Intermediate

Debugging Network Issues

Build a systematic, layer-by-layer diagnostic process using ping, mtr, dig, ss, and tcpdump/Wireshark, then apply it cold to a deliberately broken version of the load-balanced setup from the previous project.

~3h
0 / 7 steps
🚀
Intro

A network problem is a stack of possible failures, not one

Prerequisites: TCP/IP Fundamentals; DNS Deep Dive; HTTP/HTTPS Internals; Load Balancing Fundamentals.

"The site is down" or "I can't connect" is a symptom, not a diagnosis — and the actual cause could be sitting at any layer this roadmap has covered: a DNS resolution failure (project 2), a TCP connection that never completes (project 1), a TLS handshake failing (project 3), an application returning errors despite the connection working fine (also project 3), or a load balancer routing to a dead backend (project 4). Randomly trying tools until something looks wrong wastes time; working through the layers in a fixed order doesn't.

This project builds exactly that fixed order — a repeatable sequence of checks that isolates which layer is broken before you go deep on any single tool — and then has you use it cold, without knowing in advance what's actually wrong, against a deliberately sabotaged version of the previous project's load-balanced setup.

🔨

Layer 1: is the host even reachable?

🔨

Layer 2: is the path itself the problem?

🔨

Layer 3: sockets, ports, and what's actually listening

🔨

Layer 4: look at the actual packets

🎯
Secret Mission

Secret Mission: diagnose a sabotaged load balancer cold

🧹
Wrap Up

Before You Go

Pro

Test what you just learned

Self-testing is one of the best ways to retain new skills. Unlock project quizzes to check your understanding.

Log in to unlock