spf-discuss
[Top] [All Lists]

Re: a grand unified theory of MARID

2004-06-20 14:32:54
On Sat, 19 Jun 2004 23:35:05 -0400, Meng Weng Wong wrote:

Folks who prefer picture to go along with their text may
like to see this slideshow:

  http://spf.pobox.com/slides/unified%20spf/

I like the picture, makes the words easier to digest, plenty of 
scope for an alternate food analogy

Anyway, it raised a few questions as I went through it.

0040-alwayscheckhelo 
http://spf.pobox.com/slides/unified%20spf/0040.html

This slide implies (very strongly) that in SPF Classic, 
checking the HELO is a MUST, but the latest posted RFC proposal 
(200405), section 2.2.1 says 
        "SMTP+SPF receivers MAY check the HELO argument"
My opinion is that it should be a MUST, but I am concerned 
about potential issues with potentially matching .local domains 
at both ends of the transaction.


0429-linuxbroadband 
http://spf.pobox.com/slides/unified%20spf/0429.html

This was a big surprise, since it seems an ISP can not publish 
(effective) policy stating that their IP address space is not 
to be used to send emails from directly. Even DHCP Dialup can 
not have an enforced no direct email policy.

I sympathize with the "poor linux user" on a consumer grade 
broadband connection, after all I fit that description, though 
my MTA is actually on a Windoze server, since I like my Win32 
MTA.

If an ISP has a "no direct mail" or a "no server policy", then 
they should expect a published policy for these IP addresses to 
be honoured. Ideally I would hope that users allowed to send 
mail directly could switch off this record, if enabled by 
default on new connections.

Though since many admins use DHCP/Dialup DBLs (or maintain 
their own like AOL) to block these IP addresses (to which some 
ISPs submit their DHCP IP's for addition), so allowing it to 
pass may be in vain.

Regards
Karl Prince





______________________________________________________________
Email via Mailtraq4Free from Enstar (www.mailtraqdirect.co.uk)


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