Pasi,

inline...

> -----Original Message-----
> From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
> Sent: Friday, May 23, 2008 10:08 AM
> To: Rainer Gerhards; [email protected]
> Subject: RE: [Syslog] fingerprint vs PSK
> 
> 
> If you have one transport sender and one transport receiver, there's
> probably no big difference (except that most TLS implementations,
> at least in their official releases, don't support RFC 4279 yet).

A very good point indeed.

> If you have a large number of transport senders, things get different.
> Especially if you're satisfied with authenticating only the server,
> PSKs would mean more administrative work (when new clients are added),
> since it has to be configured on both ends (while the server
> fingerprint needs to be configured only on the new client -- and since
> it's not secret, you can e.g. put it in your installation scripts or
> instructions).
> 
> (That said, it's clear that fingerprints are not the *only* possible
> solution here -- but I'd really like to pick one option that's
> good enough and get this document published, rather than spend
> lot of time analyzing all the possible solutions.)
> 

Thanks for the explanation. I agree to one good enough solution is fine.
I would suggest to add a few sentences on the reasoning to the ID.

Rainer
> Best regards,
> Pasi
> 
> > -----Original Message-----
> > From: ext Rainer Gerhards
> > Sent: 23 May, 2008 09:44
> > To: [email protected]
> > Subject: [Syslog] fingerprint vs PSK
> >
> > Hi all,
> >
> > in my implementation effort (now mostly completed), I asked several
> > people for advise on implementing fingerprints. In almost all
> > cases the
> > initial reply was "why use non-standard fingerprints when we
> > have PSK"?
> > I know that RFC 4279 in section 1.1 says:
> >
> >    If the main goal is to avoid Public-Key Infrastructures (PKIs),
> >    another possibility worth considering is using self-signed
> >    certificates with public key fingerprints.  Instead of manually
> >    configuring a shared secret in, for instance, some configuration
> >    file, a fingerprint (hash) of the other party's public key (or
> >    certificate) could be placed there instead.
> >
> > However, I think it would be useful to add some short text why
> > fingerprints are more desirable. And the real question is: are they
> > actually more desirable?
> >
> > Rainer
> > _______________________________________________
> > Syslog mailing list
> > [email protected]
> > https://www.ietf.org/mailman/listinfo/syslog
> >
_______________________________________________
Syslog mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/syslog

Reply via email to