Re: Interop lsh and SSH-2.0-GitLab-SSHD
Mouse <[email protected]> Wed, 25 Oct 2023 22:37:38 -0400 (EDT)
| Newsgroups | gmane.ietf.secsh |
|---|---|
| Message-ID | <[email protected]> |
>> I note you (pgut001@) XXed out the username in your trace. Is it a >> valid gitlab.com username and you're just using a key they don't >> accept? If so, perhaps that's relevant - [...] > No, it was a random test key for use on another system that I used > because it was convenient. So the expected result from it would have > been SSH_MSG_USERAUTH_FAILURE, same as for your 'mouse'. Even more puzzling, then. > It looks like we'd really need someone at the Gitlab end of things to > look at this, I concur. > we seem to be sending more or less the same thing, the only thing > that's left is the binary data at the start of the sig and then the > sig itself - are you sending the SHA-256 OID with or without NULL > parameters? If you mean the user-authentication sig, I'm not getting to the point of sending a sig. I'm sending the public portion of the key, as usual; the expected response is either "if that's really you, prove it" or "nope, that key won't do". Instead, I'm seeing the connection die ungracefully. If you mean kex, that's completing normally; I'd have to dig more than I have the leisure to right now to tell exactly what I'm sending - I can send a complete test-connection trace, including kex, to you and/or the list, if you'd like to have a look at it. /~\ 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