spf-discuss
[Top] [All Lists]

Re: DNS RRtypes: changing PI

2003-10-20 15:43:17
Mark, Jonathan & Izzy,

On Monday 20 October 2003 21:04, Mark Lentczner wrote:
I don't understand this.  Surely Exim doesn't have any code to
implement SPF at this time.  

Not SPF specific, but it has the functionality as it stands. This is why I 
became so active on this list - SPF could be implemented NOW if we needed 
early adoption.

And surely any code that that got added to
Exim to support SPF would be constructing it's own queries to DNS.

All DNS queries are expected to have valid RFC2821 domains, hence my real 
objection to the underscore thing. I can configure it to allow underscores, 
but it's an all-or-nothing scenario, and I don't really want to break the 
current validation checks.

On Monday 20 October 2003 21:51, Jonathan Steinert wrote:
What is it about SPF doing a lookup on _smtp_client.example.com that
makes it different?

Sorry, nothing at all! Just me getting the subdomain in context confused!

On Monday 20 October 2003 22:23, Izzy Kindred wrote:
      That depends upon what is meant by "SMTP context".  As I
understand SPF at present, there is no use of the underscores in the
actual SMTP transaction.

      If SPF starts to pass information inside the SMTP transaction
(as some recent posts have suggested), then we will have to document
our changes/additions to SMTP, and we can easily document the use of
underscores at that time.

SPF should (and in my mind must) do all the work within the SMTP transaction. 
I must be able to issue a 5xx response to any attempt to send 
SPF-unauthorised email. I am not going to be responsible for generating a 
bounce, because that is not the way it should be done.

      I am somewhat opposed to basing the design of SPF on quirks
that exist in some SMTP implementations.  RCFs should drive
implmenetations.  Implementations should not drive RFCs.

That is fair and valid. But Exim veers towards the "RFC picky adherant" model, 
rather than the "introduce quirks" approach. And I think an easy, practical 
and effective approach should drive RFC development.

      That said, it may be that the anti-underscore camp will win
this argument, simply because they care more.

:-)
Ahhh. Thank you!


-------
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.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>