Hi, On Sat, Jul 13, 2024 at 08:03:13PM +0200, a1ex@dismail.de wrote:
I was doing some tests on browserleaks.com with TLS fingerprinting.
https://tls.browserleaks.com/iframe/ja3_hash
On Dillo with LibreSSL, this hash changes every time I reload the page. On Firefox (which uses NSS), the hash stays the same. I don't know if I understand fully, but it seems like LibreSSL is resistant to this type of fingerprinting.
Very nice discovery. I can also reproduce the same behavior with LibreSSL, but with OpenSSL it stays the same. Unfortunately, the JA3n hash stays the same for LibreSSL. I also found a way to access the "raw" data in JSON using: https://tls.browserleaks.com/tls With this little patch, the response can be shown in Dillo too: --8<-- diff --git a/src/IO/mime.c b/src/IO/mime.c index 9d5e6738..2361a459 100644 --- a/src/IO/mime.c +++ b/src/IO/mime.c @@ -112,6 +112,9 @@ void a_Mime_init() /* Add a major type to handle all the text stuff */ Mime_add_major_type("text", a_Plain_text); + + /* Display JSON as plain text too */ + Mime_add_major_type("application/json", a_Plain_text); } --8<-- There is also JA4[1,2], which can be displayed here: https://ja4db.com/id/ja4/ https://ja4db.com/id/ja4h/ [1]:https://developers.cloudflare.com/bots/concepts/ja3-ja4-fingerprint/ [2]:https://github.com/FoxIO-LLC/ja4 Both JA4 hashes seem to stay the same on reloading the page using LibreSSL. Best, Rodrigo.