On Fri, Sep 12, 2014 at 05:09:13PM -0400, Jeffrey Haas wrote:
> With some help from Kent, Dean and Alia, I've put together a rough first
> draft of requirements I2RS has on netmod/netconf.  It should be strongly
> noted that due to a confluence of a lot of bad timing (travel, vacation,
> etc.) I didn't have time to more broadly reach out and involve interested
> parties.
> 

[...]

> 
> Comments are appreciated.  Flames are not unexpected.
> 

Thanks Jeff for putting this together. Concerning section 2.1.1, I am
not sure I agree with:

   The SSH transport does not mandate authentication be done; it is an
   optional feature.

RFC 6242 says:

   The identity of the SSH server MUST be verified and authenticated by
   the SSH client according to local policy before password-based
   authentication data or any configuration or state data is sent to or
   received from the SSH server.  The identity of the SSH client MUST
   also be verified and authenticated by the SSH server according to
   local policy to ensure that the incoming SSH client request is
   legitimate before any configuration or state data is sent to or
   received from the SSH client.  Neither side should establish a
   NETCONF over SSH connection with an unknown, unexpected, or incorrect
   identity on the opposite side.

I also think that NC over TLS requires that both sides authenticate
and verify the certificates. I think as far as NETCONF is concerned,
all transports do actually mutual authentication.

Perhaps there is confusion here with RESTCONF where the current
wording may be read as authentication is optional. But then, this
is not past the IESG yet. ;-)

/js

-- 
Juergen Schoenwaelder           Jacobs University Bremen gGmbH
Phone: +49 421 200 3587         Campus Ring 1, 28759 Bremen, Germany
Fax:   +49 421 200 3103         <http://www.jacobs-university.de/>

_______________________________________________
i2rs mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/i2rs

Reply via email to