spf-discuss
[Top] [All Lists]

Re: Re: SPF implementations

2005-08-16 07:39:35
Graham Murray wrote:
"Stuart D. Gathman" <stuart(_at_)bmsi(_dot_)com> writes:


On Tue, 16 Aug 2005, Graham Murray wrote:

I would disagree. I suspect that it is common for 'role' addresses to
be aliased to a personal address, often with the same person receiving
more than one 'role' address. If user(_at_)example(_dot_)com receives mail for
postmaster(_at_)example(_dot_)com and abuse(_at_)example(_dot_)com, they may 
well want
replies to mails sent to those addresses to show the appropriate
'role' account rather than user(_at_)example(_dot_)com(_dot_)

Yes, but we are not talking about internal forwarding.  The above
scenario causes no problems with SPF.


Neither am I talking about internal forwarding. I am talking about the
replies to the internally forwarded or aliased email. If my mailbox is
user(_at_)example(_dot_)com it also receives mail addressed to
postmaster(_at_)example(_dot_)com, then when I reply to mail addressed to
postmaster I may well want to change the RFC(2)821 envelope return
path and header "From" to "postmaster(_at_)example(_dot_)com" rather than the
default "user(_at_)example(_dot_)com". This, I think is one situation where
RFC2831 section 7.1 is legitimate.

Yes, but that's all within the same domain. With the limited exception of records that use the localpart of the e-mail address in macros, this is nothing to do with SPF.

I would say it's a an internal routing/forwarding issue however because once I deliver to your MX, how you route it afterwards (the local part) is of no concern to me as long as you don't bounce it back to me after you've accepted responsibility for it....

Scott K


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