CVE-2024-3094 (XZ Utils Supply Chain Backdoor)
This training was a deep dive into supply chain attacks, focusing on how attackers compromise third-party libraries to infiltrate systems. 🌳 ROOT: The Core Lesson 🔹 Your code is only as secure as its weakest dependency. 🔹 Attackers don’t always target your app—they infect the libraries and tools you trust. 🔹 A single update from upstream can spread malware downstream into thousands of systems. 🌲 BRANCHES: Key Takeaways 1️⃣ Trunk: The Major Incidents (Real-World Cases) 📌 Log4j (CVE-2021-44228) – A simple logging library led to RCE attacks on millions of apps. 📌 XZ Utils Backdoor (CVE-2024-3094) – Attackers planted a hidden SSH backdoor inside a widely used Linux tool. 📌 SolarWinds Attack – A trusted software update infected top enterprises & governments. 2️⃣ Branches: How These Attacks Work? 🌿 Compromised Upstream – Hackers inject malicious code into open-source projects. 🌿 Silent Propagation – CI/CD pipelines & OS distros auto-fetch infected updates. 🌿 Exploitation in Production – The attacker gains remote access, RCE, or data leaks. 3️⃣ Leaves: Defensive Actions You Must Take! 🍃 Pin Dependencies – Use fixed versions instead of "latest". 🍃 Verify Integrity – Check hashes, signatures, and changelogs before updating. 🍃 Scan Your Stack – Use SCA tools like Dependabot, Trivy, or Snyk. 🍃 Restrict CI/CD Auto-Updates – Require manual reviews for third-party updates. 🍃 Monitor for Compromise – Set alerts for vulnerable dependencies. 🌟 TOP OF THE TREE: The Final Takeaway Supply chain security is not an option—it's a necessity! If upstream is compromised, everything downstream is at risk. Never blindly trust software updates—always verify before deploying. Your security is only as strong as the weakest library you import! Be proactive, not reactive—because the next Log4j or XZ Backdoor could already be in your pipeline!Question for members: your most rebellious labs
Hello!, I think it would be interesting to share in this Community those labs that have been the most difficult for us to complete; or those that are resisting us and we have invested a significant amount of time: trying tactics and techniques, reading carefully their documentation and references, blog posts about the exploits, testing options or just going step by step. Let's get started :)!: .: I find it hard to finish labs related to access policies or permissions in Cloud: maybe it's the syntax required to give permission to a S3 bucket or to the access point ... but I invest a lot of time to complete them. I am close to having finished 2,400 labs but when I have to write the concrete policy in that json file I struggle :). .: Esoteric labs, as I like to call them ^^. Example: CAN bus. Don't ask me the specific reason, but I have been trying for some time to finish the last few!: I love them, but I'm stuck at the moment. [...] So: which are the labs you have had the hardest time finishing (no matter the difficulty) and which are the ones you are investing the most time in?. Thank you and good luck!.
