37C3: Fuzzing the TCP/IP Stack

events.ccc.de

135 points·rurban··8 comments

Video: https://media.ccc.de/v/37c3-12235-fuzzing_the_tcp_ip_stack

Played around with coverage-guided fuzzing using honggfuzz and Intel BTS/PT a while back. Didn't have much time to fine-tune, but it seemed solid. It'd need a couple of months to properly dial in and see real results. The coverage was increasing, and aligning with the symbol tables, it was hitting the right parts of the kernel.

Harness: https://github.com/google/honggfuzz/tree/master/examples/lin...

With coverage guided fuzzing you wont get into the deep states of such protocols. If it's TCP/IP or TLS or similar. You'd really need a client/server simulator, and fuzz this.

Interesting historical overview of IP stack bugs. But nothing new, what he calls pre-alpha.

All-in-all quite a disappointing session IMHO the worse I attended at 37c3, nothing compared to the excellent https://events.ccc.de/congress/2023/hub/event/numerical_air_... that was running at the same time.

> All-in-all quite a disappointing session IMHO the worse I attended at 37c3, nothing compared to the excellent https://events.ccc.de/congress/2023/hub/event/numerical_air_... that was running at the same time.

Yeah, was there for this one, can highly recommend. Will also be building my own DIY air quality sensor to add to their network.

Which network are you considering for your DIY air quality sensor? They mention on a slide:

- https://sensor.community/

- https://luftdaten.at/

- https://openair.cologne/

Since I'm neither in Austria nor Cologne, I'd go with https://sensor.community

That session on air quality wasn’t bad, but just didn’t fit my needs in the moment. Glad you liked it so much.

Reply on news.ycombinator.com