Re: Terrapin
Mouse <[email protected]> Thu, 28 Dec 2023 13:08:51 -0500 (EST)
| Newsgroups | gmane.ietf.secsh |
|---|---|
| Message-ID | <[email protected]> |
>> In the EXT_INFO case, I really think the right thing would have been >> to make it a global request. > Unfortunately, you need to send it before userauth if you're going to > use it to change RSA signature types _during_ userauth. And global > requests are part of the ssh-connection subprotocol, [...] ! Thank you; I wasn't reading carefully enough. This sounds to me like an argument for something like a global-request but at the BPP layer, ideally the 1-19 range but that is likely to take significant time. I'd use something in the 192-255 range to experiment with and, if it seems sound, then someone with a stronger stomach for politics than I can try to get it turned into an RFC. For concreteness, I'll propose using 213, which I picked randomly (by reading single bytes from /dev/urandom until I got one in the right range), for experimenting. As a strawman proposal in case this sounds good, I propose a packet that looks just like GLOBAL_REQUEST but with a different packet type: byte SSH_MSG_whatever_we_call_it string request name boolean want reply .... request-specific data /~\ The ASCII Mouse \ / Ribbon Campaign X Against HTML [email protected] / \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B