spf-discuss
[Top] [All Lists]

Re: Re: For SPF council review: Syntax error = Permerror = Message should be rejected?

2005-05-03 15:51:00
...... Original Message .......
On Wed, 04 May 2005 00:30:31 +0200 Frank Ellermann 
<nobody(_at_)xyzzy(_dot_)claranet(_dot_)de> wrote:
Scott Kitterman wrote:

"
[...]
Well make sure your record has correct syntax and doesn't
exceed any processing limits before you publish because if
you get it wrong, your mail will get rejected."

I think that will be a hard sell on spf-help.

Wayne's -00 draft is now the "official" draft for more than
four months.  If some tools like the SPF wizard still have a
serious problem to count to ten, then they are partially
broken.

The fact that it should stay away from proposing ptr is also
not new, I can't guess how often you had to post the same
question in spf-help, "are you sure that you need the ptr ?".

If users with difficulties to count to ten try to publish SPF
policies I can only pray that it's only about their own mail
and domain, and not affecting other users of the same domain.

You think that it could discourage new users.  I was much more
worried about Radu's results with incredibly complex sender
policies, among others Pobox, RR, your policy, and IIRC Terry
also said something about some dummy include-s referencing
non-existing policies as a good idea.  It's not, it's an error.

                          Bye, Frank

Yes, but what's mostly implemented is Meng's unknown=none.  I've yet to see 
anyone ask about permerror=reject.  I'm pretty sure if it was in the wild 
we'd have had questions (almost certainly impolite).

FYI, my record is much cheaper now thanks to one of my isp's responding to 
my request to simplify their record.  I was outside the schlitt processing 
limits for many months and got no rejections as a result.  This (if only 
weakly) supports my contention that the new permerror is not fielded.

Scott Kitterman


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