-
Notifications
You must be signed in to change notification settings - Fork 15
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
[BUG] CIPHER_LIST
on Windows is wrong since in v0.6.x
#54
Comments
It's |
I can verify that you are right. Not only the safari fingerprints on Windows is not correct, but also all other fingerprints are incorrect. The different parts is:
which should be
The fingerprints on macOS and Linux are correct. It seems that the BoringSSL behavior on Windows is not the same with other platforms. I'm not sure it's our mis-configuration or a bug/feature in BoringSSL, still investigating. Thanks for reporting it! |
No problem! Thank you for the rapid replies and for creating all of this. Let me know when you figured it out so I can test on my machine too :) |
CIPHER_LIST
on Windows is wrong since in v0.6.x
I'm moving this issue to upstream, since the bug is not within Python code. |
I may have found the root cause, it should be a side effect of #12, when we enabled If // Order AES ciphers vs ChaCha ciphers based on whether we have AES hardware.
//
// TODO(crbug.com/boringssl/29): We should also set up equipreference groups
// as a server.
size_t num = 0;
if (has_aes_hw) {
for (uint16_t id : kAESCiphers) {
co_list[num++].cipher = SSL_get_cipher_by_value(id);
assert(co_list[num - 1].cipher != nullptr);
}
}
for (uint16_t id : kChaChaCiphers) {
co_list[num++].cipher = SSL_get_cipher_by_value(id);
assert(co_list[num - 1].cipher != nullptr);
}
if (!has_aes_hw) {
for (uint16_t id : kAESCiphers) {
co_list[num++].cipher = SSL_get_cipher_by_value(id);
assert(co_list[num - 1].cipher != nullptr);
}
} This will be automatically sovled once #20 is merged. |
Based on the issue title, does 0.5.10 still have correct fingerprints? |
Yes. |
@T-256 There is no build yet. |
@yifeikong ping me when I should try it out then |
@gamer191 The CVE-2023-38545 was also fixed by upgrading to 8.5.0 of curl. |
Can confirm it works for me, thank you 👍 |
While using
impersonate="safari17_2_ios"
, I get different JA3 digest results between my Windows and macOS machines (8be0b641abb257fae7b13bcfd2657032
on Mac anda76d766e1e01aa4cfaee1331b1bada3b
on Windows).Unfortunately this issue is triggering cloudflare on the Windows machine while on Mac it works just fine every time.
Just for testing I've tried using different python versions with an older openssl version but that is not the problem as the digest doesn't change.
The text was updated successfully, but these errors were encountered: