Skip to content

Conversation

@R6CG
Copy link
Contributor

@R6CG R6CG commented Aug 29, 2023

With the news that Firefox plans to start enabling ECH in the next releases, I wanted to suggest people check censorship measurements for use-application-dns.net, because that canary domain is meant to disable DoH, and ECH depends on DoH.

use-application-dns.net had originally been added in #504 2019-09-19, but then it was deleted in #727 (a dead-domain check) on 2021-02-23.

Maybe prune-dead-urls.py was run on a network that reported NXDOMAIN for use-application-dns.net, since that is the signal that is supposed to signal DNS filtering is in place?

This domain is meant to be used as a signal that overt DNS filtering is
in effect. Applications can try to resolve the name as a test for the
presence of DNS filtering. DNS filters that wish to announce their
presence can respond with NXDOMAIN rather than an IP address, to cause
some applications to disable DNS over HTTP support.

https://support.mozilla.org/en-US/kb/canary-domain-use-application-dnsnet
https://support.mozilla.org/en-US/kb/configuring-networks-disable-dns-over-https
@Lanius-collaris
Copy link
Contributor

The canary domain only applies to users who have DoH enabled as the default option. It does not apply for users who have made the choice to turn on DoH by themselves.
https://support.mozilla.org/en-US/kb/canary-domain-use-application-dnsnet

dnscrypt-proxy also blocks *.use-application-dns.net .
DNSCrypt/dnscrypt-proxy#1205
https://github.com/DNSCrypt/dnscrypt-proxy/blob/92e842126d40f6fcb919bce72983748ddb34dac9/dnscrypt-proxy/plugin_firefox.go

Copy link
Collaborator

@bassosimone bassosimone left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think we should merge to collect data about which networks seem to block use-application-dns.net. For the analysis, I think we should limit ourselves with analyzing queries generated by getaddrinfo, which should be the automatically configured resolver. But we should also keep in mind that, under specific setups, (e.g., dnscrypt-proxy and possibly others), we could get false positives (i.e., the block depends on the local user's configuration and not on the network).

🐳

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

Successfully merging this pull request may close these issues.

5 participants