[Fwd: Re: Django-DMARC making it easier to manage DMARC reports - Beta 0.1.3 on PyPI]

Alan Hicks ahicks at p-o.co.uk
Sun Mar 15 11:19:40 EDT 2015


---------------------------- Original Message ----------------------------
Subject: Re: Django-DMARC making it easier to manage DMARC reports - Beta
0.1.3 on PyPI
From:    "Alan Hicks" <ahicks at p-o.co.uk>
Date:    Sun, March 15, 2015 2:10 pm
To:      "Chris Angelico" <rosuav at gmail.com>
--------------------------------------------------------------------------

On 15/03/2015 12:29, Chris Angelico wrote:
> On Sun, Mar 15, 2015 at 11:20 PM, Alan Hicks <ahicks at p-o.co.uk> wrote:
>> Mailing lists are an issue on many levels, and dmarc has an faq that may
>> help http://dmarc.org/faq/receivers/#r_2
>
> No, they're not; the issue is only with DMARC. I have no problems with
> mailing lists and SPF records, because the list software changes the
> envelope addresses. Mailing lists are excellent, and I run a number of
> them - but not nearly as many as I'm a member of - and spam can be
> dealt with in the absence of DMARC. I've seen no benefit from it,
> unlike with SPF which has blocked no end of dead mail; and it causes
> issues on several lists.
>
> ChrisA
>

DMARC is only reinforcing SPF and DKIM so DMARC is not the core issue.

If issues are not surfacing it is more likely that both SPF and DKIM are
not strong spam indicators rather than there are no issues.

SPF is rarely implemented decisively ~all instead of -all so is not very
strong.

This email failed dkim from gmail.com but was not rejected (2015-03-15
12:30:13 1YX7gL-000JG2-Iv DKIM: d=gmail.com s=20120113 c=relaxed/relaxed
a=rsa-sha256 [verification failed - signature did not verify (headers
probably modified in transit)]).

This issue only raised itself by my company implementing a stronger
policy than currently necessary.  With the volume of spam increasing, I
don't see things getting easier, DMARC is just encouraging use of SPF
and DKIM and appears to be gaining traction in a global world where
reputation matters.

Alan





More information about the Python-list mailing list