Re: Rekey issue
[email protected] (Niels Möller)
| Newsgroups | gmane.ietf.secsh |
|---|---|
| Message-ID | <[email protected]> |
Mouse <[email protected]> writes: > I can see at least three ways of dealing with this: (1) when rekeying, > always accept the host key regardless of what it is, (2) when rekeying, > require that the host key be what it was the first time around, always > accepting if it is and erroring if not, and (3) when rekeying, behave > as normal except that a second copy of the host key from the first time > around is, effectively, added as a trusted key for the host. I think I'm doing (3). That's what made most sense to me when I implemented it, but maybe there are better options. When an unknown key is received at the initial keyexchange, there are three possible outcomes depending on the user interaction: A. Disconnect. B. Accepted only for this connection. Key is added to the in-memory list of trusted host keys. C. Accepted. As above, but in addition, the key is appended to the host key list on disk. Regards, /Niels -- Niels Möller. PGP-encrypted email is preferred. Keyid C0B98E26. Internet email is subject to wholesale government surveillance.