ietf-asrg
[Top] [All Lists]

Re: [Asrg] 6. Proposals - DNS + PKI - Yahoo's "Domain Keys"

2003-12-09 14:16:21
Mark Baugher wrote:

Yakov,

At 11:46 AM 12/9/2003, Yakov Shafranovich wrote:

Markus Stumpf wrote:

On Mon, Dec 08, 2003 at 03:35:28PM -0500, Yakov Shafranovich wrote:

The signature attests to the fact that the domain name or server from which the message originated, is not forged.


*lol*
I don't see any more security here as with a "paranoid" dns lookup.
If I do a reverse DNS lookup and get a name and do a lookup of the name
and get the IP I can assume#1 that it is correct.
   #1 with drawbacks as to DNS spoofing and DNS security.
Now, if the sending MTA has a signature on the message and I use DNS
to get the public key to verify the signature #1 from above still
applies. So the win for using PKI and not paranoid DNS lookups is zero.
[..]


Would there be a difference if the message is forwarded through a list, or is transfered via multiple MTAs?


Is it fair to say that there may be many MTAs at the sender's domain, many at the receiver's domain, but there should at most be one MTA(i) in the forwarding path?

If I understand correctly, that would mean that three MTAs are involved: sender's (s), intermediate(i) and receiver's(r). Therefore, LMAP will work for MTA(s) to MTA(i) connection, but will fail at MTA(i) to MTA(r) connection unless the sender's domain lists MTA(i) in LMAP. This presents a problem mainly for forwarding and is addressed in the LMAP discussion draft, section 3.4.

> From the standpoint of mail signing, might the
multiple MTAs at the sender be considered as a single MTA and similarly for the receiver?

For DK, since the message itself is signed, forwarding is not as big of an issue, and how many MTAs are there does not matter, since they are not been authenticated anyway (unless some kind of a ESMTP extension is used, like Hector mentioned, or modified STARTTLS like Alan mentioned).

The point is that while LMAP does need some minor changes in forwarding, DK does not. BUT, as stated in the LMAP discussion draft, that is not a reason to choose one over the other, unless significant negative impact can be shown.

But then again, it is unclear what exactly is being done, since not too many details are available.

Yakov
-------
Yakov Shafranovich / asrg <at> shaftek.org
SolidMatrix Technologies, Inc. / research <at> solidmatrix.com
"Be liberal in what you accept, and conservative in what you send" (Jon Postel)
-------


_______________________________________________
Asrg mailing list
Asrg(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/asrg



<Prev in Thread] Current Thread [Next in Thread>