Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

correlation via packet loss? #4

Open
thejh opened this issue Feb 3, 2021 · 1 comment
Open

correlation via packet loss? #4

thejh opened this issue Feb 3, 2021 · 1 comment

Comments

@thejh
Copy link

thejh commented Feb 3, 2021

near_path_nat.md says:

For HTTP/3 traffic the browser assembled UDP packets are sent via MASQUE to the IPPS and then the IPPS sends the UDP packets to the target server, essentially just adding the IP header. The IPPS forwards data received from the target server back to the browser via the same MASQUE streams that the outgoing data traversed.

Is there an analysis somewhere of how much the ability of a group of servers to directly observe packet loss (because forwarding happens at the UDP level) allows them to correlate connections? (The same question obviously also exists for connections that either drop entirely or strongly change their latency because the user switched between wifi and mobile networks, but I guess that's less avoidable.)

@DavidSchinazi
Copy link

Hi! I'm the editor of the MASQUE specifications. I think this is something we'll want to add to our Security Considerations section, but I'm not particularly worried about this attack in practice. These proxies will carry many simultaneous connections so the resources required to perform this limited correlation will be incredibly high. I'm not aware of a mathematical analysis though, and if someone were to make one I'd be happy to refer to it in the spec.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants