spf-discuss
[Top] [All Lists]

Re: TXT Records

2003-11-23 06:20:40
On Saturday 22 November 2003 4:18 pm, Philip Gladstone wrote:
I'm doubtful that changing the ESMTP protocol will fly -- there are 
still lots of people using SMTP and it isn't clear how quickly the 
firewall vendors will change to incorporate the new command. On the 
other hand, it does have simplicity and cleanness on it's side.

Notice that the format of the XSPF query allows the ESMTP server to respond to 
queries about any domain:

'XSPF' <SP> (SENDER-HOSTNAME|SENDER_IP) <SP> DOMAIN <SP> LOCALPART

For servers that do not wish to be used as a 'public oracle' giving opinions 
about third party domains, a 'try elsewhere' response should be given for 
domains it is not an mx for:

551 User not local; please try <other-domain-mx>

However there is nothing to prevent your ISP's SMTP server from providing you 
with a response for any domain you ask about, by calling out to the correct 
server (is it would do for incoming mail anyway). It may even be faster since 
your ISP's XSPF response cache will then be available to you.

When responding as a proxy in this way, an ESMTP server should return a 251 
success code instead of 250:

251 <SP> ('ALLOW'|'DENY') <SP> 'FOR' <SP> TTL-VALUE <SP> 'SECONDS' [<SP> 'FOR 
ALL']

This mechanism need not be restricted to ISP's. It could be used to provide 
arbitrary opinions concerning non-XSPF domains in the manner of the 'fallback 
domain' that has already been discussed for DNS-based SPF. A server that is 
constructing it's own 'best guess' rather than forwarding an authoritative 
response should respond with code 252 to distinguish it.

There are other interesting possibilities which could be explored:
1. Returning a percentage 'degree of trust' rather than boolean allow|deny.
2. Query for other known fallback oracles, with a trust weighting for each.

This would enable receiving MTAs to build distributed trust networks that 
share their opinions about permitted senders for non-participating domains, 
yet the protocol would remain trivially simple.

- Dan

-------
Sender Permitted From: http://spf.pobox.com/
Archives at http://archives.listbox.com/spf-discuss/current/
Latest draft at http://spf.pobox.com/draft-mengwong-spf-02.6.txt
To unsubscribe, change your address, or temporarily deactivate your 
subscription, 
please go to 
http://v2.listbox.com/member/?listname(_at_)©#«Mo\¯HÝÜîU;±¤Ö¤Íµø?¡


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