or other perimeter/network devices and change access control…
Sub-technique of Disable or Modify System Firewall (T1686).
Sigma detection rules (1)
T1686.002 Disable or Modify System Firewall: Network Device Firewall
title: "T1686.002 Disable or Modify System Firewall: Network Device Firewall"
id: 6b3433ac-724e-49b8-ada0-f3cbaef9403c
description: 'Detects Linux host-level firewall permissive changes (iptables/nft flush or default-ACCEPT policy) as the closest auditd-observable proxy for T1686.002, which natively concerns disabling firewall protections on dedicated network appliances (routers/switches) - a control plane not covered by Linux auditd. This rule only applies when the monitored Linux host itself functions as a network gateway/firewall (e.g. netfilter-based router); it cannot observe configuration changes on non-Linux network appliance operating systems. Tune using: TrustedAdminIPs, ConfigChangeWindow, RuleScopeThreshold, NewUserPrivilegeThreshold.'
references:
- https://attack.mitre.org/techniques/T1686/002
author: Shahrukh Khan
date: 2026-04-15
license: MIT
tags:
- attack.defense-impairment
- attack.t1686.002
logsource:
category: process_creation
product: linux
service: auditd
detection:
selection_iptables_flush:
type: EXECVE
a0|endswith:
- '/iptables'
- '/ip6tables'
a1:
- '-F'
- '--flush'
selection_iptables_policy_accept:
type: EXECVE
a0|endswith:
- '/iptables'
- '/ip6tables'
a1:
- '-P'
- '--policy'
a3: 'ACCEPT'
selection_nft_flush:
type: EXECVE
a0|endswith: '/nft'
a1: 'flush'
a2:
- 'ruleset'
- 'table'
condition: 1 of selection_*
falsepositives:
- Network administrators and configuration management tools legitimately flush or reset firewall rules/policies during planned maintenance windows; this cannot alone confirm malicious intent.
level: high
Related sub-techniques
Explore