Support for type 64 and 65?

[ImLunaHey]

I'm seeing requests with a type of UNKNOWN_64 and UNKNOWN_65.

[silverwind]

These are SVCB and HTTPS (ref), PR welcome.

[valenting]

We have implemented support for the HTTPS record in the dns-packet fork we use for testing Firefox
https://searchfox.org/mozilla-central/source/testing/xpcshell/dns-packet/types.js#49
I'm not sure I'll have time to upstream the changes, but if anyone is interested I'm happy to help.
SVCB has the same wire format, so it should be easy enough to use the same implementation for both.

[maxpain]

Any updates?

[maxpain]

@mafintosh Could you please merge this? Or maybe you @valenting could make a PR?

[dolvany]

+1 for this

[valenting]

Feel free to have a look at Mozilla's fork differences
master...valenting:moz-dns-packet:master

It only implements HTTPS, but svcb should be the exact same code.

[fishstiqz]

Hello. I added support for type 65 in fishstiqz@513da0f

This is based on the code by @valenting for mozilla but expanded a bit, especially in decoding the parameters.

I added a bunch of test functions based on the test vectors in RFC 9460. Would someone like to review it or would you like a PR?

[fishstiqz]

Today I added support for type 64 (SVCB) in fishstiqz@0a20879

This was trivial as HTTPS is a type of SVCB and they share the same parser code (and the same RFC). I added some specific test cases for this RR. Let me know if you would like a PR or I might send you one in a few days if I haven't heard anything.

[msimerson]

These are SVCB and HTTPS (ref), PR welcome.

As already mentioned above, please do submit PRs, and thanks!

[fishstiqz]

These are SVCB and HTTPS (ref), PR welcome.

As already mentioned above, please do submit PRs, and thanks!

Ok awesome. I will make a PR soon. I'd like to give it a bit more testing in the real world first.

[fishstiqz]

Hey I just wanted to bump this issue and note the new PR.

[catscarlet]

Last publish: 2 years ago
Last legit commit: Aug 25, 2023

What da hall?

[fishstiqz]

Hey just a reminder to run that CI/CD again so we can get this issue closed.....