On Wed, 2006-01-18 at 09:50 -0800, Eric Hibbard wrote:
> > Maybe I was not completely clear. I think we should go the TLS route
> and 
> > let the operator decide whether he wants authenticated or
> > unauthenticated TLS (or asymmetric authentication, e.g. the server is
> > authenticated but the client is not just like in HTTPS) So I fully
> agree
> > with Rainer on this one.

> 
> This is a way to go, but it is important to note that the absence of
> client-side certificates (authentication) potentially exposes you to
> hostile clients attempting to masquerade as a legitimate client. 

_I_ know this, and I would not recommend doing this in any way :) And of
course should be documented in the RFC as well.

> It also makes it more difficult to guard against man-in-the-middle attacks.

If the server is authenticated then MiTM does not apply, does it?


-- 
Bazsi


_______________________________________________
Syslog mailing list
[email protected]
https://www1.ietf.org/mailman/listinfo/syslog

Reply via email to