spf-discuss
[Top] [All Lists]

Re: an other approach for SPF

2004-08-03 07:05:28
On Tue, Aug 03, 2004 at 03:54:20PM +0200,
 Daniel Gueniche <gueniche(_at_)grenoble(_dot_)cnrs(_dot_)fr> wrote 
 a message of 28 lines which said:

first, yes you are right, we should refuse mails from domains with no 
A/MX records, we are going to fix it.

And the lame delegation to ns2.nic.fr, please.

~ % check_soa grenoble.cnrs.fr
ns2.nic.fr is not authoritative for grenoble.cnrs.fr
calypso.urec.cnrs.fr has serial number 2004072801
labs.grenoble.cnrs.fr has serial number 2004072801
 
secondly, in my mail, please read anti-virus and no anti-spam. I
mistyped, of course anti-spams do not send back mails, but
anti-virus do.

They are wrong, badly wrong, since almost all Microsoft worms use
forged email addresses.

We have sent a mail from reflector.grenoble.cnrs.fr (147.173.3.39)
simulating a mail from user(_at_)another(_dot_)existing(_dot_)domain(_dot_) 
My question was
why SPF does not base its analyses on that IP address ?

It does. SPF uses the domain in the MAIL FROM for *identification* and
the IP address of the SMTP client for *authentication*.




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