Protocol in Code · Track

ICE

NAT traversal as discovery, candidate gathering, and priority-ordered connectivity checks

5 sessions Course material (English)

Sessions

ICE track sessions

Session 01

STUN Tells You Your Own Address

What does a STUN server actually tell you, and how do you turn that answer into knowledge that you are behind a NAT?

Open Session 01
Session 02

NATs Have Personalities

Given a set of observed mappings toward different destinations, what does the pattern of change (or lack of it) tell you about the NAT's personality?

Open Session 02
Session 03

Candidates Are Gathered, Not Chosen

Which addresses become candidates, and what decides the order a peer should try them in — and how are those two questions kept completely separate?

Open Session 03
Session 04

Pairs Are Checked in Priority Order

Gathering (Session 03) produced a pile of maybes on each side. How does that pile become an ordered plan, and what decides which pair actually gets used?

Open Session 04
Session 05

Build the Toy Connectivity Check Loop

Four sessions built STUN discovery, NAT personality classification, candidate gathering, and priority-ordered checklists separately. What does it look like when two agents run all of that against each other, end to end, against a NAT they never had to understand?

Open Session 05